Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
The goal of this clinical trial is to learn whether an artificial intelligence (AI) tool called FibroX can help primary care providers better diagnose significant liver fibrosis (≥F2) and clinically significant portal hypertension in adults with metabolic dysfunction-associated steatotic liver disease (MASLD).
The main questions it aims to answer are:
Researchers will compare FibroX-assisted care to usual care to see if FibroX improves diagnostic accuracy, provider trust, and supports better decision-making.
Participants will:
This study will help determine whether FibroX can be integrated into real-world primary care workflows to support earlier and more accurate detection of liver fibrosis and portal hypertension, potentially reducing missed diagnoses, unnecessary referrals, and improving patient outcomes.
This study is a 12-month pilot clinical trial designed to evaluate the feasibility, usability, provider trust, and preliminary effectiveness of FibroX, an explainable artificial intelligence (AI) tool developed to improve the diagnosis of significant liver fibrosis (≥F2) and clinically significant portal hypertension in adults with metabolic dysfunction-associated steatotic liver disease (MASLD). MASLD is a common and progressive liver condition that can lead to cirrhosis, liver failure, and increased cardiovascular risk. Early detection of these conditions is critical because current guidelines recommend initiating therapy (e.g., resmetirom or semaglutide for ≥F2 fibrosis and beta-blockers for portal hypertension). However, existing tools like FIB-4 often lack accuracy and usability in routine primary care.
FibroX addresses these limitations by using routinely available clinical data-such as age, liver enzymes, platelet count, BMI, and kidney function-to estimate the probability of significant fibrosis and portal hypertension. It provides a triage band (rule-out, indeterminate, rule-in) and a one-line explanation of which clinical factors most influenced the prediction. This transparency is achieved using Shapley Additive Explanations (SHAP), which helps clinicians understand how the AI reached its conclusion.
In retrospective studies, FibroX demonstrated superior diagnostic performance compared to FIB-4 (AUROC 0.97 vs. 0.62) and was associated with long-term mortality risk, suggesting prognostic value beyond diagnostic utility.
This pilot trial will simulate real-world primary care workflows to test whether FibroX can be effectively used by clinicians. The study will recruit 30-40 primary care providers (MDs, DOs, NPs, PAs) from 4-6 diverse clinics. Each provider will participate in two simulation periods, each involving 16 synthetic or de-identified patient cases reflecting adults with MASLD risk factors. Ground truth for fibrosis stage and portal hypertension will be determined by biopsy or expert consensus using Vibration-Controlled Transient Elastography (VCTE) and guideline-based criteria.
Providers will be randomly assigned to review cases in one of two sequences:
After a one-week washout period, providers will switch to the other condition. For each case, providers will make a management decision (e.g., no action, order VCTE, refer to hepatology), record their confidence level, and complete surveys on usability, trust in AI, and cognitive workload.
Primary Outcomes
Secondary Outcomes
All provider actions and decision times will be automatically logged. Post-period surveys and qualitative debriefs will explore barriers and facilitators to using FibroX.
Study Significance This pilot study will generate critical data to support a future multi-center trial and potential integration of FibroX into electronic health records. If successful, FibroX could enable scalable, guideline-concordant screening for significant liver fibrosis and portal hypertension in primary care, reducing missed diagnoses and unnecessary referrals. This aligns with national priorities for precision medicine and responsible AI implementation in healthcare.
