Digital Eye Screening Support System Final Project report Submitted to the school of Biomedical Engineering for the partial fulfillment of requirements for the degree of Bachelor of science in Biomedical Engineering Jimma Institute of Technology (JIT) Jimma University Jimma, Ethiopia Team members Name 1. 2. 3. 4. 5. 6. ID No Ajaeba Mustefa…………….0137/07 Bisrat Amanuel…………….0430/07 Fikeremariam Getachew….. 0708/07 Lemlem Solomon…………. 1070/07 Lemlem Tiruneh…………... 1073/07 Mekdes Endashaw………....04776/06 Advisors Gizeaddis Lamesgin (Ph.D.) Melese Umma (MSc.) June 13, 2019 Digital Eye Screening Support System 2019 Executive summary Human eye is a vital organ of vision which reacts to light and pressure. It has a number of components which include but are not limited to the cornea, iris, sclera, pupil, lens, retina, optic nerve and vitreous. Retina is essential component of the eye and serves the primary purpose of photoreception. The center of the retina is the optical disc, a circular to oval white area measuring about 3 mm (about 1/30 of retina area). There are various things that affect structure of the eye, causing visual impairment and blindness, such as Glaucoma, Cataract, Age related macular degeneration, retinopathy etc. In Ethiopia eye diseases and blindness occurs especially for elder people. From several type of eye disease glaucoma, ARMD and cataract are common. These diseases can be prevented or treated if it caught earlier. But in Ethiopia, due to lack of awareness, less organized hospitals diagnosis system, and lack of devices in low level health facilities there is no habit to do early checkup before it aggravates. Diseases like glaucoma and retinopathy do not show symptom in early stage and they are irreversible if they already damage the eye. Ophthalmoscopy, slit-lamp microscopy, Gonioscopy and manual optical devices are few technologies that are used to diagnose/screen the eye. The quality of images acquired using optical imaging modalities in developing countries including Ethiopia is very less. High-tech imaging methods like funds-camera are not available in most health facilities in the country. In this project, an eye screening support system that is user friendly, cost effective, and portable is proposed. The system includes image acquisition, enhancement, segmentation and cup to disc ratio calculation. Images acquired can also be saved digitally for later use. In addition, pressure of the cornea is measured using an IR sensor. The developed system will have a great impact in supporting physicians in their decision in the diagnosis of the eye. i Digital Eye Screening Support System 2019 Disclosure statement We hereby declare that this report, submitted to school of Biomedical Engineering at the Jimma Institute of Technology as a partial fulfillment of the intervention and evaluation report guideline requirements for the degree of Bachelor of Science in Biomedical Engineering, has not been submitted as an exercise for a degree at any other university. We also certify that the work described herein is entirely our own work with the exception of paraphrased or quoted work whose sources are appropriately cited in the references. This report may be made available within the university library and may be Photocopied or loaned to other libraries as a reference for other work. Students Name Signature 1. Ajaeba Mustefa ----------------- --------------- 2. Bisrat Amanuel ---------------- ---------------- 3. Fikeremariam Getachew ----------------- ---------------- 4. Lemlem Solomon ----------------- ----------------- 5. Lemlem Tiruneh ---------------- ---------------- 6. Mekdes Endashaw ----------------- ----------------- Name of Advisor Signature Date Date 1. Gizeaddis Lamesgin (PhD) ---------------------- ------------------- 2. Melese Umma (MSc) ---------------------- ------------------- Name of Evaluator Signature Date ---------------------------- --------------------- -------------------- ---------------------------- ---------------------- ------------------- ii Digital Eye Screening Support System 2019 Statement of approval On behalf of the school of Biomedical Engineering at the Jimma Institute of Technology, we the mentor(s) of this project, digital eye screening support system and I, the evaluator, confirm that this project was approved as the topic for the intervention and evaluation project for the students, Ajaeba Mustefa, Bisrat Amanuel, Fikremariam Getachew, Lemlem Solomon, Lemlem Tiruneh and Mekdes Endashaw. Mentor name signature date signature date 1. Gizeaddis Lamesgin (PhD) 2. Melese Umma (MSc) Evaluator name iii Digital Eye Screening Support System 2019 Acknowledgement First of all, we would like to give our deepest thanks to our Almighty God. Next our deepest gratitude is also going to Jimma University Institute of Technology (JIT) and School of Biomedical Engineering (BME) for preparing work area and materials for implementation. Then we would like to thank our instructor Dr. Timothy Kwa for giving us his advises. we heart fully thank our clients Dr. Tizazu Sisay, DR. Dagmawit, Dr Tarekegn from ophthalmology department in Jimma Medical Center We also give special thanks for our advisors Gizeaddis Lamesgin (PhD) and M/r Melesse Umma (MSc). Our department head Engineer Wasihun Alemayehu (MSc) and instructor Shimelis Nigussie (BSc) also give as their advice and we are very grateful for their contribution. We also give our enormous thanks to Abel Worku (Msc) and Habtamu Abafoge for their unforgettable contributions. We are grateful for BME Lab assistants and other willing people for the interview. iv Digital Eye Screening Support System 2019 Table of Contents Executive summary ........................................................................................................................ i Disclosure statement ..................................................................................................................... ii Statement of approval.................................................................................................................. iii Acknowledgement ........................................................................................................................ iv List of Figure ................................................................................................................................ ix List of Table ................................................................................................................................... x Acronym........................................................................................................................................ xi CHAPTER ONE ........................................................................................................................... 1 INTRODUCTION ....................................................................................................................... 1 1.1. Clinical background ......................................................................................................... 1 1.1.1 Glaucoma ........................................................................................................................ 1 1.1.2. Cataract .......................................................................................................................... 2 1.1.3. Age-related macular degeneration ................................................................................. 2 1.1.4. Diabetic retinopathy ...................................................................................................... 3 1.2. Literature review .............................................................................................................. 4 1.3. Existing solutions ............................................................................................................. 5 1.3.1. Perimeter ................................................................................................................... 5 1.3.2. Ophthalmoscope ....................................................................................................... 6 1.3.3. Gonioscopy ............................................................................................................... 6 1.3.4. Pachymetery.............................................................................................................. 6 1.3.5. Optical Coherence Tomography ............................................................................... 6 1.4. Gap analysis ..................................................................................................................... 7 1.4.1. Subjective decision ................................................................................................... 7 1.4.2. Accessibility.............................................................................................................. 7 v Digital Eye Screening Support System 2019 1.4.3. Cost ........................................................................................................................... 7 1.4.4. Portability.................................................................................................................. 7 1.4.5. Manual diagnosis ...................................................................................................... 8 1.4.6. Time .......................................................................................................................... 8 1.5. Problem statement ............................................................................................................ 8 1.6. Objective of the project and constraints ........................................................................... 9 1.6.1. General Objective ..................................................................................................... 9 1.6.2. Specific Objectives ................................................................................................... 9 1.6.3. Constraints ................................................................................................................ 9 1.7. Significance of the study .................................................................................................. 9 1.7.1. Store images .............................................................................................................. 9 1.7.2. Location independent .............................................................................................. 10 1.7.3. Time saving ............................................................................................................. 10 1.7.4. Reliable result ......................................................................................................... 10 1.7.5. User friendly ........................................................................................................... 10 1.8. Scope of the study .......................................................................................................... 10 1.9. Breakthrough .................................................................................................................. 