GUI based Thoracic Disease Detection using Segmented Deep Convolutional Neural Network

Deep learning has shown exciting results in the field of medical imaging. Thoracic problems can be serious, and if not detected and treated in a timely manner, they can result in death. Chest X-ray imaging is being used to diagnose numerous thoracic disorders. Generally, we required a knowledgeable...

Full description

Saved in:
Bibliographic Details
Main Authors: Pant, Himanshu, Lohani, Manoj Chandra, Bhatt, Ashutosh Kumar, Pant, Janmejay, Singh, Devendra
Format: Conference Proceeding
Language:English
Subjects:
Chest X-ray
Classification
Classification algorithms
Computer architecture
Convolutional Neural Network
Convolutional neural networks
Deep learning
Disease Screening System
Segmentation
Thoracic Disorders
Thorax
Training
X-ray imaging
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Deep learning has shown exciting results in the field of medical imaging. Thoracic problems can be serious, and if not detected and treated in a timely manner, they can result in death. Chest X-ray imaging is being used to diagnose numerous thoracic disorders. Generally, we required a knowledgeable and experienced radiologist to examine chest X-ray images and determine whether or not a person has certain thorax abnormalities. Everyone doesn't have the opportunity to consult an expert in a timely manner to treat chest disease because in certain circumstances, speedy diagnostics are required. As a consequence, we present an image classification and segmentation-based algorithm that would predict a specific thoracic condition and produce a mask, empowering doctors to make critical decisions. Deep Learning approaches have proven their worth in a range of contexts, outperforming many state-of-the-art models. We used the U-Net framework with ResNet as the backbone and got good results. In medical image processing and semantic segmentation, U-Net outperforms.
ISSN:2771-1358
DOI:10.1109/ICCUBEA54992.2022.10011072