INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
Diagnosis of Pneumonia on human Chest through X-ray using CNN algorithm of Deep Learning
Authors Name:
Avinash Bharatbhai Korat
, Prof. Prakash Patel
Unique Id:
IJSDR2102037
Published In:
Volume 6 Issue 2, February-2021
Abstract:
There are lots of advantages of using deep learning in the era of increasing medical data. It is easy to know about the relationships between data algorithmically without doing any laborious work. It is difficult to diagnose lung disease without a thorax x-ray. There is a shortage of radiologists in third world countries to diagnose CT images. Our approach to removing this kind of shortage and the use of technology can close the gap of the radiologist. Our proposed model can detect pneumonia from the thorax x-ray images with precise manner and accuracy. A transfer learning approach in which a pre-trained model is used and have to modify only a few layers of the existing model to diagnose pneumonia. In our proposed work, use of RSNA pneumonia detection dataset which contains enough images to train a deep learning model. Our model provides better prediction of pneumonia on the basis of lung infection in the human body using a deep learning approach.
Keywords:
Convolution Neural Network, Deep Learning, Medical Image Analysis
Cite Article:
"Diagnosis of Pneumonia on human Chest through X-ray using CNN algorithm of Deep Learning", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.6, Issue 2, page no.257 - 262, February-2021, Available :http://www.ijsdr.org/papers/IJSDR2102037.pdf
Downloads:
000337067
Publication Details:
Published Paper ID: IJSDR2102037
Registration ID:192926
Published In: Volume 6 Issue 2, February-2021
DOI (Digital Object Identifier):
Page No: 257 - 262
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn