Artificial intelligence has made it possible to independently identify cardiotoxic… » Medvestnik


Artificial intelligence-based software has made it possible to detect cardiomyopathy in women receiving anthracyclines for breast cancer at an early stage using ECG monitoring. Using this technique will allow patients to independently monitor their heart condition.

Researchers from the Mayo Clinic and Leuven University Hospital in Belgium assessed the effectiveness of artificial intelligence (AI) software in detecting abnormalities in left ventricular ejection fraction after anthracycline chemotherapy. The study results were published in the European Journal of Preventive Cardiology.




The study included 989 women without cardiovascular disease who received anthracycline chemotherapy for breast cancer without metastasis. In all participants, the left ventricular ejection fraction remained normal before the start of chemotherapy with anthracycline drugs.

After excluding patients without ECG recording within 90 days of transthoracic echocardiography, data from 20 women in the chemotherapy group and 683 participants in the control group were analyzed. AI-based software used ECG data to detect a decrease in left ventricular ejection fraction below 50% and below 35% after anthracycline therapy with AUC values ​​of 0.93 and 0.94, respectively.

All patients underwent ECG monitoring and transthoracic echocardiography before and after the course of treatment. The diagnostic effectiveness of the developed software was assessed based on ECG registration based on the area under the AUC curve with a left ventricular ejection fraction of less than 50%. During 9.83 years of observation, a decrease in left ventricular ejection fraction was detected in 22 women.

The authors believe that the new method provides a reliable, easy-to-use, and cost-effective alternative to cardiac imaging. It will allow patients to independently control their condition for a long time.



Source link