11 CHAPTER TWO ........................................................................................................................ 12 DESIGN STRATEGY .............................................................................................................. 12 2.1. Background information ................................................................................................ 12 2.2. Evaluation of brainstormed idea .................................................................................... 12 2.2.1 Wearable and wireless intraocular pressure measurement ........................................... 12 2.2.2 Image acquisition, enhancement and segmentation system for glaucoma detection .................................................................................................................................... 14 2.2.3 Integrating the slit lamp (bio microscope) with computer application ......................... 15 vi Digital Eye Screening Support System 2019 2.3. How the matrix is obtained?........................................................................................... 16 2.4. Proposed initial design ................................................................................................... 18 2.3.1. Image acquisition ........................................................................................................ 19 2.3.2. Image Enhancement .................................................................................................... 19 2.3.3. Edge detection ............................................................................................................. 19 2.3.4. Image segmentation ..................................................................................................... 20 2.3.5. Cup-to-disc ratio ..................................................................................................... 20 2.3.6. IOP measurement .................................................................................................... 21 2.4. Components lists ............................................................................................................ 21 2.4.1. Hardware part.......................................................................................................... 21 2.4.2. Software part ........................................................................................................... 24 2.5. Implementation plan ....................................................................................................... 24 CHAPTER THREE .................................................................................................................... 25 FINAL DESIGN ....................................................................................................................... 25 3.1. Steps for construction of prototype ................................................................................... 25 3.3. Final product ...................................................................................................................... 27 3.4. Final components list ......................................................................................................... 27 3.5. Graphical user interface ..................................................................................................... 28 3.6. Simulation and results .................................................................................................... 33 CHAPTER FOUR ....................................................................................................................... 35 TESTING AND RESULTS ...................................................................................................... 35 4.1. Design criteria .................................................................................................................... 35 4.1.1. Accuracy ...................................................................................................................... 35 4.1.2. Cost ............................................................................................................................. 35 4.1.3. Portability ................................................................................................................... 35 vii Digital Eye Screening Support System 2019 4.1.4. Easy to use ................................................................................................................... 35 4.1.5. Sensitivity .................................................................................................................... 36 4.1.6. Time saving ................................................................................................................. 36 4.2. Testing plan ........................................................................................................................ 36 4.2.1. Market Testing ............................................................................................................. 36 4.2.3. Sensitivity test.............................................................................................................. 36 4.3. Testing conducted .............................................................................................................. 37 CHAPTER FIVE ........................................................................................................................ 38 CONCLUSION AND RECOMMENDATION ........................................................................ 38 5.1. Conclusion.......................................................................................................................... 38 5.2. Recommendations .............................................................................................................. 39 References ................................................................................................................................................ 40 Appendices ............................................................................................................................................... 43 Appendix A ............................................................................................................................... 43 Pair wise comparison chart .................................................................................................... 43 Appendix B ............................................................................................................................... 44 Sharp IR sensor datasheet ...................................................................................................... 44 Appendix C ............................................................................................................................... 45 Camera specification ............................................................................................................. 45 Appendix D ............................................................................................................................... 46 Arduino code ......................................................................................................................... 46 Appendix E................................................................................................................................ 48 MATLAB code ...................................................................................................................... 48 Appendix F ................................................................................................................................ 58 Statement of approval from ophthalmology department ....................................................... 58 viii Digital Eye Screening Support System 2019 List of Figure Figure 1: different types of glaucoma indications .......................................................................... 4 Figure 2: block diagram of IOP measurement .............................................................................. 13 Figure 3: block diagram of image processing method .................................................................. 14 Figure 4: block diagram of slit lamp and computer integration for analysis ................................ 15 Figure 5: algorithm or flow chart .................................................................................................. 18 Figure 6: image of normal eye and glaucoma eye ........................................................................ 19 Figure 7: volk 90D ........................................................................................................................ 22 Figure 8: CCD ............................................................................................................................... 22 Figure 9: Arduino .......................................................................................................................... 22 Figure 10: LCD ............................................................................................................................. 23 Figure 11: Sharp IR sensor ........................................................................................................... 23 Figure 12: potentiometer ............................................................................................................... 24 Figure 13: prototype construction: A) shows IOP design on Proteus B) Prototype construction of image C) graphical user interface D) final simulation Part .......................................................... 26 Figure 14: final prototype ............................................................................................................. 27 Figure 15: graphical user interface image ..................................................................................... 28 Figure 16:image acquisition in GUI ............................................................................................. 29 Figure 17: Image enhancement on GUI ........................................................................................ 30 Figure 18: edge detection on GUI ................................................................................................. 31 Figure 19: image segmentation on GUI ........................................................................................ 32 Figure 20: cup to disc ratio on GUI .............................................................................................. 33 Figure 21: A) Proteus results for normal B) IOP Proteus results for high IOP ............................ 34 Figure 23: Sharp IR sensor Datasheet A ....................................................................................... 44 Figure 24:Sharp IR sensor Datasheet B ........................................................................................ 44 Figure 25: camera specification .................................................................................................... 45 Figure 26: statement of approval from JUSH ophthalmology ward ............................................. 