AI-Echo verslaat AI-ECG en klinische score voor diagnose cardiale amyloïdose (AUROC 0,93)
Retrospectieve cohortstudie bij 598 patiënten verwezen voor cardiale scintigrafie voor verdenking transthyretine-cardiale amyloïdose (ATTR-CA), waarin drie risico-modellen werden vergeleken: de ATTR-CM klinische score, AI-ECG en AI-Echo.
Mediane leeftijd 76 jaar; 30% had bewezen ATTR-CA. AI-Echo presteerde superieur: AUROC 0,93 (95%-BI 0,91-0,95) versus 0,79 voor AI-ECG en 0,87 voor ATTR-CM score (p<0,001). Sensitiviteit/specificiteit waren respectievelijk 86%/85% (AI-Echo), 80%/64% (AI-ECG) en 86%/69% (klinische score).
Bij een drempel van 0,25 vermeed AI-Echo de meeste onnodige scintigrafieën (45 per 100 verwezen patiënten, tegenover 24 voor AI-ECG en 37 voor ATTR-CM). AI-Echo lijkt het meest waardevolle screeningsinstrument voor ATTR-CA in een hoog-risicopopulatie en kan onnodige nucleaire beeldvorming voorkomen.
Abstract (original)
BACKGROUND: To improve screening for cardiac amyloidosis (CA), several models using artificial intelligence (AI) and conventional statistics have been developed. However, few data are available to compare the relative utility of these tools. In this study, models were compared to determine their potential roles in optimizing diagnostic algorithms. METHODS: In this retrospective cohort study at our tertiary medical center, patients referred for cardiac scintigraphy for detection of transthyretin CA (ATTR-CA) who had ECG and transthoracic echocardiography within 6 months along with clinical characteristics for risk score calculation were included. The performance of previously developed and validated clinical and AI risk models for ATTR-CA, including the transthyretin ATTR-CM clinical score and AI models applied to electrocardiography (AI-ECG) and echocardiography (AI-Echo) were compared in a population referred for cardiac scintigraphy. Previously defined thresholds were used for each model. As the ATTR-CM score was validated following exclusion of AL amyloidosis, 28 patients with AL amyloidosis were excluded. AL and ATTR-CA were defined per guideline criteria. RESULTS: Among 598 patients (median age 76 [67-82] years; 72.6% male), 181 (30%) had ATTR-CA. AI-Echo identified ATTR-CA with 86% sensitivity and 85% specificity, compared to 80% and 64% for AI-ECG, and 86% and 69% for the ATTR-CM score. AUROC was 0.93 (95% CI 0.91-0.95) for AI-Echo, 0.79 (0.76-0.83) for AI-ECG, and 0.87 (0.84-0.90) for ATTR-CM score, p < 0.001). In this cohort, use of AI-Echo could have avoided more unnecessary scintigraphy than AI-ECG or ATTR-CM score (45 vs 24 vs 37, per 100, respectively) at a threshold probability of 0.25 (one case of ATTR-CA per 4 referrals for scintigraphy). CONCLUSION: Within a high-risk population for cardiac amyloidosis, the AI-Echo model demonstrated superior diagnostic discrimination and clinical utility for identification of ATTR-CA compared with the AI-ECG model and the ATTR-CM clinical score.
Dit artikel is een samenvatting van een publicatie in Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography. Voor het volledige artikel, alle details en referenties verwijzen wij u naar de oorspronkelijke bron.
Lees het volledige artikelDOI: 10.1016/j.echo.2026.05.007
Lid worden van HartVaat.nl?
Gratis — en we stemmen het nieuws en de literatuur af op uw vakgebied.
