Artificial Intelligence Predicts Inherited Cardiac Arrhythmia Using ECG, According to University of British Columbia Research
New research conducted by the University of British Columbia reveals that artificial intelligence (AI) combined with electrocardiography (ECG) data can accurately diagnose long QT syndrome, an inherited heart disorder. Long QT syndrome affects roughly one in 7000 people, with 80% of cases linked to genetic mutations in around 15 genes. Standard ECG testing measures the heart’s electrical signals to detect irregular heartbeats.
The study, featured in JAMA Cardiology, utilized a deep-learning neural network to effectively diagnose long QT syndrome and distinguish between the two most common genotypes associated with the condition. Patients with long QT syndrome experience occasional irregular heartbeats triggered by factors like exercise, stress, or exposure to cold water. Early diagnosis is crucial for managing the condition since sudden arrhythmias can be fatal if left untreated.
Individuals with long QT syndrome typically exhibit longer than normal QT intervals on ECG tests, although this is not always apparent. While genetic testing can supplement an ECG-based diagnosis, it may not be readily available or capable of detecting rarer gene mutations. Consequently, a more accurate and quick method for diagnosing long QT syndrome using ECG data can significantly benefit healthcare professionals.
Artificial intelligence has seen increasing use in medical diagnostics, including the interpretation of ECG data. In this study, Andrew Krahn, a professor at the University of British Columbia, and his team developed a deep learning neural network to identify signs of long QT syndrome in ECG readouts. The network was trained using 4521 ECGs from 990 individuals enrolled in the Hearts in Rhythm Organization Registry. A separate validation group of 101 patients was used to test the network’s accuracy in diagnosing long QT syndrome and differentiating between genotypes.
The neural network demonstrated impressive accuracy in diagnosing long QT syndrome and distinguishing between genotypes. The researchers highlight the network’s ability to identify the condition even in cases with normal or borderline QTc intervals. This feature is crucial for effective screening, particularly for patients who may require further testing or are at risk of QT-mediated arrhythmias when exposed to QT-prolonging drugs.
Further validation over a larger population may enable the widespread use of this model for assessing torsade de pointes risk in patients suspected of having long QT syndrome. The integration of AI and ECG data provides a potential solution for more efficient and accurate diagnoses, benefiting both clinicians and patients alike.
Analyst comment
Overall, this news can be evaluated as positive. The use of artificial intelligence combined with electrocardiography (ECG) data has shown promising results in accurately diagnosing long QT syndrome, an inherited heart disorder. This method can significantly benefit healthcare professionals by providing a more accurate and quick diagnosis, improving patient management. The integration of AI and ECG data provides a potential solution for more efficient and accurate diagnoses, benefiting both clinicians and patients alike. The market for AI in medical diagnostics is expected to grow as more successful applications are developed.