58 ix Digital Eye Screening Support System 2019 List of Table Table 1: screening Pugh matrix .................................................................................................... 16 Table 2: scoring Pugh matrix ........................................................................................................ 17 Table 3: Iteration of the device ..................................................................................................... 26 Table 4: testing plan and result ..................................................................................................... 37 Table 5: pair wise comparison chart ............................................................................................. 43 x Digital Eye Screening Support System 2019 Acronym 1. ARMD………………………….... Age related macular degeneration 2. CCD………………………………. Charged capacitive device 3. CCT………………………………..Central corneal thickness 4. CDR………………………………. Cup to disc ratio 5. FT…………………………………. Fourier transform 6. GUI………………………………. Graphic user interface 7. HRT……………………………….. Heidelberg retinal tomography 8. IOP……………………………….. Intraocular pressure 9. IR………………………………….. Infrared 10. LCD……………………………….. Liquid crystal display 11. MATLAB………………………….. Matrix Laboratory 12. OAG……………………………. .. Open angle glaucoma 13. OCT……………………………….. Optical coherence tomography 14. PSD…………………………………Positive sensitive detector 15. ROI………………………………. Region of interest 16. RPE………………………………. Retinal pigment epithelium 17. USB……………………………….. Universal serial bus 18. WHO……………………………… World health organization xi Digital Eye Screening Support System 2019 CHAPTER ONE INTRODUCTION 1.1. Clinical background Human eye is a vital organ of vision which reacts to light and pressure. It has a number of components which include but are not limited to the cornea, iris, sclera, pupil, lens, retina, optic nerve and vitreous [1]. The protective outer layer of the eye is called the sclera. Retina is essential component of the eye and serves the primary purpose of photoreception [2]. Retina is approximately 0.5mm thick and covers the inner side at the back of the eye. The center of the retina is the optical disc, a circular to oval white area measuring about 3 mm (about 1/30 of retina area). The mean diameter of the vessels is about 250μm [3]. All other structures of the eyes are subsidiary and act to focus images on the retina, to regulate the amount of light the eye or to provide nutrition, protection or motion [3]. The eye is often compared to a camera. Each gathers light and then transforms that light into pictures. Extraocular muscles help move the eye in different directions. Nerve signals that contain visual information are transmitted through the optic nerve to the brain [4]. The eye plays a very important role not only in life but also in human body. Also it gives us the sense of sight, allow us to see and interpret colors, dimension of object [5]. There are various things that affect structure of the eye, cause visual impairment and blindness. The main visual disorders are listed below. 1.1.1 Glaucoma Glaucoma is an acquired disease of the optic nerve (neuropathy) characterized by specific changes of the optic nerve and by visual field defects that correspond to the areas of optic nerve structural damage. It is one of the many eye diseases can lead to the blindness if it is not detected and treated in proper time [6]. It is often associated with the increased in the intraocular pressure (IOP) of the fluid known as aqueous humor in eye. High eye pressure, or intraocular pressure, is often present and is the only modifiable risk factor for glaucoma [7]. Open-angle glaucoma (OAG) is the most common form of glaucoma. It occurs when the fluid in the anterior chamber reaches the angle between the cornea and iris and passes too slowly through the meshwork drain. As the fluid builds up, the pressure inside the eye rises to a level that may damage the optic nerve [7]. 1 Digital Eye Screening Support System 2019 Glaucoma has been shown by numerous epidemiological studies to be a leading cause of blindness worldwide, ranging from 1% - 2% of the population >40 years of age among various regions. Of those with glaucoma, primary open-angle glaucoma is generally the most prevalent type. It is found to be one of the blinding eye diseases, causing 62,000 blind people in Ethiopia according a national survey in 2006 [8]. Glaucoma is detected through a comprehensive eye examination by an eye care professional (ECP). 1.1.2. Cataract Cataracts are a clouding of the lens, resulting in a blurred image. The lens is made from water and protein which arranged in a way that keeps the lens clear and allows light to pass through [9]. Oxidative damage caused by free radicals is considered to be an important factor in aging and the development of chronic diseases, including cataract formation [10] . As age increases, some of the protein may clump together and start to cloud a small area of the lens, resulting in a cataract that reduces the amount of light reaching the retina. Over time, the cataract may slowly grow larger and cloud more of the lens, making vision gradually duller or more blurred [9]. Severe cataracts are a major cause of treatable blindness throughout the world [10]. There are three main types of cataract: • Nuclear cataracts: are the most common type and named because they occur in the center of the lens • Cortical cataracts: it starts in the cortex (periphery) of the lens and gradually extend to the center of the lens • Subcapsular cataracts: are those in which the opacities are concentrated beneath or within the capsule of the lens. Cataract development is part of the ageing process. Hence, although lifestyle changes may reduce the rate of development for some people, cataracts cannot be prevented. However, surgery can correct the condition [9]. 1.1.3. Age-related macular degeneration In Age-related macular degeneration (ARMD), clumps of yellowish material known as drusen gradually accumulate within and beneath the retinal pigmented epithelium (RPE). The RPE cells may die and therefore no longer support the photoreceptors, which then cannot function, resulting in loss of vision in that part of the retina [10]. If the photoreceptors in the macula are affected, this can seriously impair fine visual skills; this situation is sometimes referred to as 2 Digital Eye Screening Support System 2019 ‘dry’ macular degeneration (or central, geographic atrophy), and is the most common form of ARMD [11]. The prevalence (for all degrees from minor to severe) of ARMD rises from approximately 2% in the 54-64 years old age group to 30% in the 75-85 years old age group [11]. 1.1.4. Diabetic retinopathy Diabetic retinopathy is a common complication of diabetes. Poor glucose control during diabetes affects the tiny blood vessels of the retina. The arteries become weakened and leak blood, leading to swelling or edema in the retina and decreased vision [12]. As diabetes progresses, circulation problems can also deprive the retina of oxygen. If abnormal blood vessel growth continues, scar tissue formation may cause retinal detachment and glaucoma may also develop [9]. There are multiple tests available for diagnosis and detection of eye diseases A. Ophthalmoscopy This diagnostic procedure helps the doctor to view the interior of the eye especially the retina. Eye drops are used to dilate the pupil and then examine the shape and color of the optic nerve by using small device with a light on the end to light and magnify the optic nerve. If the intra ocular pressure is not within the normal range or if the optic nerve looks unusual, it is necessary to take the exam many times [8]. B. Slit-Lamp Microscopy Slit-lamp microscopy is of fundamental importance in the treatment of eye disease like glaucoma. In this examination, the conjunctivae, anterior chamber, eye iris, lens, etc., are observed, but an auxiliary lens may also be used in combination in order to observe the anterior chamber angle and ocular fundus [13]. C. Gonioscopy Gonioscopy is a method of evaluating the anterior chamber angle to provide information regarding the type of glaucoma. It can also be utilized therapeutically for procedures such as laser trabeculoplasty and goniotomy. It is indispensable in the treatment and diagnosis of glaucoma. In Gonioscopy, it is important to properly recognize the various structures composing the anterior chamber angle, such as Schwalbe's line, the trabecular meshwork, the scleral spur, and the ciliary band [14]. 3 Digital Eye Screening Support System 2019 D. Perimeter Perimetry is a visual field test that produces a map of a complete field of vision. The normal visual field has an elongated elliptical shape, and with respect to the fixation point, it measures 60 degrees superiorly and medially, 70-75 degrees inferiorly, and 100-110 degrees temporally. This test determines whether the vision is affected. The two means for measuring the visual field are dynamic and static measurement. Perimeters express the brightness of the target in units of apostilbs (asb) [3]. Figure 1 shows the different types of glaucoma indications. Figure 1: different types of glaucoma indications 1.2. Literature review According to K. Nirmala et.al in Analysis of CDR Detection for Glaucoma Diagnosis to segment the color fundus camera image and calculate the features to segment optic disc and cup separately. The study indicates that according to WHO, Glaucoma is the second leading cause of blindness that contributes to approximately 5.2 million cases of blindness (15% of total blindness. The objective of this study is to detect Glaucoma using K-Means clustering segmentation and Gabor Wavelet Transform of fundus image to obtain the accurate boundary delineation. In order to extract the optic disc and cup, a region of interest around the optic cup and disc must first be delineated. The disc and cup extraction can be performed in the entire image localizing the ROI (region of interest) would help to reduce the computational cost as well as improve the segmentation accuracy. Article by Akira Sawada et.al on Vertical cup-to-disc ratio measurement for diagnosis of glaucoma on fundus images mainly focus on analyzing the optic disc on a retinal fundus image, 4 Digital Eye Screening Support System 2019 which is important for diagnosis of glaucoma. According to this article the vertical C/D ratio is most important factor for diagnosis of glaucoma, because the contrast of the cup region and the rim one was high. Thus, it is needed to measure C/D ratio automatically. Here, glaucoma cases tend to have enlarged cup regions as against the normal cases. The method used in this article is achieved a concordance rate of 86%, extraction rate of 96%, and over-extraction rate of 11% relative to the disc areas determined by an ophthalmologist. In this study, we also presented a method for recognizing glaucoma by calculating C/D ratio. The method correctly identified 80% of glaucoma cases and 85% of normal cases. Although the proposed method is not error-free, the results indicated that it can be useful for the analysis of the optic disc in glaucoma examinations. As JG Hillman et.al stated on central corneal thickness and optic disc relationship in an elderly population, CCT has been shown to be a powerful risk factor for the progression of ocular hypertension and preperimetric glaucomatous optic neuropathy to primary open angle glaucoma. In this study approximately 1500 subjects are participated with aim of determining population based normative CCT data for elderly subjects with and without diabetes, and to find the relationship between CCT and IOP. Both CCT and optic disc was measured by ultrasound pachymetry and morphometrically stated using Heidelberg retinal tomography (HRT2). This study show CCT is positively related to IOP with thinner corneas requiring less force than expected to achieve applanation by Goldmann applanation Tonometry and diabetic patients had significantly greater CCTs than non-diabetic patients. 