Artificial Intelligence in detection of acute coronary occlusion in NSTEMI patients

Comprehensive Summary

This study aims to assess the use of artificial intelligence in detecting acute coronary occlusions (ACO) in non ST-elevation myocardial infarction (NSTEMI) as previous methods have not been able to do so. The diagnosis of ACOs is highly reliant on pattern recognition in the EKG, Ai programs have been implemented in several previous studies to help detect ACOs through this. The authors analyzed the results of the studies providing an overview of the accuracy of AI use in this scenario. In a study completed by Zaiti and colleagues, an AI program created, was able to lower the amount of patients placed into the intermediate risk group by accurately classifying more patients into the low risk group. In addition, several other studies created deep neural network models which have helped enhance validation across diverse populations in detection of ACO. Overall, in the majority of the reviewed literature, the AI programs proved to be more accurate than the programs already in place. However, there still needs to be randomized data tested with the AI program in order to validate the AI’s effectiveness before implemented in daily practice.

Outcomes and Implications

The research completed by this study shows the potential implementation of the AI programs in detections of ACOs, which will hopefully allow for earlier detection and treatment. However, the authors mention the limitations of currently implementing these AI programs including problems with patient data protection and the need for basic AI training in healthcare workers. Additionally, the authors mention future research that should be complemented using different paradigms of ACO to detect it using AI.