Metabolic dysfunction-associated steatotic liver disease (MASLD) is a highly prevalent chronic liver condition globally, with its management spanning risk assessment, early diagnosis, staged intervention, and long-term follow-up. The integration of artificial intelligence (AI) and smart devices offers innovative approaches to address challenges in precision diagnosis and personalized treatment. This review synthesizes current applications of AI and smart devices across the MASLD care continuum, discusses their benefits and limitations, and outlines future directions to advance precision hepatology.
Introduction
MASLD, characterized by hepatic lipid accumulation linked to metabolic dysfunction, affects approximately 30% of adults worldwide. Its progression ranges from simple steatosis to steatohepatitis (MASH), fibrosis, cirrhosis, and hepatocellular carcinoma. The recent nomenclature shift from NAFLD to MASLD reflects a more inclusive diagnostic framework. Early identification and intervention are critical to mitigating disease progression and reducing healthcare burden. AI and smart devices are emerging as transformative tools in MASLD research and clinical practice, enabling improved risk prediction, non-invasive diagnosis, and personalized management.
Overview of AI and Smart Devices
AI encompasses machine learning (ML), deep learning, natural language processing (NLP), and data mining techniques. In healthcare, AI enhances early detection, diagnostic accuracy, and personalized therapy by analyzing multidimensional data from clinical indicators, biomarkers, and medical imaging. Smart devices—such as wearables and smartphones—equipped with sensors enable continuous monitoring of metabolic parameters (e.g., weight, glucose, blood pressure). Combined with AI analytics, these devices support dynamic health feedback and tailored interventions, facilitating real-time tracking and adaptive management of MASLD.
Application of AI in MASLD Risk Prediction and Diagnosis
Liver biopsy remains the gold standard but is invasive and unsuitable for population screening. AI models leveraging electronic health records (EHR), laboratory data, and multi-omics information have shown strong performance in predicting MASLD risk and severity. For instance, the N3-MASH model integrates CXCL10, CK-18, and adjusted BMI to improve MASH detection accuracy. In medical imaging, AI enhances the interpretation of ultrasound, CT, and MRI by automating steatosis quantification and reducing subjectivity. Deep learning models applied to ultrasound images achieve area under the curve (AUC) values up to 0.97, demonstrating high diagnostic reliability. MRI-based proton density fat fraction (PDFF) estimation, augmented by AI, offers precise non-invasive fat quantification, though cost and accessibility limit its use to high-risk patients or research settings.
AI-Based Analytical Models for Liver Histopathology
AI and digital pathology are revolutionizing histological assessment by automatically detecting and quantifying key features such as steatosis, inflammation, ballooning degeneration, and fibrosis. Quantum machine learning (QML) approaches show promise in improving classification accuracy while addressing data privacy concerns. AI-enabled dynamic monitoring of histological changes during therapy enhances sensitivity in detecting fibrosis regression, supporting more objective evaluation of treatment efficacy.
Treatment of MASLD
Lifestyle modification remains the cornerstone of MASLD management, complemented by pharmacotherapy when necessary. AI and smart devices enable personalized lifestyle interventions through real-time monitoring of diet, physical activity, and metabolic indicators. AI chatbots and mobile health applications provide adherence support, education, and psychological assistance. In drug development, AI accelerates target identification, virtual screening, and clinical trial design. For example, AI contributed to the development of resmetirom, the first FDA-approved drug for MASH, by optimizing candidate selection and trial endpoint assessment. Traditional Chinese medicine compounds have also been explored through AI-driven target prediction and molecular docking.
Ethical, Regulatory, and Data Security Considerations
The integration of AI in MASLD care raises important ethical and regulatory challenges, including algorithmic bias, lack of transparency, and data privacy. Bias in training data may exacerbate health disparities, necessitating diverse and inclusive datasets. Explainable AI (XAI) methods are critical to improving clinician trust. Regulatory frameworks such as HIPAA (USA), GDPR (EU), and PDPO (Hong Kong) provide guidance on data protection, though gaps remain. Technical strategies like federated learning, differential privacy, and synthetic data generation help safeguard patient information while enabling collaborative research.
Clinical Translation: Beneficial Populations and Scenarios
AI tools offer significant value for healthcare professionals in hepatology and primary care by reducing diagnostic variability and automating workflows. In resource-limited settings, AI-enabled portable ultrasound can decentralize screening. High-risk individuals with obesity or diabetes benefit from AI-driven early detection models. In clinical trials, AI enhances endpoint assessment through reproducible quantification of histological and imaging changes. Key application scenarios include risk stratification, lifestyle adherence promotion, remote monitoring, and drug response evaluation.
Strengths, Challenges, and Future Directions
AI strengthens MASLD management by improving diagnostic precision, enabling personalized interventions, and accelerating drug discovery. However, challenges persist: limited generalizability due to retrospective single-center data, "black-box" opacity of deep learning models, underrepresentation of diverse etiologies, and inequities in access to technology. Future research should focus on developing interpretable, multimodal AI systems validated across large multicenter cohorts. Ethical frameworks and privacy-preserving techniques must evolve alongside technological advances. Real-world implementation studies are needed to assess usability, cost-effectiveness, and long-term impact.
Conclusion
AI and smart devices are reshaping MASLD management by enhancing early detection, diagnostic accuracy, and personalized therapy. From imaging analysis and digital pathology to wearable-enabled monitoring and AI-driven drug discovery, these technologies support a shift toward precision medicine. Overcoming technical, ethical, and translational barriers will require collaborative efforts in algorithm development, standardized validation, regulatory harmonization, and inclusive deployment. With continued innovation, AI-integrated systems promise to make MASLD care more intelligent, accessible, and effective.
Full text
https://www.xiahepublishing.com/2310-8819/JCTH-2025-00406
The study was recently published in the Journal of Clinical and Translational Hepatology .
The Journal of Clinical and Translational Hepatology (JCTH) is owned by the Second Affiliated Hospital of Chongqing Medical University and published by XIA & HE Publishing Inc. JCTH publishes high quality, peer reviewed studies in the translational and clinical human health sciences of liver diseases. JCTH has established high standards for publication of original research, which are characterized by a study's novelty, quality, and ethical conduct in the scientific process as well as in the communication of the research findings. Each issue includes articles by leading authorities on topics in hepatology that are germane to the most current challenges in the field. Special features include reports on the latest advances in drug development and technology that are relevant to liver diseases. Regular features of JCTH also include editorials, correspondences and invited commentaries on rapidly progressing areas in hepatology. All articles published by JCTH, both solicited and unsolicited, must pass our rigorous peer review process.
