A major medical study has shown that artificial intelligence (AI) is more accurate than humans in analyzing long-term heart rhythm monitoring, which could help improve the diagnosis of heart disorders and reduce human errors.
اضافة اعلان
According to the study, published in Nature Medicine, the human heart beats up to 120,000 times a day, making the analysis of electrocardiogram (ECG) recordings over several days or weeks a challenging and time-consuming task.
Researchers reviewed data from 14,606 patients who used ECG devices for an average of 14 days.
Initially, the data was reviewed by human technicians using traditional methods, then re-analyzed using an AI algorithm called Deep Rhythm AI, developed by the Polish company Medical Logarithmics.
The results showed that human technicians failed to detect acute arrhythmia in 4.4% of patients, while this rate dropped to just 0.3% when using AI.
The model was also able to exclude non-critical cases with 99.9% confidence in the 14-day recordings.
In a comment on the study, Linda Johnson, the lead researcher from Lund University in Sweden, noted that the shortage of specialists in analyzing ECG during walking presents a major challenge for global healthcare.
She pointed out that extending the duration of ECG recordings could be beneficial for patients, but it requires additional resources.
She added: "We believe that AI could be the solution to this problem, contributing to improved healthcare and more accurate medical data analysis."