Politecnico di Torino (logo)

Mobile application which makes diagnosis of lung diseases by detecting anomalies from X-Ray images

Atabay Heydarli

Mobile application which makes diagnosis of lung diseases by detecting anomalies from X-Ray images.

Rel. Giovanni Malnati. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Informatica (Computer Engineering), 2022

PDF (Tesi_di_laurea) - Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB) | Preview

The COVID-19 pandemic continues to have a enormous influence on the health and well-being of the world's population. An important step in the fight against COVID-19 is using the successful screening methods of infected patients, where one of the key approaches for screening is the chest X-ray radiological imaging. This study was designed create mobile application which uses machine learning techniques to automatically detect COVID-19 pneumonia patients checking their chest X-rays with maximum detection accuracy using Deep Neural Networks (DCNN). According to the studies analysed in second chapter of the thesis, COVID-19 detection accuracy utilizing CNN on chest X-rays is quite high and accurate. Despite the fact that there have been numerous studies on COVID-19 detection, no research has produced a quick and cellular-based COVID-19 detection system that uses CNN. The Single Shot Identification (SSD) MobileNet object detection model was employed in this research to generate a quick yet accurate detection of COVID-19. For solving this issue, we provide a mobile application which uses MobileNet neural network technology to make a diagnosis from chest X-Ray images. The model was designed by using MobileNet V2 which were trained with two datasets combined together for this thesis work. In this study, the MobileNet V2 learning model was proposed to identify coronavirus-infected pneumonia patients through chest X-rays and provides classification accuracy of more than 92% (training accuracy of 95 % and 91% accuracy of validation).

Relators: Giovanni Malnati
Academic year: 2022/23
Publication type: Electronic
Number of Pages: 84
Corso di laurea: Corso di laurea magistrale in Ingegneria Informatica (Computer Engineering)
Classe di laurea: New organization > Master science > LM-32 - COMPUTER SYSTEMS ENGINEERING
Ente in cotutela: Warsaw University of Technology (POLONIA)
Aziende collaboratrici: UNSPECIFIED
URI: http://webthesis.biblio.polito.it/id/eprint/24694
Modify record (reserved for operators) Modify record (reserved for operators)