Pragmatic Study of CNN Model and Different Parameters Impact on It for the Classification of Diabetic Retinopathy

Autor:	Manaswini Jena, Smita Prava Mishra, Debahuti Mishra
Rok vydání:	2020
Předmět:	Blindness Computer science business.industry Deep learning Pooling Activation function Pattern recognition Diabetic retinopathy medicine.disease Convolutional neural network medicine Artificial intelligence business Parametric statistics
Zdroj:	Smart Innovation, Systems and Technologies ISBN: 9789811559709
DOI:	10.1007/978-981-15-5971-6_74
Popis:	Diabetic retinopathy, a cause of permanent blindness in diabetic patients is required to be detected in right time to avoid the fatal consequences. Deep learning through convolution neural network is used in this work for the automatic detection of diabetic retinopathy. The convolution neural network model basically depends on its layers of architecture and the parameters like pooling layer, activation function and learning rate which has a huge impact on the model’s behaviour. Here, the study is about the response of model according to those parametric values and their types to achieve a better performative model towards the detection of diabetic retinopathy.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::e127cc189576f8b77b0a59179f319f65 https://doi.org/10.1007/978-981-15-5971-6_74 Zobrazit plný text záznamu