Not provided
Not provided
Not provided
Not provided
| Label | Type | Description | Intervention Names |
|---|---|---|---|
| FibroX-Enabled Care | Experimental | In this arm, primary care providers use FibroX, an AI-powered clinical decision support tool, to assess simulated patient cases for significant liver fibrosis (≥F2) and clinically significant portal hypertension. FibroX displays a risk probability score, a triage band (rule-out, indeterminate, rule-in), and a one-line explainability panel showing which clinical factors most influenced the prediction. Providers use this information to make diagnostic and referral decisions (e.g., order VCTE, refer to hepatology, initiate guideline-based therapy). Each provider reviews 16 cases during this intervention period. The goal is to evaluate FibroX's impact on diagnostic accuracy, provider trust, usability, and workflow efficiency compared to usual care. |
|
| Usual Care | Active Comparator | In this arm, primary care providers assess simulated patient cases using standard clinical tools available in routine practice. These include laboratory results, vital signs, problem lists, medications, and prior imaging. Providers may optionally use the FIB-4 calculator to estimate liver fibrosis risk. Each provider reviews 16 cases during this period. No AI decision support is provided. This arm serves as the comparator to evaluate whether FibroX improves diagnostic accuracy for significant liver fibrosis (≥F2) and clinically significant portal hypertension, as well as provider trust, usability, and workflow efficiency over usual care. |
|
| Name | Type | Description | Arm Group Labels | Other Names |
|---|---|---|---|---|
| FibroX | Device | FibroX is an explainable artificial intelligence (AI) tool designed to assist primary care providers in diagnosing significant liver fibrosis (≥F2) and clinically significant portal hypertension in patients with metabolic dysfunction-associated steatotic liver disease (MASLD). It uses routinely available clinical data (e.g., age, AST, ALT, platelets, BMI, HbA1c, creatinine) to generate a risk probability score, a triage band (rule-out, indeterminate, rule-in), and a one-line explainability panel using Shapley Additive Explanations (SHAP). Providers use FibroX during simulated patient encounters to guide diagnostic and referral decisions (e.g., order VCTE, refer to hepatology, initiate guideline-based therapy). The tool aims to improve diagnostic accuracy, increase provider trust, reduce missed diagnoses, and support guideline-concordant triage in primary care. |
| Measure | Description | Time Frame |
|---|---|---|
| Diagnostic Accuracy for Significant Liver Fibrosis (≥F2) and Clinically Significant Portal Hypertension Using FibroX Compared to Usual Care | Within-provider diagnostic accuracy for detecting significant liver fibrosis (≥F2) and clinically significant portal hypertension in simulated primary care encounters. Accuracy will be assessed using sensitivity, specificity, and AUROC at clinically relevant thresholds. Ground truth for fibrosis stage and portal hypertension will be derived from biopsy, Vibration-Controlled Transient Elastography (VCTE)-based expert consensus, and guideline-defined criteria. Unit of Measure: Proportion (sensitivity and specificity in %, AUROC as a unitless value) | Immediately after each simulation period, up to 24 weeks |
| System Usability Scale (SUS) Score for FibroX Integration | Usability of FibroX assessed using the System Usability Scale (SUS), a validated 10-item questionnaire scored from 0 to 100, where higher scores indicate better usability. Unit of Measure: Score (range: 0-100; higher scores = better usability) | Immediately after each simulation period, up to 24 weeks |
| Provider Trust in AI Tool (FibroX) | Provider trust in FibroX assessed using the validated AI-Trust Scale, which includes 12 items scored on a Likert scale. Higher scores indicate greater trust in the AI tool. Unit of Measure: Score (range: 12-60; higher scores = greater trust) | Immediately after the FibroX-enabled simulation period, up to 24 weeks |
| Median Decision Time per Case | Median time (in minutes) taken by providers to complete management decisions for simulated MASLD cases using FibroX versus usual care. Unit of Measure: Minutes | Immediately after each simulation period, up to 24 weeks |
| Measure | Description | Time Frame |
|---|---|---|
| Appropriate Referral Rate | Proportion of simulated cases where provider referral decisions (e.g., hepatology referral, VCTE order) align with guideline-concordant triage rules for MASLD risk stratification. Unit of Measure: Proportion (%) | Immediately after each simulation period, up to 24 weeks |
| Net Reclassification Improvement (NRI) |
Not provided
Inclusion Criteria:
Exclusion Criteria:
Not provided
Not provided
Not provided
Not provided
Not provided
| Name | Role | Phone | Extension | |
|---|---|---|---|---|
| Basile Njei, MD, MPH, PhD, FRCP | Contact | 475-227-5537 | basile.njei@yale.edu | |
| Ulrick S Kanmounye, MD, MPH, MSc | Contact | 5705404973 | ulricksidney@gmail.com |
Not provided
Not provided
| PubMed Identifier | Type | Citation | Retractions |
|---|---|---|---|
| Result | Njei, B., et al., FIBROX: an explainable AI model for accurate prediction of advanced liver fibrosis and cardiovascular mortality in MASLD. Gastroenterology, 2024. 169(1): p. S-131-S-132. | ||
| 38615137 | Result | Njei B, Osta E, Njei N, Al-Ajlouni YA, Lim JK. An explainable machine learning model for prediction of high-risk nonalcoholic steatohepatitis. Sci Rep. 2024 Apr 13;14(1):8589. doi: 10.1038/s41598-024-59183-4. | |
| 15940625 |
| Label | URL |
|---|---|
| FibroX WebApp | View source |
Not provided
This pilot study involves simulated case reviews by primary care providers. No patient-level data is collected, and there is no current plan to share individual provider-level data with other researchers.
