A NEW tool may be reshaping early diagnosis, as an artificial intelligence programme significantly improved detection of transthyretin cardiac amyloidosis across a large health system. The trial found that targeted AI screening led to substantially more confirmed cases and timely treatment.
Expanding Efforts to Recognise Transthyretin Cardiac Amyloidosis Earlier
Efforts to improve recognition of transthyretin cardiac amyloidosis have intensified as treatment options expand and awareness of the condition grows. Yet ATTR-CM remains underdiagnosed, often presenting with subtle cardiac and orthopaedic signs that delay specialist referral. Identifying cases earlier could transform outcomes, and AI-supported programmes are now being examined as a possible route to reducing missed or late diagnoses. The need for improved detection is particularly important as the prevalence of transthyretin cardiac amyloidosis rises with ageing populations.
AI Screening Trial Shows Higher Diagnostic Yield Than Usual Care
The nonrandomized clinical trial evaluated ATTRACTnet, an AI model integrating electrocardiogram waveforms, echocardiographic measures, demographics, and diagnosis codes for orthopaedic manifestations of amyloidosis. The model achieved strong discrimination for transthyretin cardiac amyloidosis, with an area under the receiver operator characteristic curve of 0.85 using 5-fold cross-validation in an internal test set of 799 patients and 0.82 (95% CI, .81-0.83) in an external test set of 422 patients. During the study, 1471 patients were identified as having positive AI scores of 0.5 or higher, with 256 meeting criteria for inclusion. Of these, 50 underwent further diagnostic testing, leading to 24 confirmed diagnoses of ATTR-CM, of whom 21 (88 percent) began treatment within three months. The positivity rate was more than 2.8 times higher than historical controls at 15.3 percent (95 percent CI, 13.1 percent to 17.9 percent; P < .001) and resulted in an 18 percent relative increase in new diagnoses compared with the previous year.
Clinical Implications and the Future Of AI-Driven Detection
For clinicians, the findings highlight the potential for AI-assisted screening to close diagnostic gaps for transthyretin cardiac amyloidosis. Integrating such models into routine care may support earlier diagnosis, faster initiation of disease-modifying therapy, and improved patient pathways. However, prospective randomised trials are needed to establish whether enhanced detection translates into improved long-term outcomes. Continued refinement of AI tools, combined with clinician oversight, is likely to shape future cardiovascular diagnostic strategies.
Reference
Jain SS et al. Detecting transthyretin cardiac amyloidosis with artificial intelligence: a nonrandomized clinical trial. JAMA Cardiol. 2025;DOI:10.1001/jamacardio.2025.4591.
