Abstract

Deep neural networks (DNNs) are a kind of algorithm that has been used to make significant improvements in dermatology in recent years. Deep neural networks are a type of algorithm that is used to make significant advances in dermatology (DNN). 10-13 In computing, neural networks are tiny computer programmes that take in input data and analyse it in order to provide predictions. Examples include making a diagnosis of a skin condition based on a clinical picture or identifying tumour areas on a pathology slide. NNs are taught for a certain task by seeing instances with known outcomes and learning from them. The NNs then make predictions based on these instances, and the performance of these predictions is assessed. The performance of the learning is used to update internal parameters known as weights, and new predictions are produced as a result of this updating. In this manner, the network’s predictive capacity is refined until it is judged acceptable, at which time it may be considered for wider use. In comparison to conventional training of DNNs, transfer learning represents a significant improvement. As a result of transfer learning, the quantity of training data needed drops significantly. This is especially relevant in medicine, where instances with known outcomes may be difficult to come by. Unsurprisingly, many of the dermatological deep learning research that have been published to far, including some of those addressed in this review, have utilised transfer learning to train their deep neural networks (DNNs). This book focuses on in-depth Dermatology and further Deep learning in it. This book mainly intended for the budding and novice researchers to know the rudiments on this particular domain.

Keywords:

Deep Learning, Dermatology

Download Sample Pages

How to cite this Book:

APA:
Gopalakrishnan, S., & Ebenezer Abishek, B (2021). Deep Learning in Dermatology (1st ed., pp. 1-178), Jupiter Publications consortium,ISBN: 978-93-91303-00-6, DOI: https://doi.org/10.47715/JPC.B.67.2021.978939130300610.47715/JPC.B.64.2021.9788194706953