Not provided
Not provided
Not provided
Not provided
Not provided
| ID | Term |
|---|---|
| D008103 | Liver Cirrhosis |
| ID | Term |
|---|---|
| D008107 | Liver Diseases |
| D004066 | Digestive System Diseases |
| D005355 | Fibrosis |
| D010335 | Pathologic Processes |
Not provided
Not provided
This study uses a provider-level randomized crossover design in simulated clinical settings. Each primary care provider will complete two intervention periods in random order: one using usual care tools (standard labs and optional FIB-4 calculator), and one using FibroX, an AI-based decision support tool. Each period includes 16 simulated patient cases with MASLD risk factors. A one-week washout separates the periods. This crossover model allows within-provider comparison of diagnostic accuracy, usability, and decision-making between FibroX-assisted care and usual care, minimizing inter-provider variability and enhancing internal validity.
Not provided
Not provided
Not provided
Not provided
|
| Usual Care | Other | In the usual care condition, primary care providers assess simulated patient cases using standard clinical tools available in routine practice. These include laboratory results, vital signs, problem lists, medications, and prior imaging. Providers may optionally use the FIB-4 calculator to estimate liver fibrosis risk. No AI decision support is provided. This intervention serves as the comparator to evaluate whether FibroX improves diagnostic accuracy for significant liver fibrosis (≥F2) and clinically significant portal hypertension, as well as provider trust, decision-making quality, and workflow efficiency compared to usual care. |
|
Change in classification accuracy for MASLD risk categories (rule-out, indeterminate, rule-in) when using FibroX compared to usual care. Unit of Measure: NRI score (unitless) |
| Immediately after each simulation period, up to 24 weeks |
| Calibration of Risk Predictions | Calibration of FibroX predictions compared to observed outcomes, assessed using calibration intercept, slope, and calibration plot. Unit of Measure: Intercept and slope (unitless) | Immediately after each simulation period, up to 24 weeks |
| Provider Confidence in Decision-Making | Provider-reported confidence in MASLD management decisions during simulated cases, measured on a 5-point Likert scale (1 = very low confidence; 5 = very high confidence). Unit of Measure: Score (range: 1-5; higher scores = greater confidence) | Immediately after each simulation period, up to 24 weeks |
| Cognitive Load During Case Review | Provider cognitive workload assessed using NASA Task Load Index (NASA-TLX), which provides an overall workload score from 0 to 100 across six dimensions. Unit of Measure: Score (range: 0-100; higher scores = greater workload) | Immediately after each simulation period, up to 24 weeks |
| Intended Downstream Testing Burden | Number and type of additional tests or referrals (e.g., VCTE, hepatology consult) that providers intend to order after each simulated case. Unit of Measure: Count (number of tests/referrals per case) | Immediately after each simulation period, up to 24 weeks |
| Adoption and Fidelity to Triage Recommendations | Proportion of cases where providers follow FibroX triage recommendations (e.g., rule-out, indeterminate, rule-in) without override. Unit of Measure: Proportion (%) | Immediately after each simulation period, up to 24 weeks |
| Override Rate and Reasons | Proportion of cases where providers override FibroX recommendations and the documented reasons for override. Unit of Measure: Proportion (%) | Immediately after each simulation period, up to 24 weeks |
| Fairness Analysis Across Subgroups | Performance of FibroX (sensitivity, specificity, calibration) across demographic subgroups (age, sex, BMI, race/ethnicity). Unit of Measure: Proportion (%) and AUROC (unitless) | Immediately after each simulation period, up to 24 weeks |
