Artificial Intelligence-Based Opportunities in Liver Pathology—A Systematic Review

Allaume, Pierre; Rabilloud, Noémie; Turlin, Bruno; Bardou-Jacquet, Édouard; Loréal, Olivier; Caldéraro, Julien; Khene, Z.; Acosta, Oscar; Crevoisier, R. de; Rioux‐Leclercq, Nathalie; Pécot, Thierry; Kammerer‐Jacquet, Solène‐Florence

doi:10.3390/diagnostics13101799

Cited by 9 publications

(6 citation statements)

References 74 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This suggests that a larger sample of patients is needed to investigate these markers since our population with CK19+ was small. Translating this data into molecular profiling and resorting to AI algorithms may also be helpful [64,65].…”

Section: Discussionmentioning

confidence: 99%

Prognostic Factors after Hepatectomy for Hepatocellular Carcinoma—The Importance of Pathological Immunophenotyping, the Steatohepatitic Subtype and the Impact of the Hepatic Pedicle Clamping

Viana,

Oliveira,

Martins

et al. 2024

Gastrointestinal Disorders

View full text Add to dashboard Cite

Introduction: Hepatectomy (HP) is, along with liver transplantation, the only potentially curative treatment for Hepatocellular Carcinoma (HCC). The high prevalence of Metabolic Syndrome (MS) may be causing a shift in the HCC spectrum. Hepatic Pedicle Clamping (HPC), used to reduce perioperative bleeding during HP, has been theorized to increase the risk of recurrence. Cytokeratin 19 (CK19) and glypican-3 (GLP-3) have been identified as markers of worse prognosis in HCC. Materials and Methods: A clinical and pathological review of 59 patients undergoing HP for HCC between 2005 and 2013 was performed. Chronic liver disease was observed in 53 patients (89.8%), with cirrhosis in 54.2% [most frequent etiologies: ethylism (47.5%), HCV (25.4%) and HBV (11.9%)]. MS was in 36% of patients. In addition, 95% of patients had Child–Pugh class A and 5% class B, and there was a median MELD of 8 (6–18). A single nodule was observed in 46 patients (78%) with an average size of 5.4 cm. Microscopic vascular invasion (MiVI) was in 49% of patients and macroscopic (MaVI) in 17. HPC was in 43 patients (74.1%). Statistical analysis was performed with SPSS™ 21.0. Survival tests (Kaplan–Meier, log-rank and Cox regression). Statistical significance was with p < 0.05. Results: Major morbidity in 22% of patients. Mortality in 5.1%. Median overall survival (OS) of 71 months and median disease-free survival (DFS) of 37. In a multivariate analysis: MaVI (p = 0.001), MiVI (p = 0.005) and HCV infection (p = 0.002) were associated with worse OS; MS was associated with better OS (p = 0.001); MaVI (p = 0.000), MiVI (p = 0.035) and HPC (p = 0.012) were associated with worse DFS. CK19+/GLP-3− (p = 0.007) and CK19−/GLP-3+ (p = 0.029) patients were associated with worse DFS and CK19−/GLP-3− (p = 0.031) with better DFS. Discussion/Conclusions: HPC was an independent factor of worse DFS. The ischemia-reperfusion injury (IRI) produced by HPC could promote a more angiogenic and angioinvasive phenotype of tumor cells, resulting in higher recurrence. HCV etiology was associated with worse OS. MS was associated with better OS, highlighting the importance of a hepatectomy in these cases. The combined detection of CK19 and GLP-3 was an independent prognostic factor in HCC patients allowing for the identification of more aggressive tumors.

show abstract

Section: Discussionmentioning

confidence: 99%

Prognostic Factors after Hepatectomy for Hepatocellular Carcinoma—The Importance of Pathological Immunophenotyping, the Steatohepatitic Subtype and the Impact of the Hepatic Pedicle Clamping

Viana,

Oliveira,

Martins

et al. 2024

Gastrointestinal Disorders

View full text Add to dashboard Cite

show abstract

“…These reasons have led to the increased demand for analysis using CAD systems, which require some steps for the inspection of histopathological images, including the division of WSI's into patches of reduced size (with or without overlap), the subsequent steps of pre-processing, segmentation, features extraction, and application of ML or DL approaches. Of note, a recent article provided a comprehensive, systematic, and updated discussion of the literature evidence on the application of DL approaches on WSIs obtained via liver histopathology [98]. In particular, the authors evaluated the selected articles by highlighting their performance and bias through a useful tool known as the Quality Assessment of Diagnostic Accuracy Studies version 2 (QUADAS-2).…”

