A collaborative effort between academics from Hitit University's medical and engineering faculties has resulted in the development of AI-powered software capable of diagnosing anterior cruciate ligament (ACL) tears.
Following Hitit University's specialization in machinery and manufacturing technologies, the Machinery and Manufacturing Technologies Application and Research Center (MİTAM) was established, initiating research focused on the use of artificial intelligence in various fields.
A team was formed within the university, comprised of academics from Hitit University's medical and engineering faculties, to conduct research on the use of artificial intelligence in diagnosing ACL tears.
Approximately 400 people's MRI images were used to train the AI.
As part of the study, the MRI images of approximately 400 individuals with ACL tears were trained on the AI by the orthopedics and traumatology unit.
Associate Professor Dr. Taner Alıç, a faculty member of the Department of Orthopedics and Traumatology at Hitit University's Faculty of Medicine, stated that the ACL is one of the most important parts of the knee and that injuries in this area are particularly common in athletes.
Emphasizing the importance of accurate diagnosis in the treatment of anterior cruciate ligament (ACL) tears, Alıç stated, "In the accurate diagnosis of ACL tears, MRI images are as crucial as clinical examination. However, in evaluating these MRI images, it is important to notice very important details that humans might inevitably miss and to make an accurate diagnosis and provide the correct treatment."
"Up to 92% accuracy rate"
Drawing attention to the fact that artificial intelligence and deep learning models are beginning to take their place in the healthcare sector, especially in orthopedics, Alıç continued:
"Most studies are evolving in this direction. This process is also the case for ACL tears. With artificial intelligence models, we can diagnose detailed tears that humans might miss accurately and very quickly, within seconds. We collaborated with valuable professors from the engineering faculty in making this diagnosis. We focused on how we could diagnose ACL tears more accurately and quickly, and we achieved quite good results in our studies. There is an accuracy rate of up to 92%. This was a very important figure for a start."
Alıç added that in the future they will conduct AI-assisted studies targeting different areas such as meniscus and cartilage damage, and that such software is a technological tool that will help and support physicians, especially in diagnosis.
Images from different MRI devices will also be processed.
Meryem Yalçınkaya, Assistant Professor in the Department of Industrial Engineering at the Faculty of Engineering, explained that they are conducting numerous studies on artificial intelligence technology at MITAM.
Yalçınkaya stated that during a consultation with the Chief Physician's Office of Hitit University Erol Olçok Training and Research Hospital approximately 1.5 years ago, they decided to conduct AI-assisted studies for the orthopedics and traumatology department. She added that after two months of technical procedures conducted with Assoc. Prof. Dr. Taner Alıç, they succeeded in teaching artificial intelligence about anterior cruciate ligament tears.
Yalçınkaya stated, "In the subsequent process, we focused on model development in the digital laboratory here. We developed a deep learning model to detect whether or not there is a tear in the cruciate ligament area. We developed multiple models, and the best one detected the presence or absence of a cruciate ligament tear with 92% accuracy."
Yalçınkaya also indicated that they will process images from different MRI devices in the next phase to further improve the software.
Foreseeing that artificial intelligence will be an integral part of early diagnosis and detection in the future and stating that they are conducting AI-supported studies in this direction, Yalçınkaya noted:
"We have conducted studies not only on cruciate ligament tears but also on the detection of previous cancer cells, whether or not there is a foreign body in the bronchoscopy area, and whether or not there is loss of emotion on the human face after cosmetic surgery. We aim to combine all of these into a single program, under different modules, and transform this product into a service product in the healthcare sector within the Hitit Technology company."