| Result |
| Ratziu V, Charlotte F, Heurtier A, Gombert S, Giral P, Bruckert E, Grimaldi A, Capron F, Poynard T; LIDO Study Group. Sampling variability of liver biopsy in nonalcoholic fatty liver disease. Gastroenterology. 2005 Jun;128(7):1898-906. doi: 10.1053/j.gastro.2005.03.084. |
| 34987607 | Result | Decharatanachart P, Chaiteerakij R, Tiyarattanachai T, Treeprasertsuk S. Application of artificial intelligence in non-alcoholic fatty liver disease and liver fibrosis: a systematic review and meta-analysis. Ther Adv Gastroenterol. 2021 Dec 21;14:17562848211062807. doi: 10.1177/17562848211062807. eCollection 2021. |
| 25715361 | Result | Meng F, Zheng Y, Zhang Q, Mu X, Xu X, Zhang H, Ding L. Noninvasive evaluation of liver fibrosis using real-time tissue elastography and transient elastography (FibroScan). J Ultrasound Med. 2015 Mar;34(3):403-10. doi: 10.7863/ultra.34.3.403. |
| 21898504 | Result | Boursier J, de Ledinghen V, Zarski JP, Fouchard-Hubert I, Gallois Y, Oberti F, Cales P; multicentric groups from SNIFF 32, VINDIAG 7, and ANRS/HC/EP23 FIBROSTAR studies. Comparison of eight diagnostic algorithms for liver fibrosis in hepatitis C: new algorithms are more precise and entirely noninvasive. Hepatology. 2012 Jan;55(1):58-67. doi: 10.1002/hep.24654. |
| 20101745 | Result | Wong VW, Vergniol J, Wong GL, Foucher J, Chan HL, Le Bail B, Choi PC, Kowo M, Chan AW, Merrouche W, Sung JJ, de Ledinghen V. Diagnosis of fibrosis and cirrhosis using liver stiffness measurement in nonalcoholic fatty liver disease. Hepatology. 2010 Feb;51(2):454-62. doi: 10.1002/hep.23312. |
| 25007047 | Result | Yoon JH, Lee JM, Joo I, Lee ES, Sohn JY, Jang SK, Lee KB, Han JK, Choi BI. Hepatic fibrosis: prospective comparison of MR elastography and US shear-wave elastography for evaluation. Radiology. 2014 Dec;273(3):772-82. doi: 10.1148/radiol.14132000. Epub 2014 Jul 7. |
| 38143613 | Result | Mondal A, Debnath A, Dhandapani G, Sharma A, Lukhmana S, Yadav G. Prevalence of High and Moderate Risk of Liver Fibrosis Among Patients With Diabetes at a Noncommunicable Diseases (NCD) Clinic in a Primary Healthcare Center in Northern India. Cureus. 2023 Nov 23;15(11):e49286. doi: 10.7759/cureus.49286. eCollection 2023 Nov. |
| 29886156 | Result | Estes C, Anstee QM, Arias-Loste MT, Bantel H, Bellentani S, Caballeria J, Colombo M, Craxi A, Crespo J, Day CP, Eguchi Y, Geier A, Kondili LA, Kroy DC, Lazarus JV, Loomba R, Manns MP, Marchesini G, Nakajima A, Negro F, Petta S, Ratziu V, Romero-Gomez M, Sanyal A, Schattenberg JM, Tacke F, Tanaka J, Trautwein C, Wei L, Zeuzem S, Razavi H. Modeling NAFLD disease burden in China, France, Germany, Italy, Japan, Spain, United Kingdom, and United States for the period 2016-2030. J Hepatol. 2018 Oct;69(4):896-904. doi: 10.1016/j.jhep.2018.05.036. Epub 2018 Jun 8. |
| 38228377 | Result | Targher G, Byrne CD, Tilg H. MASLD: a systemic metabolic disorder with cardiovascular and malignant complications. Gut. 2024 Mar 7;73(4):691-702. doi: 10.1136/gutjnl-2023-330595. |
| 39389571 | Result | Maher S, Rajapakse J, El-Omar E, Zekry A. Role of the Gut Microbiome in Metabolic Dysfunction-Associated Steatotic Liver Disease. Semin Liver Dis. 2024 Nov;44(4):457-473. doi: 10.1055/a-2438-4383. Epub 2024 Oct 10. |
| 38727678 | Result | Younossi ZM, Mangla KK, Berentzen TL, Grau K, Kjaer MS, Ladelund S, Nitze LM, Coolbaugh C, Hsu CY, Hagstrom H. Liver histology is associated with long-term clinical outcomes in patients with metabolic dysfunction-associated steatohepatitis. Hepatol Commun. 2024 May 10;8(6):e0423. doi: 10.1097/HC9.0000000000000423. eCollection 2024 Jun 1. |
| 41974444 | Derived | Njei B, Kanmounye US. An Explainable AI Tool (FibroX) for Detecting Advanced Liver Fibrosis in Adults With Type 2 Diabetes: Protocol for a Pilot Crossover Trial. JMIR Res Protoc. 2026 May 21;15:e90456. doi: 10.2196/90456. |
| D013568 |
| Pathological Conditions, Signs and Symptoms |