Software developed by academics from Hitit University's Faculty of Medicine and Faculty of Engineering detects cancerous cells using artificial intelligence.
Following Hitit University's specialization in machine and manufacturing technologies, the "Digital Transformation Laboratory" was established, initiating studies specifically focused on the use of artificial intelligence in various fields.
A team consisting of academics and students from Hitit University's Faculty of Medicine and Faculty of Engineering was formed to conduct research on the use of artificial intelligence in the diagnosis of various cancer diseases.
As part of the study, approximately 12,000 labeled samples taken from leukemia patients by the hematology and medical microbiology units were used to train artificial intelligence on 10 different cancer types.
Furthermore, a digital apparatus integrated into a microscope was developed, enabling the conversion of laboratory studies into digital data.
The first study focuses on blood cells.
Associate Professor Dr. Ünsal Savcı, Head of the Department of Medical Microbiology at Hitit University's Faculty of Medicine, stated that they trained artificial intelligence on thousands of blood cells, both those carrying and not carrying a cancer risk, using photographs taken with an advanced microscope.
Prosecutor Savcı stated that they achieved a 90% success rate in diagnosing with artificial intelligence through three different trials, saying, "Here, acute and chronic leukemia diagnoses have truly begun to be made. While the success rate in some cancer types in cells is approximately 100%, we have achieved an average success rate of 90% in distinguishing cancer from normal cells."
Savcı added that with increasing data, the success rate of artificial intelligence in diagnosis could approach 100% and that it could be used as a decision support mechanism, even if not a decision-making mechanism, especially in hospitals and health institutions where there are no specialists in the hematology branch.
Drawing attention to the fact that diagnoses can be missed in health institutions where there are no specialists in the relevant field, Savcı said, "Reaching this diagnosis is important in remote places, small towns, and small health institutions. Being able to make this diagnosis is important. Of course, who will make the final decision? A clinician, a doctor, will make this decision. Artificial intelligence is a decision support mechanism here. It will guide people."
The prosecutor added that they are also conducting studies on the use of artificial intelligence in the diagnosis of other diseases.
Professor Dr. Akif Akgül, Head of the Computer Engineering Department at Hitit University, explained that they are developing AI-based disease detection studies in blood cells with doctors from the medical faculty and continued as follows:
"We detected 10 different classes with 12,000 labeled samples on approximately 1200 data points. While some of these classes had a success rate close to 100%, we have generally achieved an average success rate of 90%. The development process continues with different algorithms, and we aim to exceed 90%."
Artificial Intelligence to Send Diagnosis to Doctors' Mobile Phones
Explaining that the study on disease detection in blood cells began with the idea of eliminating diagnostic problems in healthcare facilities where there are insufficient specialist physicians, Akgül stated:
"We prepared a project to solve these problems. We developed an apparatus for microscopes. By attaching this apparatus, cells can be visualized, and different classes can be identified from the images, determining which type of disease these classes represent. Furthermore, through an interface, this apparatus sends messages or emails to specialist physicians, providing information as a decision support system, and doctors can make the final diagnosis."
Akgül also pointed out that with the mobile application they are working on, doctors can follow the AI's diagnostic suggestions from their mobile phones.
Prof. Dr. Akgül emphasized that by including students in the project alongside academics, they contributed to students learning about the use of artificial intelligence and the project development process.
