Tıbbi tedavi verilerinden sebep sonuç ilişkilerinin makina öğrenim yöntemleri ile belirlenmesi

The healthcare industry is equipped with a treasure trove of data gathered from various sources. This massive amount of data can be used to work wonders. In this study, we used a carefully selected subset of data from the Medical Information Mart for Intensive Care database. We adopted principal component analysis as the main method, to observe and capture the daily medical changes in intensive care unit patients from their medical treatment data. The results can be used to inform the attending physicians, which laboratory tests are exhibiting variances after an intervention along with their associated epiphenomenon. This will be used to make decisions such as, which treatment or diagnosis to apply further or which prescriptions to give/avoid. Experimental analysis results indicate that principal component analysis was able to capture daily patient changes after a certain treatment or intervention. Model validity and stability are evaluated using permutation and bootstrap tests. In both cases, the model exhibited an acceptable significance level. Moreover, causal impact analysis is conducted using Bayesian structural time-series models. Results showed that the approach provides promising results for interpreting large quantities of patient data for establishing a cause-effect relationship from medical interventions and be used as an early warning system. The study reflected the capability of principal component analysis to monitor and provide an alert to the clinicians about the patient's conditions, thereby providing opportunities for timely interventions. If combined with other AI models, the approach can be able to support medical decision making and enable effective patient-tailored care for better patient health outcomes.