A Contemporary Technique for Lung Disease Prediction using Deep Learning

IEEE BASE PAPER ABSTRACT:

The field of computer science known as machine learning explores algorithms that learn from examples. Classification requires machine learning algorithms that understand how to apply a class label to data from the problem area. An easy-to-understand example is categorizing emails as “spam” or “not spam.” Binary classification predicts one of two classes, whereas multi-class classification predicts one of several classes. In most binary classification tasks, there is one type of normal state and another type of aberrant state. This Project involves how lung disease prediction using x-ray images will predict through the binary classification model implemented, and various python libraries like Tensor Flow, Keras, NumPy, etc. are used. This research project will observe the prediction of lung diseases by using x-ray images and further the output will be predicted with the detailed example and a detailed source code. The implementation will be shown step by step with possible screenshots.

OUR PROPOSED ABSTRACT:

The impact of disease on health is escalating quickly as a result of environmental changes, climate change, adjustments in lifestyle, and other reasons. Health problems are now more likely as a result. In 2016, 3.4 million individuals died from chronic obstructive pulmonary disease (COPD), which is typically brought on by smoking and pollution, while asthma claimed the lives of 400,000 people. Particularly in developing and low-income nations, where millions of people struggle with poverty and air pollution, there is a very high risk of lung disorders.

According to WHO estimates, illnesses linked to home air pollution, such as asthma and pneumonia, cause nearly 4 million preventable deaths each year. Worldwide, lung illness is a prevalent occurrence. These include pneumonia, asthma, TB, fibrosis, chronic obstructive pulmonary disease, and others. The early detection of lung illness is important.

The method of discovering and classifying lung disorders into different groups using medical imaging has been improved because to the development of deep learning. For this, a variety of machine learning and image processing models have been created. Convolutional neural networks (CNNs), one type of existing deep learning approach, are used to forecast lung illness. CNN’s basic version is inadequate. As a result, we propose a novel deep learning framework for predicting lung diseases based on the VGG16 Architecture.

The goal of this research is to develop a VGG16 architecture-based lung disease detection model. Early detection and diagnosis of lung disease are essential in the medical field because doing so will make it easier to manage patients’ future clinical care. The X-ray picture dataset obtained from the Kaggle source is subjected to the VGG16 Architecture. The dataset’s sample and full versions are taken into consideration.

The VGG16 Architecture beats existing techniques for both whole and sample datasets in terms of measures including precision, recall, F1 score, and validation accuracy. Therefore, the proposed VGG16 Architecture will make it easier for both professionals and clinicians to detect lung problems. This improvement has greatly helped the medical community’s ability to treat patients quickly.

PROJECT OTUPUT VIDEO:

ALGORITHM / MODEL USED:

VGG16 Architecture.

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

System : Pentium i3 Processor.
Hard Disk : 500 GB.
Monitor : 15’’ LED
Input Devices : Keyboard, Mouse
Ram : 4 GB

SOFTWARE REQUIREMENTS:

Operating System : Windows 10 / 11.
Coding Language : Python 3.8.
Web Framework : Flask.
Frontend : HTML, CSS, JavaScript.

REFERENCE:

V Pradyotan Raju, Senduru Srinivasulu, P Vishnu, Jeberson Retna Raj, Gowri S, Jabez J, “A Contemporary Technique for Lung Disease Prediction using Deep Learning”, 2022 International Conference on Applied Artificial Intelligence and Computing (ICAAIC), IEEE Conference, 2022.

Python IEEE Projects

A Contemporary Technique for Lung Disease Prediction using Deep Learning