Please use this identifier to cite or link to this item: https://gnanaganga.inflibnet.ac.in:8443/jspui/handle/123456789/14920
Title: Proposed Model for Detection of Pneumonia Using Deep Learning
Authors: Rai, Bipin Kumar
Srivastava, Anoop Kumar
Sharma, Shivani
Kamboj, Shashank
Keywords: Convolution Neural Network
Deep Learning
Flask Framework
Python
Seaborn
Issue Date: 2024
Publisher: Lecture Notes in Electrical Engineering
Springer Science and Business Media Deutschland GmbH
Citation: Vol. 1096; pp. 563-573
Abstract: Accurate detection of Pneumonia is highly challenging. Pneumonia is first diagnosed by a doctor through the x-ray, but it can be time taking and can have a lot of investments. We used a Deep Learning algorithm to solve this problem. This paper presents a proposed model called Totally Automated Pneumonia Discovery (TAPD) Model for detection of Pneumonia using Deep Learning. We developed an algorithm which utilizes Convolution Neural Network (CNN) and with the help of some other layers we made a custom Deep Learning Model. Deep Learning algorithms are widely used in analyzing medical images and CNN has become very useful for disease classification. We have used a dataset containing images and created an algorithm which can detect whether a person is suffering from Pneumonia or not? Unlike other methods, our model handles the problem of overfitting with ease and also eliminates the problem of vanishing and exploring gradients. Our model is less complex and has less complexity. Our model is developed on the Flask framework which is totally based on Python and also our model has provided a user interface to test. Accuracy of the model is great as compared to the other models. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.
URI: https://doi.org/10.1007/978-981-99-7137-4_56
http://gnanaganga.inflibnet.ac.in:8080/jspui/handle/123456789/14920
ISBN: 9.78982E+12
ISSN: 1876-1100
Appears in Collections:Conference Papers

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.