Artificial Intelligence Improves Pathologist Agreement for Fibrosis Scores in Nonalcoholic Steatohepatitis Patients

Envisioning how to advance the MASH field

Since 1980, the cumulative effort of scientists and health-care stakeholders has advanced the prerequisites to address metabolic dysfunction-associated steatotic liver disease (MASLD), a prevalent chronic non-communicable liver disease. This effort has led to, among others, the approval of the first drug specific for metabolic dysfunction-associated steatohepatitis (MASH; formerly known as nonalcoholic steatohepatitis). Despite substantial progress, MASLD is still a leading cause of advanced chronic liver disease, including primary liver cancer. This Perspective contextualizes the nomenclature change from nonalcoholic fatty liver disease to MASLD and proposes important considerations to accelerate further progress in the field, optimize patient-centric multidisciplinary care pathways, advance pharmacological, behavioural and diagnostic research, and address health disparities. Key regulatory and other steps necessary to optimize the approval and access to upcoming additional pharmacological therapeutic agents for MASH are also outlined. We conclude by calling for increased education and awareness, enhanced health system preparedness, and concerted action by policy-makers to further the public health and policy agenda to achieve at least parity with other non-communicable diseases and to aid in growing the community of practice to reduce the human and economic burden and end the public health threat of MASLD and MASH by 2030.

Digital quantitation of bridging fibrosis and septa reveals changes in natural history and treatment not seen with conventional histology

BACKGROUND AND AIMS

Metabolic dysfunction-associated steatohepatitis (MASH) with bridging fibrosis is a critical stage in the evolution of fatty liver disease. Second harmonic generation/two-photon excitation fluorescence (SHG/TPEF) microscopy with artificial intelligence (AI) provides sensitive and reproducible quantitation of liver fibrosis. This methodology was applied to gain an in-depth understanding of intra-stage fibrosis changes and septa analyses in a homogenous, well-characterised group with MASH F3 fibrosis.

METHODS

Paired liver biopsies (baseline [BL] and end of treatment [EOT]) of 57 patients (placebo, n = 17 and tropifexor n = 40), with F3 fibrosis stage at BL according to the clinical research network (CRN) scoring, were included. Unstained sections were examined using SHG/TPEF microscopy with AI. Changes in liver fibrosis overall and in five areas of liver lobules were quantitatively assessed by qFibrosis. Progressive, regressive septa, and 12 septa parameters were quantitatively analysed.

RESULTS

qFibrosis demonstrated fibrosis progression or regression in 14/17 (82%) patients receiving placebo, while the CRN scoring categorised 11/17 (65%) as 'no change'. Radar maps with qFibrosis readouts visualised quantitative fibrosis dynamics in different areas of liver lobules even in cases categorised as 'No Change'. Measurement of septa parameters objectively differentiated regressive and progressive septa (p < .001). Quantitative changes in individual septa parameters (BL to EOT) were observed both in the 'no change' and the 'regression' subgroups, as defined by the CRN scoring.

CONCLUSION

SHG/TPEF microscopy with AI provides greater granularity and precision in assessing fibrosis dynamics in patients with bridging fibrosis, thus advancing knowledge development of fibrosis evolution in natural history and in clinical trials.

Deep learning and digital pathology powers prediction of HCC development in steatotic liver disease

BACKGROUND AND AIMS

Identifying patients with steatotic liver disease who are at a high risk of developing HCC remains challenging. We present a deep learning (DL) model to predict HCC development using hematoxylin and eosin-stained whole-slide images of biopsy-proven steatotic liver disease.

APPROACH AND RESULTS

We included 639 patients who did not develop HCC for ≥7 years after biopsy (non-HCC class) and 46 patients who developed HCC <7 years after biopsy (HCC class). Paired cases of the HCC and non-HCC classes matched by biopsy date and institution were used for training, and the remaining nonpaired cases were used for validation. The DL model was trained using deep convolutional neural networks with 28,000 image tiles cropped from whole-slide images of the paired cases, with an accuracy of 81.0% and an AUC of 0.80 for predicting HCC development. Validation using the nonpaired cases also demonstrated a good accuracy of 82.3% and an AUC of 0.84. These results were comparable to the predictive ability of logistic regression model using fibrosis stage. Notably, the DL model also detected the cases of HCC development in patients with mild fibrosis. The saliency maps generated by the DL model highlighted various pathological features associated with HCC development, including nuclear atypia, hepatocytes with a high nuclear-cytoplasmic ratio, immune cell infiltration, fibrosis, and a lack of large fat droplets.

