Ischemic heart disease and respiratory infections are listed among the leading causes of death globally. Hence, immediate diagnosis is essential to prevent further complications and initiate the treatment of patients as early as possible. However, disease symptoms may vary among patient population causing a major challenge in the diagnosis especially for the early stages, where they are not observable to be analysed as a basis of a disease. This delays the diagnosis, limits the medical resources, and extends the costs for the treatment.
The most severe manifestation of ischemic heart disease is myocardial infarction, and its early detection is crucial for saving lives of patients. The earliest signs of myocardial infarction are visible in cardiac ultrasound, which is echocardiography. In her doctoral dissertation, Aysen Degerli Ahishali proposes novel techniques to detect the signs of myocardial infarction using echocardiography with the support of Artificial Intelligence algorithms to provide computer-aided diagnosis in clinical practise.
In addition to decision support, the proposed algorithms provide visual enhancements over echocardiography recordings as capturing the movement of heart, which serve as assistive tools for cardiologists during the assessment and diagnosis of myocardial infarction.
COVID-19, listed in respiratory infections, is a major health concern that can lead to hospitalization, intubation, and even death. In fact, the World Health Organization reported around 7 million confirmed deaths worldwide due to COVID-19 by 2024. Especially during the pandemic, it became vital to detect COVID-19 immediately to prevent the spread of coronavirus for the sake of public health. However, time-consuming and inaccurate manual diagnostic tools prevent the spread of the disease from being controlled.
In her doctoral dissertation, Aysen Degerli Ahishali addresses these challenges by proposing automatic detection algorithms over chest X-rays taken from COVID-19 patients. The proposed AI-based algorithms provide computer-aided diagnosis to medical doctors by automatically annotating the pneumonia regions on patients’ lungs as well as detecting the early signs of COVID-19 that are challenging to catch by naked eye.
The integration of computer-aided diagnostic tools and algorithms into clinical practice is gaining increasing importance in medical diagnostics as AI continues to develop.
“The algorithms proposed in the dissertation for myocardial infarction detection have received an honourable mention for Technical Creativity Award 2022 by the City of Tampere. This shows that such advancements in computer-aided diagnosis become increasingly important in the future to build smart cities and enhance healthcare services,” says Aysen Degerli Ahishali.
Public defence on Friday 24 May
The doctoral dissertation of M.Sc. Aysen Degerli Ahishali in the field of Computing and Electrical Engineering titled Machine Learning Algorithms for Computer-aided Diagnosis of Myocardial Infarction from 2D Echocardiography and COVID-19 from Chest X-rays will be publicly examined at the Faculty of Information Technology and Communication Sciences at Tampere University at 12:00 on Friday 24 May 2024 at Hervanta Campus in the auditorium SA203 S3 of Sähkötalo building (Korkeakoulunkatu 3, Tampere).
The Opponent will be Prof. Tuan D. Pham from the Queen Mary University of London. The Custos will be Prof. Moncef Gabbouj from the Faculty of Information Technology and Communication Sciences at Tampere University.