1.3. Existing solutions Different technical methods are used for diagnosis of eye disease. 1.3.1. Perimeter Perimetry is a visual field test that produces a map of a complete field of vision. The normal visual field has an elongated elliptical shape, and with respect to the fixation point, it measures 60 degrees superiorly and medially, 70-75 degrees inferiorly, and 100-110 degrees temporally. This test determines whether the vision is affected. The two means for measuring the visual field are dynamic and static measurement. Perimeters express the brightness of the target in units of apostilbs (asb) [3]. 5 Digital Eye Screening Support System 2019 1.3.2. Ophthalmoscope It is an instrument for inspecting the interior of the eye. The device devise consists of a strong light that can be directed in to the eye by small mirror or prism. The light reflects of the retina and back through small hole in the ophthalmoscope, through which the examiner sees a nonstereoscopic magnified image of the structures at the back of the eye, including the optic disc, retina, retinal blood vessels, macula and choroid. It is particularly useful as a screening tool for various ocular diseases. Such as: diabetic retinopathy [3]. 1.3.3. Gonioscopy This diagnostic exam helps determine whether the angle where the iris meets the cornea is open and wide or narrow and closed. During the exam, eye drop is used to numb the eye. A hand-held contact lens is gently placed on the eye. This contact lens has a mirror that shows if the angle between the iris and cornea is closed and blocked or open and wide [14]. Gonioscopy is indispensable in the treatment and diagnosis of glaucoma. In Gonioscopy, it is important to properly recognize the various structures composing the anterior chamber angle, such as Schwalbe's line, the trabecular meshwork, the scleral spur, and the ciliary band [15]. 1.3.4. Pachymetery Pachymetery is a simple painless test to measure the thickness of cornea which is the clear window at the front of the eye. A probe called Pachymetery is gently placed on the front of the eye (cornea) to measure thickness. It helps to diagnosis, because corneal thickness has a potential to influence eye pressure reading [16]. 1.3.5. Optical Coherence Tomography Optical coherence tomography (OCT) is an imaging modality that performs high resolution, micrometer-scale cross sections of the eye and other biological structures. This technology presents a clearer picture of the pathophysiology of disease. Also, OCT is useful in determining response to therapy [17]. 1.3.6. Manual screening using 3D lenses Manual screening system is a technique used by the physicians with the help of 3D lens to make diagnosis on the slit lamp or fundus copy. To watch the optic nerve 3D lens is needed. They also analyze the cup to disc ratio manually through experience they guess the diameter 6 Digital Eye Screening Support System 2019 and give the ratio. If the cup is large they estimate that, the patient may have glaucoma. They also relate the result with IOP. 1.4. Gap analysis The above existing solutions have their own gaps. From those; • Manual observation/eye screening • Requires subjective decision making • Most of them are unavailable in health centers of Ethiopia • They are expensive • Not portable • There is delay on diagnosis 1.4.1. Subjective decision In the current technologies, diagnosis and analysis is made by the examiner making the decision subjective. This subjective decision making process is prone to error leading to misdiagnosis. 1.4.2. Accessibility Currently, the devices are less accessible in Ethiopia, like Tonometry, Ophthalmoscopy which is mostly used for diagnosing purpose. They are found in some hospital in limited amount of number. These devices are not found at health centers and in rural areas. This inaccessibility of device leads many peoples to not diagnosed early and become blind. 1.4.3. Cost The costs of devices which are currently used are expensive. To diagnose glaucoma there is a need of different tests. These tests need screening devices. In Jimma medical center there is tonometry device used for pressure measurement and slit lamp microscope used to see the status of the eye and analyze the stage of glaucoma. Although, the devices are useful, there is limited number of those devices due to cost. 1.4.4. Portability The devices are large in size table top devices. The physician and patient must go to the device to diagnose since; the device is not portable. For diagnosis, integrating different devices and 7 Digital Eye Screening Support System 2019 correlate their output is needed, which is very tiresome for patients. In glaucoma most of patients are elders, for them to diagnose in many different wards is very difficult. 1.4.5. Manual diagnosis Currently in Ethiopia, physician most commonly use slit lamp with lens to observe and screen eye looking for abnormalities like retinopathy, glaucoma, neuropathy etc. Analysis of images are done manually and there is no means of storing the images except when using sophisticated and expensive devices like funds camera. Manual diagnosis is subjective which may lead to false estimation. 1.4.6. Time Manual observation of the eye looking for abnormalities is a time consuming procedure. Absence of digital storage system when using traditional screening methods is also another cumbersome. If a physician is unsure with the first decision, he may redo the observation because of lack of a stored image. This increases patient’s waiting time. 1.5. Problem statement Eye disease can be treated if it is caught earlier. Glaucoma, cataract, retinopathy and ARMD are the most occurring eye diseases in Ethiopia. Now days, Glaucoma becomes familiar in Ethiopia. Many people are losing their sights because of glaucoma, especially people who aged above 40 years. These problems come due to lacking habit of early checkup for eyes, unavailability of suitable detection or screening mechanisms in low level health facilities and lack of awareness, even among some health professionals in Ethiopia awareness and understanding of glaucoma is low. The worst scenario is that glaucoma is a chronic disease characterized by irreversible optic nerve damage and visual field loss. Most people go to hospital after the disease aggravates. After advanced stage of glaucoma, the treatment become only medication and surgery to preserve the rest undamaged optic nerve. Thereby, if one come and early diagnosed, his/her sight can be preserved. Currently in Ethiopia, manual eye screening procedure is practiced to diagnose the eye. This method is subjective and prone to error. High-tech devices are not accessible in most health facilities because of their expensive cost. In this project, simple, cost effective, accessible and suitable eye screening support system is proposed. 8 Digital Eye Screening Support System 2019 1.6. Objective of the project and constraints 1.6.1. General Objective The general objective of this project is to develop a digital eye screening support system. 1.6.2. Specific Objectives The specific objectives of this project are: • To digitalize the eye screening system. • To enhance retinal images and segment the optical disk • To provide cost effective, portable and easy to use eye screening system • To help physicians analyze presence of glaucoma and other disease diseases. • To simplify the work load of physicians. 1.6.3. Constraints Constrains of the project are: • Portability: small size • Sensitivity: sensitive to detect small amount of high pressure. • Cost effective: the cost is expected to be < 5000birr • Easy to use: it should not take more than 30 min training to understand • Feasibility: components which are needed to produce are available. • Accessibility: can be accessible in any health organization. 1.7. Significance of the study This project has a great significance in glaucoma disease detection. It has also other significances. 1.7.1. Store images This project enables the physician to store patient information and diagnosis result. It helps for further analysis of image when it is needed. It also helps to simplify follow up since, the physician can see the previous result when the patient comes again. 9 Digital Eye Screening Support System 2019 1.7.2. Location independent Since, the image acquiring and storing in computer is possible, it allows the clinicians to review same data at the same time even though, they are found in different locations. This enables health center health officers to ask assistance and further analysis from residents or specialists. 1.7.3. Time saving The current technology does not provide analyzation rather only input image. Analyzation is the physicians work. Therefore, watching the image and analyze it takes some period of time. This project provides simple and fast analyzer system which, used to decrease waiting time and suffering due to long waiting time. Many elders should have been diagnosed without any long bother. 1.7.4. Reliable result The current technology does not capture image and do analysis. The physician watches directly in to eye through slit lamp and analyze the coming image by him/her self. This result unreliable result sometimes and when the physician is in rush he/she may not see detail of the eye and forget some features. However, this project gives quality image that passes under image several image processing steps. If the ophthalmologist needs to see extracted image, he can see it easily by pressing the graphical interfacing module. Example, optic nerve can be seen by this technology. 1.7.5. User friendly This system is easily understandable by the user since, there is no complexity in the system. The device will finish almost more than 50% of physician work. Therefore, the physician will have satisfaction and dedication to his/ her work. It also provides comfortable diagnosing system for both the user and the patient. The user can sit without bending towards the slit lamp and the patient can even sleep or be in his supine position. Thereby, it is comfortable and user friend system. 