9/1/23, 12:17 AM s.amizone.net/NTCC/NTCC/ProjectSynPrint?CourseId=812659 AMITY UNIVERSITY -----UTTAR PRADESH----Amity School of Engineering and Technology Minor Project Student Name Enrollment No Programme Company's Name and Address Industry Guide Name Designation Contact Number PAARTH BHARDWAJ A2305220552 B.Tech (Computer Science & Engineering) Amity University Uttar Pradesh Amity Rd, Sector 125, Noida, Uttar Pradesh 201301 110058 Dr. Geetika Associate Professor Ph.(O) : 9871882460 Mobile : 9871882460 Fax : 9871882460 E-mail : ait@amity.edu (R) : 9871882460 Project Information 1) Project Duration : (68 Days) a) Date of Summer Internship commencement (31/07/2023) a) Date of Summer Internship Completion (06/10/2023) 2) Topic SkinSage XAI : Intelligent Lesion Diagnosis 3) Project Objective The "SkinSage XAI" skin lesion detection system utilizes deep learning models and Explainable AI techniques to accurately classify skin lesions. Over ten weeks, the project team trains models, extracts features, and implements Explainable AI methods. The system aims to enhance early skin lesion detection for improved medical diagnosis and treatment. 4) Methodology to be adopted • Data Collection and Cleaning: Gather a diverse dataset of skin lesion images from reliable sources and perform data cleaning to ensure consistency and quality. • Deep Learning Model Training: Train multiple deep learning models, such as CNN, ResNet, DenseNet, and InceptionV3, using transfer learning and finetuning on the collected dataset. • Feature Extraction: Extract deep features from the trained models and apply Principal Component Analysis (PCA) to reduce dimensionality (optional). • Explainable AI Implementation: Utilize Grad-CAM to visualize relevant regions in images and LIME to generate local explanations for model predictions. • Interpretation and Documentation: Analyze the results, interpret model behavior, and document findings to provide valuable insights for dermatologists. 5) Brief Summery of project(to be duly certified by the industry guide) https://s.amizone.net/NTCC/NTCC/ProjectSynPrint?CourseId=812659 1/2 9/1/23, 12:17 AM s.amizone.net/NTCC/NTCC/ProjectSynPrint?CourseId=812659 The "SkinSage XAI" system is expected to achieve high accuracy in classifying skin lesions and provide interpretable explanations for its predictions. The use of Explainable AI methods will offer visualizations of relevant regions within images, empowering dermatologists to make informed decisions based on the model's output. Through the development of the "SkinSage XAI" skin lesion detection system, the project aims to make a significant contribution to the field of dermatology. By combining state-of-the-art deep learning models with Explainable AI techniques, the system seeks to improve the early detection of skin lesions, ultimately leading to more effective medical interventions and better patient outcomes. Signature (Student) Close Signature (Industry Guide) Signature (Faculty Guide) Print https://s.amizone.net/NTCC/NTCC/ProjectSynPrint?CourseId=812659 2/2