Implementatie-intelligentie met AI: prospectieve evaluaties nodig

The publication of the first phase 3 randomized clinical trial in 2018 demonstrating benefit of a transthyretin stabilizer for the treatment of amyloid transthyretin cardiomyopathy (ATTR-CM) fueled a growing interest in improving detection of ATTR-CM. Concurrently, the emergence of artificial intelligence (AI)–based diagnostics has led to the development of a range of AI-based screening models, which leverage various combinations of structured data and unstructured data, such as electrocardiogram (ECG) waveforms and echocardiography images. These models have shown the potential to facilitate diagnosis of ATTR-CM, and the US Food and Drug Administration has cleared several AI-echo models for cardiac amyloidosis detection as Software as a Medical Device. However, prospective evaluation and clinical experience with these models remains limited and is critically needed prior to broad deployment.