1.8. Scope of the study This system mainly focused on elder people or people who aged greater than 40 years. But glaucoma sometimes happens in children also. 10 Digital Eye Screening Support System 2019 1.9. Breakthrough When we went to the hospital, we wonder to visit the ophthalmology department. Because, we heard about glaucoma and many other eye diseases. While visiting, we saw an ophthalmologist who were diagnosing a patient eye through slit lamp. He was stopped, and he took many time to diagnose one person. We asked him and he told us, that he works this for 5 years and many times a day. This will cause him back pain and also other eye problems. Therefore, we inspired to design eye screening support system which give comfort to physician with processed image and IOP integration. This method is also good for the patient also. 11 Digital Eye Screening Support System 2019 CHAPTER TWO DESIGN STRATEGY 2.1. Background information The leading causes of visual impairment are primarily age-related eye diseases including cataracts, diabetic retinopathy, glaucoma, and age-related macular degeneration. Glaucoma is an acquired disease of the optic nerve (neuropathy) characterized by specific changes of the optic nerve and by visual field defects that correspond to the areas of optic nerve structural damage. Without showing any symptom early stage glaucoma will damage optic nerve. Mostly patients don’t know the disease until it reaches the advanced stage. The damage caused by glaucoma is irreversible, but if it detected earlier the treatment can prevent its progression. Thus, early detection of glaucoma is essential. The goal of this project is to provide an accurate, low cost, easy to use, fast and portable eye disease screening device based on intraocular pressure estimation and image processing with the aim of decreasing blindness due to different eye disease. This device is mainly used in government or private hospital, health sector and clinics. In order to do these different alternatives was provided such as Wearable and wireless intraocular pressure measurement, Integrating the slit lamp (bio microscope) with computer application, Image acquisition, enhancement and extraction system for eye disease detection. 2.2. Evaluation of brainstormed idea 1. Wearable and wireless intraocular pressure measurement 2. Image acquisition, enhancement and extraction system 3. Integrating the slit lamp (bio microscope) with computer application 2.2.1 Wearable and wireless intraocular pressure measurement Intraocular pressure is a fluid pressure in the eye. It is an important aspect in the evaluation of patient at a risk of glaucoma. Eye pressure is measured in mmHg. Normal eye pressure ranges from 12-22 mmHg, and eye pressure of greater than 22 mmHg is considered higher than normal. Although the person does not show sign of glaucoma, this is referred to that ocular hypertension. This possible solution measures the eye pressure and suspect 12 Digital Eye Screening Support System 2019 probability glaucoma. This method is done by, attach the sensor at eye glass. It acquires the pressure difference at the iris and measure the distance between iris and cornea. These inputs will inter to the processing unit and analyzed. On the LCD length between iris and cornea, the eye pressure and comments will be displayed as an output. Figure shows the block diagram of this technique Wear the eye glass Receive signal wirelessly & ADC Acquiring parameters Out put Cornea length Eye pressure Figure 2: block diagram of IOP measurement Strength • Since it is non- contact with eye, its comfortable • Safe • Easy to use by everyone Weakness The main problem of this device is its accuracy. It measures the eye pressure only which is one of the risk factor of glaucoma and sometimes in normal eye pressure also glaucoma may happen. Required Materials • Pressure sensor • Eye glass • Distance sensor • Analogue to digital convertor • LCD • Electrical components 13 Digital Eye Screening Support System 2019 2.2.2 Image acquisition, enhancement and segmentation system for glaucoma detection In this method, a 3D lense will be attached with a mini camera to acquire high-quality images of the eye. The acquired images will be enhanced, segmented and cup to disc ratio is calculated to diagonise glacoma. In addition pressure sensor will also be incorporated to acquire intra-ocular presure. Graphical user interface (GUI) will be develped to acquire image and pressure signal, enhance and exract the image. The system will be developed using MATLAB. The result will be displayed on the computer. GUI Camera with lens Image acquisition Processing unit Image enhancement Pressure sensor Image segmentation Result Image CDR LCD Pressure value Figure 3: block diagram of image processing method Strength • Accurate, portable and easy to use • Cost effective • Save image and personal information is possible • Can diagnose other eye disease like retinopathy, degenerative eye diseases etc. • Improve the quality of diagnosis • Used in health center Weakness • Do not analyze image 14 Digital Eye Screening Support System 2019 • Self-diagnose not possible Required Materials • Camera • 3D lens • Raspberry pi 3 • Pressure sensor • Touch screen LCD • Electrical components 2.2.3 Integrating the slit lamp (bio microscope) with computer application This method is done by acquiring image from the slit lamp (bio microscope) and enhances that image and display in the computer. Therefore, the patient sits on the slit lamp instrument and the physician can analyze anywhere he want. Camera attached to slit lamp Image acquisition from slit lamp Computer application Image processing Processed image displayed Computer Figure 4: block diagram of slit lampapplication and computer integration for analysis Strength This method will: Computer application • Decrease the physician load • Increase time • Increase quality of diagnosis • prevents the physician from back pain and sight problem Computer application Weakness • Cost • Need slit lamp device • Need ophthalmologist Computer application Computer application 15 Computer application Digital Eye Screening Support System 2019 Required Materials • Camera • Slit lamp • PC app From the above possible solutions image acquisition, enhancement and extraction system for glaucoma detection method is selected using Pugh Matrix. Pugh matrix for screening proposed concept Evaluation criteria Wireless IOP Image processing measurement Integrating slit lamp with comp application Accuracy 0 1 1 Cost effectiveness 1 0 1 Portability 1 1 1 Accessibility 1 1 -1 Easy to use 0 1 0 Feasibility -1 1 1 Sensitivity 0 1 0 Total 0 = 3, -1 = 1, +1 = 3 0 = 1, -1 = 0, +1 = 6 0 = 2, -1 = 1, +1 = 4 Rank 3 1 2 Continue…??? No Yes Revised Table 1: screening Pugh matrix 2.3. How the matrix is obtained? Our baseline is the system we have in place at the moment, so we score this a null (zero) against our criteria. It is used as a reference. Now consider option 1. In relation to criteria 1 (Accuracy), do we consider that it is better, the same as, or worse than the baseline? If it's better, we give it a 16 Digital Eye Screening Support System 2019 +; if it's the same, we give it a 0; and if it's worse, we give it a -. As we can simply understand from the second design which is image processing, it is accurate. Therefore, it is better, and we gave a value of +. In terms of criteria 2(cost), it's worse than the baseline. For criteria 3(portability), 4(Accessibility), 5(easy to use) and criteria 6 it is better for the last criteria (sensitivity), it is better. Thus the three designs are screened. Next, we multiply all values in the above table by the corresponding weight. Design criteria weight (priority) Accuracy 0.29 Cost effective 0.24 Portability 0.19 Accessibility 0.14 Easy to use 0.07 Feasibility 0.063 Sensitivity 0.01 Pugh matrix for concept screening Feature Pressure Image Bio microscope with measurement enhancement computer app 0.29 1 0.29 3 0.87 2 0.58 Cost effectiveness 0.24 3 0.72 2 0.48 1 0.24 Portability 0.19 2 0.38 3 0.57 1 0.19 Accessibility 0.14 2 0.28 3 0.42 1 0.14 Easy to use 0.07 3 0.21 2 0.14 1 0.07 Feasibility 0.063 1 0.063 3 0.189 2 0.126 Sensitivity 0.01 1 0.01 3 0.03 2 0.02 Accuracy Total Weight 1.953 2.7 Table 2: scoring Pugh matrix 17 1.366 Digital Eye Screening Support System 2019 2.4. Proposed initial design In this project we used the system called image acquisition, enhancement and extraction system. Here, a 3D lens will be attached with a mini camera to acquire high-quality images of the eye. The acquired images will be enhanced, segmented and extracted to diagonise glaucoma. In addition sharp IR sensor also be incorporated to acquire intra-ocular presure. Graphical user interface (GUI) will be develped to acquire image and pressure signal, enhance and exract the image. The system will be developed using Raspberi pi. Rasperi pi compatible Touch screen LCD display will be used to display results.the procedure are followed image acquisission, preprocessing, segmentation, edge detection and cup-to-disc ratio. Algorithm/ flow chart Figure 5: algorithm or flow chart 18 Digital Eye Screening Support System 2019 2.3.1. Image acquisition Retinal image of an eye is acquired by USB camera. Cameras typically use 2D (area) Chargecouple device sensors, whereas scanners employ 1D (line) CCDs that move across the image as each row is scanned. CCDs have become the sensor of choice in many imaging applications because they do not suffer from geometric distortions and have a linear response to incident light [18]. Figure 6: image of normal eye and glaucoma eye 2.3.2. Image Enhancement It includes varying brightness and contrast of image. It also includes filtering and histogram equalization. It comes under pre-processing step to enhance various features of image. Here we use gamma correction or power law. The power law transformation function is described by π = π. π^πΎ where r is the original pixel value, s is the resulting pixel value, c is a scaling constant, and γ is a positive value. [18]. 2.3.3. Edge detection Edge detection is a fundamental image processing operation used in many computer vision solutions. The goal of edge detection algorithms is to find the most relevant edges in an image or scene. These edges should then be connected into meaning fullness and boundaries, resulting in a segmented image containing two or more regions [18]. The process of edge detection consists of three main steps [18]; Noise Reduction: Due to the first and second derivative’s great sensitivity to noise, the use of image smoothing techniques before applying the edge detection operator is strongly recommended. 19 Digital Eye Screening Support System 2019 Detection of Edge Points: Here local operators that respond strongly to edges and weakly elsewhere are applied to the image, resulting in an output image whose bright pixels are candidates to become edge points. Edge Localization: Here the edge detection results are post processed, spurious pixels are removed, and broken edges are turned into meaningful lines and boundaries, using techniques such as the Hough transform There are different types of Edge detection methods • First-order derivative edge detection • Second-order derivative edge detection • The canny edge detector • Edge linking and boundary detection From those we use second-order derivative edge detection, the canny edge detector and edge linking and boundary detection [18]. 