New framework improves cell nuclei segmentation models for renal pathology

Researchers have developed a novel human-in-the-loop framework to enhance cell nuclei segmentation models for renal pathology. This framework combines pseudo-labels from existing models with expert annotations of varying difficulty levels. The study systematically evaluated nine cell segmentation model configurations using seven fine-tuning strategies, demonstrating that incorporating data across different difficulty levels significantly improves model performance. LSP-DETR achieved the highest F1 score with hard-case fine-tuning, while StarDist showed the most substantial improvement with medium-case fine-tuning. AI

IMPACT Enhances the accuracy of AI models for critical tasks in medical pathology, potentially improving diagnostic capabilities.

RANK_REASON The cluster contains a research paper detailing a new framework and evaluation of models for a specific scientific task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• LSP-DETR
• StarDist

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →