Uploaded by jjstarc23

Unanswered Questions

advertisement
1. Unanswered Questions
One of the biggest obstacles in the effective treatment of ANCA-associated vasculitis is a
poor understanding of its underlying causes. Hence, a viable next step to the effective
treatment of this condition in computational modelling. Deep learning algorithms are
already being used to predict diabetes complications well in advance using patient data
(Dagliati et al., 2018). With respect to ANCA-associated vasculitis, machine learning could
be used to analyze key risk factors (such as age, environmental conditions, genetic
information) and provide medical practitioners important data on their patients’ risk of
undergoing ANCA-associated vasculitis in the future. This would allow them to begin the
necessary treatments before harmful symptoms set in, saving the patient much suffering. A
challenge to the implementation of such algorithms is the fact that there is a great deal of
randomness associated with the onset of ANCA-associated vasculitis due to the genetic
component of the illness (Chung and Monach, 2017). Hence, if a deep-learning based
approach is to be instituted, a major challenge will be to integrate a degree of stochasticity
to better simulate the randomness of genetics, which will be a most curious, and elegant,
integration of biology and computer science.
Download