2.3.4. Image segmentation Segmentation is the process of partitioning an image into a set of no overlapping regions whose union is the entire image. These regions should ideally correspond to objects and their meaningful parts, and background [18]. Watershed segmentation Watershed transform decomposes an image completely and thus assigns each pixel either to a region or a watershed. The watershed transform is used to represent regions in a segmented image (equivalent to catchment basins) and the boundaries among them (analogous to the ridge lines) [18]. 2.3.5. Cup-to-disc ratio To calculate the vertical cup to disc ratio, the optic cup and disc first have to be segmented from the retinal images. The green plane of the registered image is extracted to choose mean value for background blood vessel, cup and disc are replaced for segmenting the image. Using concatenate function, the images are mapped in a set of 4 iterations to execute for the above set of mean values. The identified mean value is replicated with the mean value within each of the array and then the 20 Digital Eye Screening Support System 2019 distance matrix is calculated. Small discs have small cups with a median C/D ratio of about 0.35 and Large discs have large cups with a median C/D ratio of about 0.55 [9]. 2.3.6. IOP measurement IOP is determined by the balance between the rate of aqueous secretion and aqueous outflow. Distribution of IOP is within a range of 11-21mmHg. Normal IOP varies with the time of day, heartbeat, blood pressure level and respiration. If the value is above this range it viewed with suspicion [9]. Among glaucoma patients, most of the cases the IOP increase to above 30mmHg, and some even higher than 100mmHg.we use Arduino and sharp IR sensor to determine the pressure of the eye by using biomechanics concept which is. π=ππ/Δπ [19]. π is elastic modulus; π is stress, and πis initial length. Initial length is 10.03cm since the sensor putted at a distance of 10cm and the distance of cornea to the eyelid is 0.03mm [20]. young’s modulus is 7500.6mmHg [20]. We will get the stress (π) value from the sensor. Therefore, we can calculate the change in length by using π=Δπ*π/π and we can get new length π by adding Δπ with initial length of π = 10.03cm. 2.4. Components lists 2.4.1. Hardware part • Lens (Volk 90D) • Camera • Arduino • LCD • sharp IR sensor Lens(volk90D) The 90D non-contact slit lamp lens is intended for general diagnosis and small pupil examinations, and features patented double aspheric glass optics that provides enhanced imaging of the retina. 21 Digital Eye Screening Support System 2019 Figure 7: volk 90D Charge coupled device (CCD) It works by converting light into a pattern of electronic charge in silicon chip. This pattern of charge is converted into a video wave form, digitalized and stored as an image file on a computer. Figure 8: CCD Arduino UNO Arduino Uno is a microcontroller board based on the ATmega328P (datasheet). It has 14 digital input/output pins (of which 6 can be used as PWM outputs), 6 analog inputs, a 16 MHz quartz crystal, a USB connection, a power jack, an ICSP header and a reset button. It contains everything needed to support the microcontroller; simply connect it to a computer with a USB cable or power it with an AC-to-DC adapter or battery to get started [22]. Figure 9: Arduino 22 Digital Eye Screening Support System 2019 LCD LCD is the technology used for displays in note book and other smaller computers. LCD consumes much less power than LED and gas display because they work on the principle of blocking light rather than emitting it. In this project 16*2 used for displaying all result, user guide and instruction for users. We used LCD to display the values of corneal distance and IOP [23]. Figure 10: LCD Sharp IR sensor It is a distance measuring sensor unit, it measures from 10cm to 80 cm and composed of an integrated combination of PSD (positive sensitive detector), IRED (infrared emitting diode) and signal processing circuit. The variety of the reflectivity of the object, the environmental temperature and the operating duration are not influenced easily to the distance detection because of adopting the triangulation method. To convert the distance value to pressure biomechanics principle is used. Initial length is 10.03cm, young’s modulus is 7500.6mmHg. Distance value can be acquired from sharp IR sensor and calculate the pressure by using π=Δπ*π/π formula. π is stress, Δπ is change in length, π is elastic modulus, and π is length. Therefore, the sensor must be very sensitive for the minimum distance variation. Figure 11: Sharp IR sensor 23 Digital Eye Screening Support System 2019 Potentiometers It is a three-terminal resistor with a sliding or rotating contact that forms an adjustable voltage divider If only two terminals are used, one end and the wiper, it acts as a variable resistor. The measuring instrument called a potentiometer is essentially a voltage divider used for measuring electric potential (voltage). We used it to adjust brightness of LCD [24]. Figure 12: potentiometer 2.4.2. Software part MATLAB: (short for MATrix LABoratory) is a special-purpose computer program optimized to perform engineering and scientific calculations. It started life as a program designed to perform matrix mathematics, but over the years it has grown into a flexible computing system capable of solving essentially any technical problem [25]. 2.5. Implementation plan For the implementation of the device, materials are obtained from Biomedical Engineering laboratories, from Jimma Medical center and also lending some materials. We are supported by our advisor Dr. Gizeadiss Lamesgin for collecting materials. There were no challenges regard to materials. After constructing the prototype, we want to test want to test our goals and constraints, such as accuracy, cost effectiveness, easy to use and sensitivity. 24 Digital Eye Screening Support System 2019 CHAPTER THREE FINAL DESIGN Our final design was completed as proposed on the first step of designing plan, aimed at obtaining user friendly, accurate, portable and safe eye screening support system. The design includes camera and lens holding system, controller system (Arduino and IR sensor) and software. The wood-made support system is used both as a light guidance system that incorporates slits (small openings) and also as a support system for the camera and 3D lens. Arduino integrated with the IR sensor is used to measure intra-ocular pressure. The result of IR sensor is displayed using 16X2 LCD. The software part is designed for ease of use. The Matlab developed GUI includes image acquisition, Enhancement, Edge detection, Segmentation and analysis part. 3.1. Steps for construction of prototype Step 1. Material collection that are required for prototype construction from biomedical department, our advisor and ophthalmology department. Step 2. Develop flow chart and Algorism for image processing part as well as IOP measurement. Step 2. Develop MATLAB GUI for image processing. Step 3. Design the IOP measurement circuit part. Step 4. Develop Arduino code for IOP measurement. Step 5. Design lens holder and camera by wood. Step 6. Capture image and test the system. Step 7. Connect IOP circuit which are sharp 0A41SKF1Z IR, Arduino and LCD on breadboard. Step 8. Finally, iteration and test done. 3.2. Iteration and design changes The device iteration and the modification of equipment listed on the table below. 25 Digital Eye Screening Support System 2019 No Design parts Proposed Implemented 1 Arduino Uno Raspberry pi Arduino Uno 2 LCD display Touch screen LCD 16*2 3 Sharp 0A41SKF1Z Sharp 0A41SKF1Z IR Sharp 0A41SKF1Z IR IR sensor sensor sensor 4 Lens Volk 90D Volk 90D 5 Camera Raspberry pi camera 6 Potentiometer Potentiometer Remarks Unavailability Unavailability Unavailability Potentiometer Table 3: Iteration of the device Figure 13: prototype construction: A) shows IOP design on Proteus B) Prototype construction of image C) graphical user interface D) final simulation Part 26 Digital Eye Screening Support System 2019 3.3. Final product Figure 14: final prototype This system is a dual function system one for image processing part and the other for pressure measurement. Therefore, two devices function can be substituted by it which are tonometeric function and slit lamp functions. To achieve this image enhancement, edge detection, segmentation and cup to disc ratio. Thereby, glaucoma, cataract, ARMD and retinopathy can be diagnosed by watching the segmented image and cup to disc ratio. The pressure analysis only useful for glaucoma. On the prototype part simple parts are integrated camera, volk 90D and computer for processing are integrated. Sharp IR sensor and LCD integrated with Arduino to give IOP for glaucoma. 3.4. Final components list 1. Arduino Uno 2. LCD 16*2 27 Digital Eye Screening Support System 2019 3. Sharp 0A41SKF1Z IR sensor 4. Volk 90D 5. USB cam 3.5. Graphical user interface The operation window is graphical user interface part. The code is generated using MATLAB it has eight push button and four axes. Each push buttons are designed for unique function. Pushbutton 1- 3…. image acquisition Push button 4…. Image save Push button 5…. Image Enhancement Push button 6…. Edge detection Push button 7 …. Segmentation Push button 8 …. For analysis of cup to disc ratio and diagnose its normality. Figure 15: graphical user interface image 28 Digital Eye Screening Support System 2019 Image acquisition Image can be acquired via camera from USB camera or loading image from folder. After Pressing push button 2 preview will be active and it can load image if we enable the loading portion of code. It also acquire from camera by enabling capturing code part. Then image can be saved via push button 4. Figure 16 shows the image acqusition user interface. Figure 16:image acquisition in GUI 29 Digital Eye Screening Support System 2019 Image enhancement Image enhancement done by using power law, which uses gamma correction factor. It can correct the camera error or noise by applying correction factor πΎ of 1.2. This part enables the physician to see enhanced image from original. Figure 17 shows the image enhancement user interface. Figure 17: Image enhancement on GUI 30 Digital Eye Screening Support System 2019 Edge detection Edge detection is found on push button six and it gives out extracted image of the edge. On the retina edge detection can extract the blood vessels and disc part. This help the physicians to see all edges clearly. The edge detection processes designed by ‘sobel’ operators, which is one of image processing technique. Figure 18 shows the edge detection user interface. Figure 18: edge detection on GUI 31 Digital Eye Screening Support System 2019 Image segmentation It is a major part of image processing part which helps to analyze several kinds of eye diseases by watching the image. It is found on push button seven and uses watershed algorithm which identify rough area from smooth one. Therefore, if there is tumor or other odd structure on retinal image, it shows clearly. On normal condition it segment the disc of retina and display on axes 3. Figure 19 shows the image segmentation user interface. Figure 19: image segmentation on GUI 32 Digital Eye Screening Support System 2019 Analysis This portion is a final part of image processing which is analysis small diameter circle and large circle area ratio called cup to disc ratio or CDR. In normal condition cup to disc ratio must be 0.3 and less. If cup increased , CDR also increased which may be an implication of glaucoma. Figure 20 shows the image analysis user interface. Figure 20: cup to disc ratio on GUI 3.6. Simulation and results The IOP sensor and controller are simulated using Proteus software. IR sensor from Proteus is used to simulate the pressure sensor. The output of the IR sensor is fed to the Arduino. Young’s modulus of the cornea using biomechanics principle is used to convert the measured distance into pressure. The output pressure is displayed in LCD. Figure 21 A and B below show the simulation part. 33 Digital Eye Screening Support System 2019 Figure 21: A) Proteus results for normal B) IOP Proteus results for high IOP 34 Digital Eye Screening Support System 2019 CHAPTER FOUR TESTING AND RESULTS To ensure that a system meets its designed specifications and other requirements, a test plan was used. Design criteria specified by the developers of the system or project are basement for testing plan. Specifically, in context of this project definition and contents of design criteria is discussed below. 