CONCLUSIONS

The ability of the DL model to capture subtle pathological features beyond fibrosis suggests its potential for identifying early signs of hepatocarcinogenesis in patients with steatotic liver disease.

Consensus statements on the current landscape of artificial intelligence applications in endoscopy, addressing roadblocks, and advancing artificial intelligence in gastroenterology

BACKGROUND AND AIMS

The American Society for Gastrointestinal Endoscopy (ASGE) AI Task Force along with experts in endoscopy, technology space, regulatory authorities, and other medical subspecialties initiated a consensus process that analyzed the current literature, highlighted potential areas, and outlined the necessary research in artificial intelligence (AI) to allow a clearer understanding of AI as it pertains to endoscopy currently.

METHODS

A modified Delphi process was used to develop these consensus statements.

RESULTS

Statement 1: Current advances in AI allow for the development of AI-based algorithms that can be applied to endoscopy to augment endoscopist performance in detection and characterization of endoscopic lesions. Statement 2: Computer vision-based algorithms provide opportunities to redefine quality metrics in endoscopy using AI, which can be standardized and can reduce subjectivity in reporting quality metrics. Natural language processing-based algorithms can help with the data abstraction needed for reporting current quality metrics in GI endoscopy effortlessly. Statement 3: AI technologies can support smart endoscopy suites, which may help optimize workflows in the endoscopy suite, including automated documentation. Statement 4: Using AI and machine learning helps in predictive modeling, diagnosis, and prognostication. High-quality data with multidimensionality are needed for risk prediction, prognostication of specific clinical conditions, and their outcomes when using machine learning methods. Statement 5: Big data and cloud-based tools can help advance clinical research in gastroenterology. Multimodal data are key to understanding the maximal extent of the disease state and unlocking treatment options. Statement 6: Understanding how to evaluate AI algorithms in the gastroenterology literature and clinical trials is important for gastroenterologists, trainees, and researchers, and hence education efforts by GI societies are needed. Statement 7: Several challenges regarding integrating AI solutions into the clinical practice of endoscopy exist, including understanding the role of human-AI interaction. Transparency, interpretability, and explainability of AI algorithms play a key role in their clinical adoption in GI endoscopy. Developing appropriate AI governance, data procurement, and tools needed for the AI lifecycle are critical for the successful implementation of AI into clinical practice. Statement 8: For payment of AI in endoscopy, a thorough evaluation of the potential value proposition for AI systems may help guide purchasing decisions in endoscopy. Reliable cost-effectiveness studies to guide reimbursement are needed. Statement 9: Relevant clinical outcomes and performance metrics for AI in gastroenterology are currently not well defined. To improve the quality and interpretability of research in the field, steps need to be taken to define these evidence standards. Statement 10: A balanced view of AI technologies and active collaboration between the medical technology industry, computer scientists, gastroenterologists, and researchers are critical for the meaningful advancement of AI in gastroenterology.

CONCLUSIONS

The consensus process led by the ASGE AI Task Force and experts from various disciplines has shed light on the potential of AI in endoscopy and gastroenterology. AI-based algorithms have shown promise in augmenting endoscopist performance, redefining quality metrics, optimizing workflows, and aiding in predictive modeling and diagnosis. However, challenges remain in evaluating AI algorithms, ensuring transparency and interpretability, addressing governance and data procurement, determining payment models, defining relevant clinical outcomes, and fostering collaboration between stakeholders. Addressing these challenges while maintaining a balanced perspective is crucial for the meaningful advancement of AI in gastroenterology.

Potential of Using qFibrosis Analysis to Predict Recurrent and Survival Outcome of Patients with Hepatocellular Carcinoma after Hepatic Resection

BACKGROUND

There remains a lack of studies addressing the stromal background and fibrosis features and their prognostic value in liver cancer. qFibrosis can identify, quantify, and visualize the fibrosis features in biopsy samples. In this study, we aim to demonstrate the prognostic value of histological features by using qFibrosis analysis in liver cancer patients.

METHODS

Liver specimens from 201 patients with hepatocellular carcinoma (HCC) who underwent curative resection were imaged and assessed using qFibrosis system and generated a total of 33 and 156 collagen parameters from tumor part and non-tumor liver tissue, respectively. We used these collagen parameters on patients to build two combined indexes, RFS index and OS index, in order to differentiate patients with early recurrence and early death, respectively. The models were validated using the leave-one-out method.

RESULTS

Both combined indexes had significant prediction value for patients' outcome. The RFS index of 0.52 well differentiates patients with early recurrence (p < 0.001), and the OS index of 0.73 well differentiates patients with early death during follow-up (p = 0.02).

