Comprehensive Evaluation and Fine-Tuning of Foundational Cell Nuclei Segmentation Models in Renal Pathology

Key points

• In this work, we extended a human-in-the-loop framework by combining 5,901 foundation-model-generated pseudo-labels from well-segmented cases (Easy), 860 newly expert-annotated unresolved challenging cases (Medium), and 198 expert-annotated consensus failure cases (Hard).1
• These annotations, spanning different levels of segmentation difficulty, enabled the systematic evaluation of seven single-source and mixed-source fine-tuning strategies across nine cell segmentation model configurations.1
• LSP-DETR achieved the highest F1 score of 0.8725 with Hard-only fine-tuning, while StarDist showed the largest improvement, increasing from 0.7380 to 0.8332 with Medium-only fine-tuning.1
• These findings show that annotations spanning multiple difficulty levels support effective model adaptation, although the optimal annotation composition remains model dependent.1

Sources (1)

• [1]Comprehensive Evaluation and Fine-Tuning of Foundational Cell Nuclei Segmentation Models in Renal Pathology
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 04:05 AM

  Accurate nuclei instance segmentation is essential for quantitative renal pathology, yet general-purpose models often struggle with low contrast, dense nuclei, complex morphology, and strong background staining.

  In this work, we extended a human-in-the-loop framework by combining 5,901 foundation-model-generated pseudo-labels from well-segmented cases (Easy), 860 newly expert-annotated unresolved challenging cases (Medium), and 198 expert-annotated consensus failure cases (Hard).

Extractive summary: sentences quoted from the sources.