4.1. Design criteria In order to achieve the desired goal of the project, it is mandatory to follow standards that make listed potential solutions tangible standards and some additional criteria as listed below. 4.1.1. Accuracy The existing solutions for glaucoma detection is manual, time wasting and physically not safe for the physician. Due to this the result may face some inaccuracy. Therefore, this project considers the above features and solves the accuracy problem by capturing image of retina and passes through several image processing method and give cup to disc ratio in number. 4.1.2. Cost Engineering designs must be cost effective, Especially, for developing countries as Ethiopia. This project minimizes the cost of slit lamp and tonometer with simple device. Since its cost effective it can be used by many health centers in Ethiopia. 4.1.3. Portability The existing solutions are heavy table top devices this makes them difficult to move. This project uses light materials and system which is easy to move to other computer or laptop if it has Matlab. 4.1.4. Easy to use This device is user friendly. It can be used by physicians with simple computer training. For image acquisition resolution selection can be done by clicking push button. The other push buttons also have their own functions. Such as, image enhancement, edge detection, segmentation, cup to disc ratio. 35 Digital Eye Screening Support System 2019 4.1.5. Sensitivity The device must be sensitive for small amount of change corneal distance to calculate the IOP which is a pre indicator of glaucoma disease. Therefore, sensitivity must fulfill for the device. 4.1.6. Time saving Time is main thing for health facility since, many patient need to diagnosed and treated as fast as possible without further process. Although, the existing solutions are manual and time consuming. But this project gives a digital system which save time and make the diagnosing system easy. 4.2. Testing plan This section verifies and ensure that a product or system meets its design specifications and other requirements. A test plan is usually prepared by significant input from group and advisors. Depending on what is to be tested, different methods will be followed. 4.2.1. Market Testing Market testing is used in determination of the cost for each component included in the project. This test is done by summing the price of each component which we use for construction of prototype and will be comparing with predetermined device cost which was 1710 ETB. 4.2.2. Accuracy test The main criteria of device to be accepted are accuracy in its measurement. To test the accuracy of device we use comparison method that is output data from measurement will be compare to the output of three or four ophthalmologist doctors from slit lamp or existing technology. 4.2.3. Test for user friendly To test the system can be operated by everyone we invite a guest he/she have no idea about package or never see before, then asses how much he/she can operate the device without any confusion and error using graphical user interface. 4.2.3. Sensitivity test Sensitivity test done by checking the sensor variability for each distance change. To analyze sensitivity several iterations must be done. After several iterations if it gives precise result it can be estimated as high sensitive device. If not, it cannot be possible to identify the change in distance and IOP. 36 Digital Eye Screening Support System 2019 4.3. Testing conducted The following table shows that how criteria are tested and evaluated. No Feature to tested Input Method Design Result specification 1 Cost Market Total cost < 4181 ETB 1925 ETB was added (Which is cost effective) 3 Time Enumerating Stop watch < 1minute 40s the time 4 User friend Simple Train the Can procedure physicians be Simple understandable by use and see how ophthalmologists they understand it Table 4: testing plan and result 37 to Digital Eye Screening Support System 2019 CHAPTER FIVE CONCLUSION AND RECOMMENDATION 5.1. Conclusion Human eye is a vital organ of vision which reacts to light and pressure. The eye plays a very important role not only in life but also in human body, also it gives us the sense of sight, allow us to see and interpret colors, dimension of object. There are various things that affect structure of the eye, cause visual impairment and blindness. From them, glaucoma, cataract, age related macular disease, and retinopathy are the main cause of blindness. Therefore, to diagnose these diseases early we design an eye screening support system. This system done by image processing and biomechanics law. Image processing part follows image acquisition, enhancement via power law, edge detection with Sobel operator, segmentation using watershed algorism and cup to disc ratio (<=0.3 is normal) and display each result on the computer with graphical user interface. For each process there is pushbutton which enable it to display the processed image. It has eight push buttons to enable the image processing axes and image displayed on axes. Biomechanics law used for glaucoma measurement only since eye pressure is an indicator of glaucoma. This method done by acquiring corneal distance via sharp IR sensor 0A41SKF1Z and correlate with initial distance, modules and change in length between original distance and change in length between new acquired length and the initial length. IOP calculated by the following formula; π=Δπ*π/π. The normal range of IOP is from 10-22 mmHg. Segmentation helps to diagnose ARMD and also retinopathy. Therefore, our device gives a dual purpose and work us a tonometer and slit lamp. To achieve this material collection from our advisor, biomedical engineering department, ophthalmology department which is found at Jimma Medical Center and electrical engineering department. After that we developed MATLAB and Arduino code for graphical user interface of image processing part and IOP part respectively. Then after, the circuit design on Proteus done and several iteration process done on the simulation. Finally, connecting components with Arduino and final prototype done. After building the prototype, some testing procedures were done. User friendly test, market test, and user friendly test are among them. The testing result shows that, the device costs 1710 ETB 38 Digital Eye Screening Support System 2019 which is very cost effective, and time saving with in less than 40 seconds and which sensitive as well as user friendly for the users. 5.2. Recommendations This project does image processing until segmentation, but it does not classify image as glaucoma, cataract, ARMD or other for self-diagnosis comparing the processed image with several train images which saved on that application. Therefore, in the future: • Self-diagnosis by using image classification • Incorporating the system with mobile application • Apply the concept of telemetry and telemedicine. • Include temperature difference and other parameters to make IOP measurement more accurate. 39 Digital Eye Screening Support System 2019 References [1] A. A. Dahl., "emedicinehealth," 19 october 2018. [Online]. https://www.emedicinehealth.com/anatomy_of_the_eye/article_em.htm/. Available: [Accessed 30 march 2019]. [2] C. S. McCaa, "The eye and visual nervous system," Environmental Health Perspectives, vol. 44, pp. 1-8, 1982. [3] e. Dnyaneshwari D. Patii, "Diagnose glaucoma by proposed image processing method," international journal of computer application , Vols. 106-No. 8, p. 14, November, 2014. [4] J. D. B. et.al, Willy series in pure and applied optics, Newyork, 1975. [5] Carmen Willings, "structure & function of the eye," [Online]. Available: http://www.teachingvisuallyimpaired.com/structure-function-of-the-eye.html. [Accessed 28 March 2019]. [6] e. Dr. V. Suresh Babu, "Detection of gaucoma in retinal image using raspberry pi," Asian journal of applied science and technology (AJAST), vol. 2, no. 2, p. 395, April, 2018. [7] e. Coleman A., "Open opthalmol J.," september 2009. [Online]. Available: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2759104/. [Accessed 1 april 2019]. [8] D. M. Devin, "The demise of direct ophthalmoscopy," april 2015. [9] A. P. Ken Nischal, Clinical Opthalmology, UK: Elsevier, 2011. [10] G. Bruno, "Eye disorders," huntington college of health sciences , 2009. [11] "ophalmology procedures manual," national health and nutrition examination survey, 2005. [12] d. o. opthalmology, "ophtalmology and eye disease," 2016-2017. [13] G. Nicole, "what is slit lamp," april 2018. 40 Digital Eye Screening Support System 2019 [14] W. L. M. A. MD, "Principles of Gonioscopy," Color Atlas of Gonioscopy, 2017. [15] R. Kara, "ophthalmoscope," Encylopaedia britannica, 05 November 2014. [Online]. Available: https://www.britannica.com/technology/ophthalmoscope#accordion-article- history. [Accessed 02 April 2019]. [16] A. C. Imitiaz, "measurement of central corneal thickness in health and disease," pp. 179-180, 23 october 2009. [17] S. B. a. A. Agarwal, "opthalmic instrument and diagnostic tes," color atlas of ophthalmoscopy , pp. 488-506. [18] O. Marques, practical image and video processing using matlab, florida: JOhn Wiley and Sons, Inc, 2011. [19] L. Waite, Applied biofluid Mechanics, USA, 2007. [20] W. A. v. d. Bosch, "topographic anattomy of the eye lids, and the effects of sex and age," Br ophthalmol, vol. 83, pp. 347-352, 1999. [21] a. alsheikh, "determination of modulus of elastisity if the human cornea," journal of refractory surgery, 2007. [22] "Arduino UNO Rev3," [Online]. Available: https://store.arduino.cc/arduino-uno-rev3. [Accessed 24 May 2019]. [23] 2. c. L. d. w. y. b. light. [Online]. Available: https://www.hwkitchen.com/products/lcddisplay-2*16-characters/. [Accessed 24 may 2019]. [24] J. Yong, testine electronic components. [25] S. J.chapman, matlab programing with applicaions for engineers, USA: global engineering, 2013. [26] N. Christian, "What is glaucoma," 15 December 2017. 