CONCLUSIONS

Combined index calculated with qFibrosis from a digital readout of the fibrotic status of peri-tumor liver specimen in patients with HCC has prediction values for their disease and survival outcomes. These results demonstrated the potential to transform histopathological features into quantifiable data that could be used to correlate with clinical outcome.

Artificial intelligence-assisted digital pathology for non-alcoholic steatohepatitis: current status and future directions

The worldwide prevalence of non-alcoholic steatohepatitis (NASH) is increasing, causing a significant medical burden, but no approved therapeutics are currently available. NASH drug development requires histological analysis of liver biopsies by expert pathologists for trial enrolment and efficacy assessment, which can be hindered by multiple issues including sample heterogeneity, inter-reader and intra-reader variability, and ordinal scoring systems. Consequently, there is a high unmet need for accurate, reproducible, quantitative, and automated methods to assist pathologists with histological analysis to improve the precision around treatment and efficacy assessment. Digital pathology (DP) workflows in combination with artificial intelligence (AI) have been established in other areas of medicine and are being actively investigated in NASH to assist pathologists in the evaluation and scoring of NASH histology. DP/AI models can be used to automatically detect, localise, quantify, and score histological parameters and have the potential to reduce the impact of scoring variability in NASH clinical trials. This narrative review provides an overview of DP/AI tools in development for NASH, highlights key regulatory considerations, and discusses how these advances may impact the future of NASH clinical management and drug development. This should be a high priority in the NASH field, particularly to improve the development of safe and effective therapeutics.

Digital pathology for nonalcoholic steatohepatitis assessment

Histological assessment of nonalcoholic fatty liver disease (NAFLD) has anchored knowledge development about the phenotypes of the condition, their natural history and their clinical course. This fact has led to the use of histological assessment as a reference standard for the evaluation of efficacy of drug interventions for nonalcoholic steatohepatitis (NASH) - the more histologically active form of NAFLD. However, certain limitations of conventional histological assessment systems pose challenges in drug development. These limitations have spurred intense scientific and commercial development of machine learning and digital approaches towards the assessment of liver histology in patients with NAFLD. This research field remains an area in rapid evolution. In this Perspective article, we summarize the current conventional assessment of NASH and its limitations, the use of specific digital approaches for histological assessment, and their application to the study of NASH and its response to therapy. Although this is not a comprehensive review, the leading tools currently used to assess therapeutic efficacy in drug development are specifically discussed. The potential translation of these approaches to support routine clinical assessment of NAFLD and an agenda for future research are also discussed.

Fibrosis severity scoring on Sirius red histology with multiple-instance deep learning

Non-alcoholic fatty liver disease (NAFLD) is now the leading cause of chronic liver disease, affecting approximately 30% of people worldwide. Histopathology reading of fibrosis patterns is crucial to diagnosing NAFLD. In particular, separating mild from severe stages corresponds to a critical transition as it correlates with clinical outcomes. Deep Learning for digitized histopathology whole-slide images (WSIs) can reduce high inter- and intra-rater variability. We demonstrate a novel solution to score fibrosis severity on a retrospective cohort of 152 Sirius-Red WSIs, with fibrosis stage annotated at slide level by an expert pathologist. We exploit multiple instance learning and multiple-inferences to address the sparsity of pathological signs. We achieved an accuracy of , an F1 score of and an AUC of . These results set new state-of-the-art benchmarks for this application.

A researcher's guide to preclinical mouse NASH models

Non-alcoholic fatty liver disease (NAFLD) and its inflammatory form, non-alcoholic steatohepatitis (NASH), have quickly risen to become the most prevalent chronic liver disease in the Western world and are risk factors for the development of hepatocellular carcinoma (HCC). HCC is not only one of the most common cancers but is also highly lethal. Nevertheless, there are currently no clinically approved drugs for NAFLD, and NASH-induced HCC poses a unique metabolic microenvironment that may influence responsiveness to certain treatments. Therefore, there is an urgent need to better understand the pathogenesis of this rampant disease to devise new therapies. In this line, preclinical mouse models are crucial tools to investigate mechanisms as well as novel treatment modalities during the pathogenesis of NASH and subsequent HCC in preparation for human clinical trials. Although, there are numerous genetically induced, diet-induced and toxin-induced models of NASH, not all of these models faithfully phenocopy and mirror the human pathology very well. In this Perspective, we shed some light onto the most widely used mouse models of NASH and highlight some of the key advantages and disadvantages of the various models with an emphasis on 'Western diets', which are increasingly recognized as some of the best models in recapitulating the human NASH pathology and comorbidities.