EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Histological AI‐Assisted Analysis Outperforms Pathologist Assessment in Predicting Disease Outcomes in PSC Patients

Anna M Salonen iD, Iiris Nyholm iD, Linn Tähti, Matilda Sinkko, Darshan Kumar, Martti Färkkilä, Johanna Arola, Mikko P Pakarinen, Sonja Boyd, Nelli Sjöblom

Hepatology Research · 29.07.2026

DOI: 10.1111/hepr.70254 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Hepatology Research, online erschienen am 29.07.2026. Verfasst von Anna M Salonen und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pathology, Helsinki University and Helsinki University Hospital (HUH), HUS Diagnostic Center, Helsinki, Finland. Die Arbeit nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Sigrid Juséliuksen Säätiö, Lastentautien Tutkimussäätiö, Syöpäsäätiö.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aims

Liver biopsy has shown prognostic significance in primary sclerosing cholangitis (PSC). However, histology is typically assessed manually based on coarse grading scales, making precise quantification difficult. We developed a neural network model to address the key histological features of liver tissue, including portal tracts, fibrosis, biliary epithelium, portal inflammation, and vasculature. Methods and Results Our training data included 603 liver sections from patients with PSC ( n = 478), other hepatobiliary diseases ( n = 119), and normal controls ( n = 6), ensuring heterogeneity of the sample material. The model was validated and evaluated for its ability to predict key clinical endpoints in PSC: liver transplantation (LT) due to end‐stage liver disease (ESLD), and a composite endpoint comprising LT, cholangiocarcinoma (CCA), and liver‐related death, allowing evaluation with all relevant clinical outcomes. The model reliably identified the targeted histological features. AI‐derived portal inflammation and fibrosis demonstrated superior discriminatory performance compared with manual scoring, particularly for prediction of LT due to ESLD (AUC 0.81–0.82 vs. 0.86–0.87). AI‐derived blood vessels and biliary epithelium were also prognostic for LT due to ESLD (AUC 0.80, 0.83). In multivariable analysis adjusted for age and sex, AI‐derived portal inflammation remained independently associated with the composite endpoint (HR 2.51, 95% CI 1.04–6.05).

Conclusions

AI‐measured portal inflammation is a robust predictor of clinical endpoints. Additionally, vascular area, fibrosis, and biliary epithelium showed prognostic value. AI‐based measures outperformed manual scores for several parameters. Overall, multiple AI‐derived parameters predicted LT due to ESLD, and comparable trends were observed for the composite endpoint. These findings highlight the importance of liver histology as a surrogate endpoint in patients with PSC.

Schlagwörter

Stichwörter der Autor:innen

artificial intelligenceliver histologymachine learningprimary sclerosing cholangitisprognosis

Autor:innen und Institutionen

  • Department of Pathology, Helsinki University and Helsinki University Hospital (HUH), HUS Diagnostic Center, Helsinki, Finland.
  • Section of Pediatric Surgery, Pediatric Liver and Gut Research Group, New Children's Hospital, University of Helsinki and Helsinki University Hospital, Helsinki, Finland.
  • Aiforia Technologies Plc, Helsinki, Finland.
  • Helsinki University and Helsinki University Hospital, Abdominal Center, Helsinki, Finland.
  • Department of Dermatopathology, University of Helsinki, Helsinki University Hospital (HUH), Helsinki, Finland.

ORCID-Profile: Anna M Salonen, Iiris Nyholm

Förderung

  • Sigrid Juséliuksen Säätiö
  • Lastentautien Tutkimussäätiö
  • Syöpäsäätiö
  • Helsinki University Hospital Fund

Zitieren

APA

Salonen, A. M., Nyholm, I., Tähti, L., Sinkko, M., Kumar, D., Färkkilä, M., Arola, J., Pakarinen, M. P., Boyd, S., & Sjöblom, N. (2026). Histological AI‐Assisted Analysis Outperforms Pathologist Assessment in Predicting Disease Outcomes in PSC Patients. Hepatology Research. https://doi.org/10.1111/hepr.70254

Vancouver

Salonen AM, Nyholm I, Tähti L, Sinkko M, Kumar D, Färkkilä M, et al. Histological AI‐Assisted Analysis Outperforms Pathologist Assessment in Predicting Disease Outcomes in PSC Patients. Hepatology Research. 2026. doi:10.1111/hepr.70254

BibTeX
@article{Salonen2026,
  author = {Salonen, Anna M and Nyholm, Iiris and Tähti, Linn and Sinkko, Matilda and Kumar, Darshan and Färkkilä, Martti and Arola, Johanna and Pakarinen, Mikko P and Boyd, Sonja and Sjöblom, Nelli},
  title = {Histological AI‐Assisted Analysis Outperforms Pathologist Assessment in Predicting Disease Outcomes in PSC Patients},
  journal = {Hepatology Research},
  year = {2026},
  doi = {10.1111/hepr.70254}
}
Kostenlos starten →