Computational Medicine Research Group specializes in artificial intelligence, machine learning, application of mathematical and computational models to routine clinical medicine.
Acceptable laboratory test intervals to be learned from hospital data using artificial intelligence algorithms, to provide a personalized reference interval with learned intervals.
The visual and audio recordings of the interaction of the physician with the patient during the clinical examination are understood by artificial intelligence algorithms and the data captured is stored as part of patient's electronic health records.
The machine-learning model involves evaluating and learning the effects of a treatment applied to a patient. Using electronic health records of patients, the model learns the cause effect relationships between the treatment and the results of the treatment and tries to bring these cause-effect relations to the attention of the physician using machine-learning algorithms.
In this project, a system is implemented that can execute and coordinate staged ordering and execution of diagnostic laboratory tests.
A system that allows narration to be taken from the patient using artificial intelligence is being implemented. Patient story combined with patient's electronic healt records is used to generate several hypotheses relevant to patient's clinical condition.
An environment for determination of reference intervals for laboratory tests for pediatric patients, using data mining techniques on electronic health records.
Application of probabilistic learning techniques in health problems.
Identification and monitoring of internal quality control nonconformities in medical laboratories.
In this study, two different databases were developed for the organization of variation-based personal genetic data. The first from these databases is the relational database, and the second is the no-sql database. In both databases, the variation data of 2504 individuals, which were published by 1000 Genomes Projects, were stored. To store this data, the spaces needed by the databases were calculated and compared. In addition, some queries that are frequently used by clinical applications were run and the response times of the databases were calculated. In this study, three new methods for three different clinical applications were also developed and the integration of databases with these methods was provided. The first method classifies individuals as disease-based, finds individuals who are genetically most similar to a person and calculates the disease risks of the individuals. The second method dynamically detects variations that may be associated with any disease or treat. The last method identifies protected regions using variation-based personal genetic data.
In the field of laboratory medicine, minimizing errors and establishing standardization is only possible by predefined processes. The aim of this study was to build an experimental decision algorithm model open to improvement that would efficiently and rapidly evaluate the results of biochemical tests with critical values by evaluating multiple factors concurrently.
Preparing visual designs by artificial intelligence.
Automatic confirmation of the results of medical laboratory tests using artificial intelligence.
F Demirci, P Akan, T Kume, AR Sisman, Z Erbayraktar, S Sevinc American journal of clinical pathology 146 (2), 227-237. Go to the Article
Çakırgöz, O., & Sevinç, S. (2015). Proceedings of the World Congress on Electrical Engineering and Computer Systems and Science (EECSS 2015), 314, 1–8.Go to the Article
Çakırgöz, O., & Sevinç, S. (2016). Proceedings of the International Conference on Computer Science and Engineering (UBMK 2016), 16, 89–99.
Çakırgöz, O., & Sevinç, S. (2016). Proceedings of the International Conference on Computer Science and Engineering (UBMK 2016), 24, 143-148.
Çakırgöz, O., & Sevinç, S. (2016). Proceedings of the International Conference on Computer Science and Engineering (UBMK 2016), 37, 214-222.
Çakırgöz, O., & Sevinç, S. (2016). International Conference on Advanced Technology & Sciences ICAT'Rome, 23–31.
Projects as Computational Medicine Group
As a result of our studies, we obtained 3 patent applications, academic publications, industrial projects, research projects and we continue to work.
The Symposium on "New Approaches Based on Artificial Intelligence in Tıpta" was held in Health Sciences University Tepecik Training and Research Hospital. Researchers interested in "Artificial Intelligence", scientists, health workers and software ...Read More
Mithatpaşa Street No: 56/20 Dokuz Eylül University Technology Olive Building Office No: 112 Balçova-İzmir
(0232) 277 55 59