NORMA eResearch @NCI Library

Multiclass classification of Covid-19, Tb, Pneumonia, and health cases using Deep Learning

Shinde, Aditya Pramod (2022) Multiclass classification of Covid-19, Tb, Pneumonia, and health cases using Deep Learning. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (645kB) | Preview


In recent times we have been made aware of the high infection rate and lethality that can be caused by the outbreak of pulmonary diseases like Covid-19. The need for an economical and quick diagnosis of such a disease is also increasing. It is important to consider that a disease like covid has a resemblance to some of the other pulmonary diseases like pneumonia and tuberculosis and has overlapping symptoms and characteristics with them. There have been many previous types of research that have used chest x-rays for the classification of covid and healthy cases or covid and pneumonia or covid and TB using binary classification. This research proposes a deep learning framework for multiclass classification of Covid-19, pneumonia, Tuberculosis, and normal cases with the use of chest X-rays. In this research, I have used a combination of three open datasets available on Kaggle to create my dataset for research which has a total of 4136 images. I have used image data augmentation and applied pre-trained models like VGG16 and DenseNet121. I have also created a CNN from scratch which can perform classification with 91 per cent overall accuracy and high precision and an F1 score which I have discussed further in the report. This investigation can help physicians and patients in early diagnosis and can act as a pre-diagnosis method to start early treatment or isolation of the patient. It can also act as a patient prioritization tool to help physicians focus on people that have a high probability to be infected in case of a high number of patients due to a sudden outbreak.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
R Medicine > RB Pathology
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 11 Mar 2023 12:32
Last Modified: 11 Mar 2023 12:32

Actions (login required)

View Item View Item