Semantic Segmentation of iPS Cells: Case Study on Model Complexity in Biomedical Imaging

TL;DR

This study demonstrates that a carefully configured DeepLabv3 model can outperform larger foundation models in segmenting iPS cell images, emphasizing the effectiveness of tailored, simpler models for specialized biomedical tasks.

Contribution

The paper shows that optimized, domain-specific configurations of DeepLabv3 can achieve superior performance over larger models like SAM2 in biomedical image segmentation.

Findings

DeepLabv3 outperforms SAM2 and MedSAM2 in iPS cell segmentation.

Model complexity does not always correlate with better performance in specialized tasks.

Open-source implementation supports small datasets and domain-specific encoding.

Abstract

Medical image segmentation requires not only accuracy but also robustness under challenging imaging conditions. In this study, we show that a carefully configured DeepLabv3 model can achieve high performance in segmenting induced pluripotent stem (iPS) cell colonies, and, under our experimental conditions, outperforms large-scale foundation models such as SAM2 and its medical variant MedSAM2 without structural modifications. These results suggest that, for specialized tasks characterized by subtle, low-contrast boundaries, increased model complexity does not necessarily translate to better performance. Our work revisits the assumption that ever-larger and more generalized architectures are always preferable, and provides evidence that appropriately adapted, simpler models may offer strong accuracy and practical reliability in domain-specific biomedical applications. We also offer an open-source implementation that includes strategies for small datasets and domain-specific encoding, with the aim of supporting further advances in semantic segmentation for regenerative medicine and related fields.