41 Digital Eye Screening Support System 2019 [27] M. A. e. Abiye, "epidimology of glaucoma in central Ethiopia," Ethiopia , 2009. [28] B. e. Frank, "imaging technologies for the diagnosis of Glaucoma," p. 16, 21 December 2009. 42 Digital Eye Screening Support System 2019 Appendices Appendix A Pair wise comparison chart no Feature Sensitivity Easy Accuracy Accessibility Feasibility portability Cost to Total effective use 1 Sensitivity X 0 0 0 0.17 0 0 0.17 2 Easy to use 1 X 0 0 0.5 0 0 1.5 3 Accuracy 1 1 X 1 1 1 1 6 4 Accessibility 1 1 0 X 1 0 0 3 5 Feasibility 0.83 0.5 0 0 X 0 0 1.33 6 Portability 1 1 0 1 1 X 0 4 7 Cost 1 1 0 1 1 1 X 5 effective Table 5: pair wise comparison chart Total =21 β’ Weight = feature/total 43 Digital Eye Screening Support System 2019 Appendix B Sharp IR sensor datasheet Figure 22: Sharp IR sensor Datasheet A Figure 23:Sharp IR sensor Datasheet B 44 Digital Eye Screening Support System 2019 Appendix C Camera specification Figure 24: camera specification 45 Digital Eye Screening Support System 2019 Appendix D Arduino code int IR=A0; float distancevalue; float length_change; float IOpressure; float initial_length=10.03; float Y=7500.6; LiquidCrystal lcd(12,11,5,4,3,2); void setup() { lcd.begin(16,2); } void loop() { distancevalue=analogRead(IR)*0.0048828125; length_change=distancevalue-initial_length; IOpressure=(length_change*Y)/initial_length; lcd.setCursor(0,0); lcd.print("corneadis="); lcd.print(distancevalue); lcd.setCursor(12,0); lcd.print("cm"); lcd.setCursor(0,1); lcd.print("IOP="); 46 Digital Eye Screening Support System 2019 lcd.print(IOpressure); lcd.setCursor(10,1); lcd.print("mmHg"); if(IOpressure<10) { lcd.setCursor(12,1); lcd.print("LowP"); delay(50); } if(IOpressure>=10 && IOpressure<=22) { lcd.setCursor(12,1); lcd.print("well"); delay(50); } if(IOpressure>22) { lcd.setCursor(12,1); lcd.print("HighP"); delay(50); } } 47 Digital Eye Screening Support System 2019 Appendix E MATLAB code function varargout = Glacouma(varargin) % GLACOUMA MATLAB code for Glacouma.fig % GLACOUMA, by itself, creates a new GLACOUMA or raises the existing % singleton*. % H = GLACOUMA returns the handle to a new GLACOUMA or the handle to % the existing singleton*. % GLACOUMA('CALLBACK',hObject,eventData,handles,...) calls the local % function named CALLBACK in GLACOUMA.M with the given input arguments. % GLACOUMA('Property','Value',...) creates a new GLACOUMA or raises the % See also: GUIDE, GUIDATA, GUIHANDLES % Edit the above text to modify the response to help Glaucoma % Last Modified by GUIDE v2.5 08-Jun-2019 15:00:29 % Begin initialization code - DO NOT EDIT gui_Singleton = 1; gui_State = struct('gui_Name', mfilename, ... 'gui_Singleton', gui_Singleton, ... 'gui_OpeningFcn', @Glacouma_OpeningFcn, ... gui_State.gui_Callback = str2func(varargin{1}); end gui_mainfcn(gui_State, varargin{:}); end 48 Digital Eye Screening Support System 2019 % End initialization code - DO NOT EDIT % --- Executes just before Glacouma is made visible. function Glacouma_OpeningFcn(hObject, eventdata, handles, varargin) % This function has no output args, see OutputFcn. % hObject handle to figure % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % varargin command line arguments to Glacouma (see VARARGIN) %set(handles.axes1,'Visible','off') %set(handles.axes2,'Visible','off') % Choose default command line output for Glacouma handles.output = hObject; % Update handles structure guidata(hObject, handles); % UIWAIT makes Glacouma wait for user response (see UIRESUME) % uiwait(handles.figure1); % --- Outputs from this function are returned to the command line. function varargout = Glacouma_OutputFcn(hObject, eventdata, handles) % varargout cell array for returning output args (see VARARGOUT); % hObject handle to figure % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Get default command line output from handles structure 49 Digital Eye Screening Support System 2019 % handles structure with handles and user data (see GUIDATA) D=menu('Select Video Input Device:',IA); if isempty(IA)||D==0 msgbox IA=IA(D,:); IA(IA==' ')=[]; x=imaqhwinfo(IA); try DeviceID=menu('Select Device ID',x.DeviceIDs); F=x.DeviceInfo(DeviceID).SupportedFormats; nF=menu('Select FORMAT',F); warndlg({'Try Another Device or ID ';... 'You Donot Have Installed This Device(VideoInputDevice)'}) return end % --- Executes on button press in axes1button. function pushbutton2_Callback(hObject, eventdata, handles) global IA DeviceID Format s try VidObj= videoinput(IA, DeviceID, Format); nBands = get(handles.VidObj, 'NumberOfBands'); set(handles.axes1,'Visible','off') axes(handles.axes1) 50 Digital Eye Screening Support System 2019 preview(handles.VidObj, hImage) catch E msgbox({'Configure The s Correctly!',' ',E.message},'s INFO') end guidata(hObject, handles); % --- Executes on button press in pushbutton6. function pushbutton9_Callback(hObject, eventdata, handles) % hObject handle to pushbutton9 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) s=getimage(handles.axes1); s=rgb2gray(s); %s=im2double(s); %gradient gmag = imgradient(s); %watershade L = watershed(s); Lrgb = label2rgb(L); %opening - closing reconstruction Iobrd = imdilate(Iobr,se); %Regional Maxima of Opening-Closing by Reconstruction fgm = imregionalmax(Iobrcbr); %Regional Maxima Superimposed on Original Image 51 Digital Eye Screening Support System 2019 %I2 = labeloverlay(Io,fgm); se2 = strel(ones(5,5)); fgm2 = imclose(fgm,se2); fgm3 = imerode(fgm2,se2); %Modified Regional Maxima Superimposed on Original Imag fgm4 = bwareaopen(fgm3,20); %I3 = labeloverlay(s,fgm4); %Thresholded Opening-Closing by Reconstruction %bw = imbinarize(Iobrcbr); threshold = graythresh(Iobrcbr); bw=im2bw(Iobrcbr,threshold); %Watershed Ridge Lines D = bwdist(bw); DL = watershed(D); bgm = DL == 0; % Compute the Watershed Transform of the Segmentation Function. gmag2 = imimposemin(gmag, bgm | fgm4); L = watershed(gmag2); %Markers and Object Boundaries Superimposed on Original Imag labels = imdilate(L==0,ones(3,3)) + 2*bgm + 3*fgm4; %I4 = labeloverlay(s,labels); %Colored Labels Superimposed Transparently on Original Image 52 Digital Eye Screening Support System 2019 axes(handles.axes3) %hold on %himage = imshow(Lrgb); %himage.AlphaData = 0.3; % --- Executes on button press in pushbutton5. function pushbutton5_Callback(hObject, eventdata, handles) % hObject handle to pushbutton5 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) s=getimage(handles.axes1); F= s;%imresize(F,0.5); axes(handles.axes2) imshow(I,'parent',handles.axes2) % --- Executes on button press in pushbutton6. function pushbutton6_Callback(hObject, eventdata, handles) % hObject handle to pushbutton6 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) %%%%%%%%%% DialateBloodVessel( RGB ) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% s=getimage(handles.axes2); %s=rgb2gray(s); axes(handles.axes2) 53 Digital Eye Screening Support System 2019 imshow(BW1,'parent',handles.axes2); % --- Executes on button press in pushbutton5. function pushbutton8_Callback(hObject, eventdata, handles) % hObject handle to pushbutton8 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) s=getimage(handles.axes2); %s=rgb2gray(s); imshow(F,'parent',handles.axes6) hold on viscircles(centers, radii,'EdgeColor','b'); %As an alternative, you can use the builtin imfindcircles functions as follows % calculate the centroid and radius of all the regions %stats = regionprops('table',grad >= 20,'Centroid', 'MajorAxisLength','MinorAxisLength'); %centers = stats.Centroid; %diameters = mean([stats.MajorAxisLength stats.MinorAxisLength],2); %radii = diameters/2; %[maxRadii, iMax] = max(radii); % select the largest circle %subplot(2, 2, 1); %viscircles(centers(iMax, :),maxRadii); % visualise the selected circle function edit1_Callback(hObject, eventdata, handles) % hObject handle to edit1 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB 54 Digital Eye Screening Support System 2019 % handles structure with handles and user data (see GUIDATA) % Hints: get(hObject,'String') returns contents of edit1 as text % str2double(get(hObject,'String')) returns contents of edit1 as a double % --- Executes during object creation, after setting all properties. function edit1_CreateFcn(hObject, eventdata, handles) % hObject handle to edit1 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles empty - handles not created until after all CreateFcns called % Hint: edit controls usually have a white background on Windows. % See ISPC and COMPUTER. if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor')) set(hObject,'BackgroundColor','white'); end % --- Executes on mouse press over figure background. function figure1_ButtonDownFcn(hObject, eventdata, handles) % hObject handle to figure1 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) function edit4_Callback(hObject, eventdata, handles) % hObject handle to edit4 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Hints: get(hObject,'String') returns contents of edit4 as text 55 Digital Eye Screening Support System 2019 % str2double(get(hObject,'String')) returns contents of edit4 as a double % --- Executes during object creation, after setting all properties. function edit4_CreateFcn(hObject, eventdata, handles) % hObject handle to edit4 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles empty - handles not created until after all CreateFcns called % Hint: edit controls usually have a white background on Windows. % See ISPC and COMPUTER. if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor')) set(hObject,'BackgroundColor','white'); end % --- Executes on selection change in listbox4. function listbox4_Callback(hObject, eventdata, handles) % hObject handle to listbox4 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Hints: contents = cellstr(get(hObject,'String')) returns listbox4 contents as cell array %contents{get(hObject,'Value')} returns selected item from listbox4 % --- Executes during object creation, after setting all properties. function listbox4_CreateFcn(hObject, eventdata, handles) % hObject handle to listbox4 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles empty - handles not created until after all CreateFcns called 56 Digital Eye Screening Support System 2019 % Hint: listbox controls usually have a white background on Windows. % See ISPC and COMPUTER. if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor')) set(hObject,'BackgroundColor','white'); end % --- Executes on button press in pushbutton9. function pushbutton10_Callback(hObject, eventdata, handles) % hObject handle to pushbutton10 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) 57 Digital Eye Screening Support System 2019 Appendix F Statement of approval from ophthalmology department Figure 25: statement of approval from JUSH ophthalmology ward 58