Expanding diagnosis and treatment of chronic liver disease with AI and digital innovations

Our Take:

Non-alcoholic fatty liver disease (NAFLD) affects 25% of the world's population, posing significant healthcare challenges as it often goes undetected due to limitations in screening. Traditional diagnostic techniques, such as liver biopsies, are invasive and costly, however, newer non-invasive alternatives like FibroScan, Magnetic Resonance Elastography (MRE), and Shear Wave Elastography (SWE) offer promising solutions. While integrating these technologies remains difficult due to cost and access barriers. These therapies hold promise in broadening access to diagnostics and treatments. Expanding these advancements via digital health platforms, mobile units, and community-based initiatives could improve outcomes for populations disproportionately affected by NAFLD, ultimately aiding in reducing healthcare costs over time.

Key Takeaways:

Digital tools like FibroScan and AI diagnostics are improving early NAFLD detection, with FibroScan achieving 90% specificity for advanced fibrosis (Bioengineering, Basel)
NAFLD affects 25% of the population globally and nearly 100M people in the U.S. alone (Hepatology; American Liver Foundation)
Among patients with NAFLD, up to 35% progress to non-alcoholic steatohepatitis (NASH), and about 20% of these patient's illness advances to cirrhosis within 2 years (Journal of Clinical Medicine)
Emerging therapies, such as semaglutide and obeticholic acid, offer promising reductions in liver fat (59%) and fibrosis (23%), providing new treatment options (Annals of Gastroenterology)

The Problem:

NAFLD has quickly become one of the leading causes of chronic liver disease globally, affecting about 25% of the world’s population. In the U.S., NAFLD is now on par with more recognized causes of cirrhosis, like alcohol-induced liver disease. This is especially true among populations experiencing increasing rates of obesity, type 2 diabetes, and metabolic syndrome. Unfortunately, NAFLD often remains undiagnosed and untreated due to screening challenges. Current diagnostic methods, such as liver biopsy, ultrasound, and MRI, can be either too invasive or too expensive to be widely used for screening. This situation has created a noticeable gap in early detection and care, as real-world evidence indicates that diagnoses of NAFLD in primary care are much lower than predicted, indicating many cases are being missed or not properly recorded. By contrast, using early screening with digital technologies presents one of the best chances to alleviate the burden of NAFLD. Longer term, this approach could improve outcomes and reduce costs by preventing advanced liver disease and its complications.

Importantly, NAFLD disproportionately affects those of lower socioeconomic status, highlighting the importance of addressing social determinants of health when implementing digital screening tools. Mobile screening units and telemedicine platforms can deliver liver health diagnostics directly to underserved areas, reducing geographic and financial barriers to care. Collaborating with community organizations can enhance awareness and boost participation in screening programs among at-risk groups that might otherwise be overlooked. While expanding NAFLD screening presents certain logistical and cost challenges, research indicates that the potential benefits of preventing serious liver diseases likely outweigh these limitations. Combined with policies focused on improving care access and tackling social determinants of health, these digital solutions can help alleviate the financial and operational concerns of mass screening opponents. A review in the Journal of Hepatology recently advocated for "an investment framework to elevate the profile of NAFLD/NASH and achieve significant cost efficiencies by directing decision-makers on effective actions to take."

One of the greatest challenges is the silent nature of NAFLD. For example, many patients don’t show symptoms complicating diagnosis and by the time symptoms become apparent evidence suggests the disease may have already advanced to more serious stages like cirrhosis, or even liver cancer. A recent scoping review published in the Journal of Clinical Medicine reveals that among patients with NAFLD, up to 35% progress to NASH, and about 20% of these rapidly decline to cirrhosis within just 2 years, greatly increasing their risk of liver-related complications and mortality. The review also points out significant inequalities in the prevalence and outcomes of NAFLD, influenced by social determinants of health such as gender, ethnicity, and socioeconomic status. The authors emphasize the need for more research on how these factors intersect and a deeper understanding of how gene-environment interactions contribute to NAFLD.

The Backdrop:

The diagnostic landscape for NAFLD has seen remarkable improvements over the past two decades. Transient elastography, commonly known as FibroScan, is an exciting development in NAFLD screening. This innovative tool uses ultrasound technology to measure liver stiffness, a vital indicator of fibrosis, which signals the progression of liver damage. It’s quick, painless, and more affordable than a biopsy or MRI, making it a viable option for broader screening. However, its use in routine clinical practice is still limited due to equipment cost, training, and its fit into workflows. Other alternatives to FibroScan include Magnetic Resonance Elastography (MRE), Shear Wave Elastography (SWE), and Acoustic Radiation Force Impulse Imaging (ARFI). While MRE is known for its accuracy, it is expensive and usually found in specialized settings. SWE works well with standard ultrasound machines, offering great flexibility, although it’s not as established in NAFLD diagnosis as FibroScan. ARFI is portable and cost-effective, but it hasn’t undergone comprehensive validation. Although FibroScan is widely adopted, we should note that its accuracy may decrease in patients with a higher body mass.

