Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 18, 2025Open Access

Automated Segmentation of Hepatic Vessels and Lobules in Whole-Slide Images Using U-Net Models

View Full Paper
Ask AI
Bookmark
Share

Authors

MBMehul BafnaMKMatthias KönigUniversity of StuttgartSSSylvia SaalfeldFachhochschule Kiel

Discussion

Loading...

Member takes

Overview

Deep-learning approach enhances segmentation accuracy of hepatic vessels and lobules, suggesting its potential for advanced histological analysis.

Key Points

  • The automated segmentation achieved Dice scores of 0.960 for lobules and 0.801 for central veins, demonstrating high accuracy.
  • Using a weight-boosted nnU-Net framework, the model improved detection rates for smaller vascular structures and class imbalances.
  • Adaptable geometric data transformations enhanced the model's robustness in segmenting complex hepatic structures.
  • Evaluations confirmed generalizability across multiple test datasets, supporting the broader application in computational liver histology.

Cite This Study

Bafna et al. (2025) studied this question.

synapsesocial.com/papers/68d463e231b076d99fa62fa8https://doi.org/10.1101/2025.09.08.674181
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Automatic liver segmentation using U-Net deep learning architecture for additive manufacturing2024 · 11 citations
  2. 2Improving automatic segmentation of liver tumor images using a deep learning model2024 · 15 citations
  3. 3Automatic hepatic tumor segmentation in intra-operative ultrasound: a supervised deep-learning approach2024 · 2 citations
  4. 4Deep Learning-Based Liver Vessel Segmentation2024 · 1 citations
  5. 5Medical Image Analysis for Liver Tumor Localization and Segmentation using Deep Learning2024