Publication Date: 2021/06/24
Abstract: Human lungs are more prone to the diseases and need to be identified at the early stages so that appropriate treatment can be provided and minimize the complications. The utility of Artificial Intelligence (AI) in predicting the pathologies in chest X-ray images was studied. Two different neural network models were applied on the National Institute of Health (NIH) chest X-ray image datasets. These models were Visual Geometry Group (VGG) model and Sequential model with dense layers. Images were converted in to arrays and are fed into each of these models which predicted the pathologies present in the chest X-ray images. The accuracy obtained using VGG model was 65.6% and the accuracy of sequential model was 73%. Although sequential model is more accurate than VGG model, there is a need for further fine tuning the models for better accuracy
Keywords: VGG Model, Sequential Model, Artificial Intelligence, Neural Network, Lung Pathologies.
DOI: No DOI Available
PDF: https://ijirst.demo4.arinfotech.co/assets/upload/files/IJISRT21JUN605.pdf
REFERENCES