Moreover, AI applications and emerging AI-driven diagnostic algorithms hold the potential to improve NAFLD screening even further. These algorithms can analyze large datasets, including imaging and lab results, to identify patients at risk for NAFLD or NASH sooner and more accurately than traditional methods. For example, a recent study in the journal Therapeutic Advances in Gastroenterology highlighted that “AI-supported systems provide promising performance improvements for diagnosing NAFLD, NASH, and identifying liver fibrosis among NAFLD patients.” Trials like this show how AI can enhance workflow efficiency, enabling clinicians to screen larger populations and focus attention on high-risk patients for more thorough follow-up. In addition, another article, "The Digital Transformation of Hepatology," highlights the importance of mobile health apps, wearables, telemedicine, and AI in improving diagnostics and management of cirrhosis and NAFLD. These solutions can personalize care, allow remote monitoring, and offer cost-effective interventions, enhancing accessibility. While challenges remain such as validating the technologies, ensuring usability, and addressing the digital divide in vulnerable groups, integrating these innovations into EMRs can streamline workflows, lessen administrative burdens, and ensure timely follow-ups for high-risk patients, ultimately improving outcomes. As these technologies advance, making them scalable, affordable, and accessible in low-resource settings remains crucial.

Therapeutically, NAFLD management has traditionally focused on lifestyle changes, such as modifying diets and increasing exercise, to help reduce liver fat and inflammation. While these adjustments can be quite effective, adherence can be challenging, especially for patients facing socioeconomic barriers like health literacy, education or access to healthy food. Several new pharmacological therapies for NAFLD have shown great promise in clinical trials in the past few years. For example, GLP-1 receptor agonists like Semaglutide, have made significant strides in reducing liver fat and fibrosis progression in a recent phase 2 trial. Currently, in late-stage trials, obeticholic acid (OCA) aims to reduce fibrosis by 23% in NASH patients by targeting bile acid pathways. Lanifibranor, a pan-PPAR agonist, has also proven to be effective in reversing fibrosis and improving liver function.

As these innovative therapies move closer to approval, they highlight the critical importance of early screening and intervention in managing NAFLD. Nevertheless, it’s important to recognize that some experts oppose mass screening for NAFLD, citing concerns over the financial strain and the lack of FDA-approved treatments, preferring the current management approach of focusing on lifestyle changes like weight loss. Critics argue that screening may not improve outcomes if patients rely heavily on lifestyle adherence. However, early detection may be crucial for preventing serious conditions like liver failure or cancer. As new treatments near approval, earlier screening can help identify those who can benefit, shifting care from costly reactive interventions to more effective preventive approaches.

Implications:

Widespread access to screening tools is crucial for NAFLD diagnosis beyond specialized settings. Technologies like FibroScan, MRE, SWE, and ARFI must be affordable for small clinics and rural centers to scale effectively. Innovative financing approaches like public-private partnerships can help subsidize costs, making them feasible for lower-resource environments, while standardized training can enable primary care providers to use these technologies, expanding their application beyond liver clinics. Technological tools like cloud platforms for shared screening data can reduce financial burdens on facilities and enhance liver health monitoring across care settings, and new therapies, including semaglutide, obeticholic acid, and lanifibranor, create new opportunities for tailored NAFLD care and reimbursement. Moreover, new digital tools like remote patient monitoring (RPM) allow real-time tracking of liver health, enabling clinicians to adjust treatments without the expense of costly follow-up visits. AI advances like predictive analytics can optimize treatment decisions based on patient profiles, improving outcomes while minimizing unnecessary interventions.

However, integrating these tools into preventive care requires greater routine screening for high-risk groups, such as those with obesity, diabetes, or metabolic syndrome. This way AI protocols can identify and flag high-risk patients, ensuring early and regular screenings rather than reactive measures after symptoms arise. As the diagnostic benefits of early detection and prevention become apparent, we expect Insurance companies to recognize them and cover non-invasive tools like FibroScan. Over time this should eliminate cost barriers for patients and prompting healthcare providers to include liver screening in check-ups. A UK study showed that integrating non-invasive liver fibrosis tests into care pathways for NAFLD increased advanced liver fibrosis detection by up to 137% while reducing healthcare costs. This led to a 25% decrease in spending per case detected, showcasing a cost-effective strategy for NAFLD management. Integrating screening data into electronic health records will facilitate ongoing patient management, allowing clinicians to track disease progression and implement early interventions.

The competition in NAFLD screening technologies offers opportunities for innovation in digital health. By combining diagnostic tools with AI and cloud platforms, we can create comprehensive liver health solutions that streamline care for providers and patients. With rapid advancements in NAFLD tools and therapies, enhancing care is possible. To unlock the full potential of these innovations, we must focus on scalability, integration into preventive care, equitable access, personalized treatments, and continuous exploration of digital health. By taking these steps, we can improve health outcomes for millions at risk for NAFLD and reduce the burden on our healthcare systems.

Related Reading: