Artificial Intelligence methods for decision making support

Lala Yagubova, BSc

Master's thesis

Abstract:

Pneumonia is the inflammation of the fluid inside the air sacs of the lungs. Pneumonia, which is caused by viral, bacterial, and sometimes fungal diseases that enter the lungs, is extremely infectious. Pneumonia is one of the top causes of mortality among children ages 0 to 4 years old. Pneumonia is one of the leading causes of death in the COVID-19 pandemic that has devastated the globe over the last several years. In situations with a huge number of cases, algorithms based on artificial intelligence may be employed to provide a more precise and timely diagnosis. By analyzing chest X-ray images using deep learning algorithms, it is possible to quickly and accurately identify hundreds of patients. As the primary focus of our work, we analyzed chest x-ray images and achieved a 97 percent accuracy rate for correct diagnosis. Since the majority of algorithms now in use take the form of black boxes, it is exceedingly difficult to understand how decisions are made. Also, we endeavored to determine which portions of the globe the AI programmer used to make judgments.

Keywords

COVID-19, X-ray, Pneumonia, Convolutional Neural Networks, Grad-Cam

Language used: English

Date on which the thesis was submitted / produced: 3. 12. 2022

Thesis defence

• Date of defence: 6. 6. 2024
• Supervisor: doc. Ing. Arnošt Veselý, CSc.
• Reader: Richard Vágner, externi

Czech University of Life Sciences Prague

Faculty of Economics and Management

Master programme / field:
Systems Engineering and Informatics / Informatics