AI-guided automated identification of patients at high-risk for hepatocellular carcinoma in large US and Global datasets.

[BACKGROUND & AIMS] Hepatocellular Carcinoma (HCC) screening targets patients with cirrhosis but lacks risk stratifi- cation and misses non-cirrhotic cases. We evaluated whether LIver cancer RIsk Computation (LIRIC) models built from routinely collected electronic health record (EHR) data can predict 3-year HCC risk in general and cirrhosis populations and generalize across diverse U.S. and international cohorts.

[METHODS] We conducted a multi-center study using longitudinal EHR data from 64 U.S. and 29 ex-U.S. (Latin America/Asia- Pacific) healthcare organizations. Adults >40 without prior HCC were included; cases were identified using ICD codes 6-36 months pre-diagnosis. U.S. models were developed in general and cirrhosis cohorts (46,679 cases; 1,128,202 con- trols), internally-externally validated by site, race/ethnicity, and time, and evaluated in a simulated "silent" deployment. We benchmarked model-derived risk against population incidence using standardized incidence ratios (SIRs). External validation applied U.S. models to ex-U.S. cohorts; region-specific ex-U.S. models were also trained. Logistic regression and neural networks (LIRICLR, LIRICNN) were compared using area under the curve (AUC), calibration, and Geometric Mean of Overestimation (GMOE) risk ratios.

[RESULTS] In the U.S. general population, LIRICNN achieved AUC 0.93 (95% CI: 0.9218-0.9289) using 46 features. Internal-external AUCs averaged 0.93 across sites and race/ethnicity strata, with good calibration (GMOE 0.89; 95% CI: 0.862-0.911). At a screening threshold corresponding to SIR≈31, sensitivity was 48% and specificity 97%. U.S.- trained models applied to ex-U.S. data yielded lower AUCs (0.84), but region-specific retraining restored performance (AUC 0.94).

[CONCLUSIONS] LIver cancer RIsk Computation (LIRIC) models accurately stratify Hepatocellular Carcinoma (HCC) risk using routine electronic health record (EHR) data in both general and cirrhosis populations and can be adapted for inter- national settings via local retraining. These results support LIRIC as a scalable foundation for risk-based HCC surveillance strategies and future prospective implementation studies.

[IMPACT AND IMPLICATIONS] Current HCC screening targets patients with cirrhosis but lacks risk stratification and misses non-cirrhotic cases; we therefore evaluated whether LIRIC models built from routinely collected EHR data can predict 3-year HCC risk in general and cirrhosis populations and generalize across diverse U.S. and international cohorts. These findings are important for hepatologists, primary care clinicians, and health systems because they show that routine EHR data can support accurate and scalable HCC risk stratification across diverse populations. In practice, LIRIC could serve as a foundation for risk- based HCC surveillance strategies and future prospective implementation studies using data already collected in routine care. Because this was a retrospective modeling study and international performance required local retraining, these findings should be interpreted as supporting future implementation rather than immediate universal adoption.