Clinical Scorecard: Utilizing Machine Learning for the Discovery of Biomarkers Aiding in the Early Detection of Metabolic Dysfunction–Associated Steatotic Liver Disease
At a Glance
Category
Detail
Condition
Metabolic dysfunction–associated steatotic liver disease (MASLD), including hepatic steatosis and metabolic dysfunction–associated steatohepatitis
Key Mechanisms
Metabolic dysfunction linked to obesity, insulin resistance, dyslipidemia, and inflammation leading to liver fat accumulation and fibrosis
Target Population
Adults with metabolic risk factors such as obesity, type 2 diabetes, and metabolic syndrome
Care Setting
Clinical settings utilizing noninvasive imaging (FibroScan) and biomarker assessment for early diagnosis and monitoring
Key Highlights
MASLD is the most common chronic liver disorder worldwide and closely associated with metabolic syndrome and cardiovascular disease.
Liver biopsy is the traditional diagnostic gold standard but is invasive; noninvasive imaging like FibroScan offers safer, repeatable assessment.
Machine learning models identified key biomarkers (age, BMI, insulin) predictive of MASLD and hepatic fibrosis with high accuracy.
Guideline-Based Recommendations
Diagnosis
Use noninvasive imaging modalities such as transient elastography (FibroScan) to measure hepatic steatosis (CAP score ≥238 dB/m) and stiffness (liver stiffness ≥7 kPa).
Incorporate biomarker profiles including demographic, metabolic, lipid, and biochemical markers to aid early MASLD diagnosis.
Management
Focus on addressing metabolic risk factors such as obesity, insulin resistance, and dyslipidemia to manage MASLD progression.
Monitoring & Follow-up
Utilize repeat FibroScan assessments for ongoing evaluation of hepatic steatosis and fibrosis.
Monitor relevant biomarkers (e.g., insulin, HbA1c, lipid profile) to assess disease status and treatment response.
Risks
Avoid liver biopsy for large-scale screening due to invasiveness and associated risks.
Consider limitations of imaging modalities including operator dependence and altered results in certain patient populations.
Patient & Prescribing Data
Adults with metabolic dysfunction and risk factors for MASLD
Early identification of MASLD using biomarker profiles and FibroScan can guide timely interventions targeting metabolic syndrome components.
Clinical Best Practices
Exclude patients with viral hepatitis and significant alcohol use when assessing MASLD to avoid confounding.
Apply machine learning techniques (Random Forest, XGBoost) to identify and validate key biomarkers predictive of MASLD and fibrosis.
Combine demographic, metabolic, lipid, and biochemical markers for comprehensive risk assessment.
by Jolie Boullion, Amanda Husein, Akshat Agrawal, Diensn Xing, Md Ismail Hossain, Md Shenuarin Bhuiyan, Oren Rom, Steven A Conrad, John A Vanchiere, A Wayne Orr, Christopher G Kevil, Mohammad Alfrad Nobel Bhuiyan
