Machine learning is a branch of artificial intelligence that aims to adapt computer algorithms to learning. The ability to solve problems without a predetermined algorithm is formed during the processing of a training dataset, which in
medicine includes the response of the patient’s body or a medical decision made in the context of a specific clinical situation. There are a number of machine learning methods, including classical methods, ensemble methods, and neural networks; depending on the method of training, there are training with a teacher, without a teacher, with partial involvement of a teacher, and training with reinforcement. The article describes the principles of operation, areas of application, advantages, and limitations of these methods in solving clinical problems encountered in ophthalmological practice.
The problems encountering at the stages of data collection, development, implementation, and further use of medical artificial intelligence systems are discussed, as well as possible ways to solve them.
Key words: Artificial intelligence, eye diseases, neural network, teacher training, decision support systems, medical visualization
Conflicts of interest. The authors have no conflicts of interest to declare.
Funding. This work was financially supported by the Foundation for Assistance to Small Innovative Enterprises in Science and Technology (contract №150ГС1ЦТНТИС5/64226 dated December 22, 2020).
For citation: Gliznitsa P.V., Takhchidi Kh.P., Svetozarskiy S.N., Bursov A.I., Shusterzon K.A. Machine
learning in the diagnosis and treatment of ophthalmic diseases. Head and neck. Russian Journal.
2022;10(1):83–90 (In Russian).
The authors are responsible for the originality of the data presented and the possibility of publishing illustrative material – tables, figures, photographs of patients.