Section: State-of-the-artmentioning

confidence: 99%

The Use of Artificial Intelligence in the Liver Histopathology Field: A Systematic Review

Grignaffini,

Barbuto,

Troiano

et al. 2024

Diagnostics

View full text Add to dashboard Cite

Digital pathology (DP) has begun to play a key role in the evaluation of liver specimens. Recent studies have shown that a workflow that combines DP and artificial intelligence (AI) applied to histopathology has potential value in supporting the diagnosis, treatment evaluation, and prognosis prediction of liver diseases. Here, we provide a systematic review of the use of this workflow in the field of hepatology. Based on the PRISMA 2020 criteria, a search of the PubMed, SCOPUS, and Embase electronic databases was conducted, applying inclusion/exclusion filters. The articles were evaluated by two independent reviewers, who extracted the specifications and objectives of each study, the AI tools used, and the results obtained. From the 266 initial records identified, 25 eligible studies were selected, mainly conducted on human liver tissues. Most of the studies were performed using whole-slide imaging systems for imaging acquisition and applying different machine learning and deep learning methods for image pre-processing, segmentation, feature extractions, and classification. Of note, most of the studies selected demonstrated good performance as classifiers of liver histological images compared to pathologist annotations. Promising results to date bode well for the not-too-distant inclusion of these techniques in clinical practice.

show abstract

“…Similarly in breast cancer, Krithiga et al examined studies where image analysis techniques were used to detect, segment and classify disease, with reported accuracies ranging from 77 to 98% 17 . Other reviews have examined applications in liver pathology, skin pathology and kidney pathology with evidence of high diagnostic accuracy from some AI models 18 – 20 . Additionally, Rodriguez et al performed a broader review of AI applied to WSIs and identified 26 studies for inclusion with a focus on slide level diagnosis 21 .…”

Section: Introductionmentioning

confidence: 99%

Artificial intelligence in digital pathology: a systematic review and meta-analysis of diagnostic test accuracy

McGenity,

Clarke,

Jennings

et al. 2024

npj Digit. Med.

View full text Add to dashboard Cite

Ensuring diagnostic performance of artificial intelligence (AI) before introduction into clinical practice is essential. Growing numbers of studies using AI for digital pathology have been reported over recent years. The aim of this work is to examine the diagnostic accuracy of AI in digital pathology images for any disease. This systematic review and meta-analysis included diagnostic accuracy studies using any type of AI applied to whole slide images (WSIs) for any disease. The reference standard was diagnosis by histopathological assessment and/or immunohistochemistry. Searches were conducted in PubMed, EMBASE and CENTRAL in June 2022. Risk of bias and concerns of applicability were assessed using the QUADAS-2 tool. Data extraction was conducted by two investigators and meta-analysis was performed using a bivariate random effects model, with additional subgroup analyses also performed. Of 2976 identified studies, 100 were included in the review and 48 in the meta-analysis. Studies were from a range of countries, including over 152,000 whole slide images (WSIs), representing many diseases. These studies reported a mean sensitivity of 96.3% (CI 94.1–97.7) and mean specificity of 93.3% (CI 90.5–95.4). There was heterogeneity in study design and 99% of studies identified for inclusion had at least one area at high or unclear risk of bias or applicability concerns. Details on selection of cases, division of model development and validation data and raw performance data were frequently ambiguous or missing. AI is reported as having high diagnostic accuracy in the reported areas but requires more rigorous evaluation of its performance.

show abstract

Artificial Intelligence-Based Opportunities in Liver Pathology—A Systematic Review

Cited by 9 publications

References 74 publications

Prognostic Factors after Hepatectomy for Hepatocellular Carcinoma—The Importance of Pathological Immunophenotyping, the Steatohepatitic Subtype and the Impact of the Hepatic Pedicle Clamping

Prognostic Factors after Hepatectomy for Hepatocellular Carcinoma—The Importance of Pathological Immunophenotyping, the Steatohepatitic Subtype and the Impact of the Hepatic Pedicle Clamping

The Use of Artificial Intelligence in the Liver Histopathology Field: A Systematic Review

Artificial intelligence in digital pathology: a systematic review and meta-analysis of diagnostic test accuracy

Contact Info

Product

Resources

About