# Beyond Classification: Pathology Foundation Models as Detection Encoders for Mitotic Figures

arXiv: https://arxiv.org/abs/2607.28007 · HTML: https://mlsanity.com/paper/2607.28007/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 31 references that are in the map.
- First posted on arXiv 30 Jul 2026.

## Builds on (most cited first, 20 of 20)

- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [An Empirical Study of Training Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.02057/) (2021)
- [MMDetection: Open MMLab Detection Toolbox and Benchmark](https://mlsanity.com/paper/1906.07155/) (2019)
- [Exploring Plain Vision Transformer Backbones for Object Detection](https://mlsanity.com/paper/2203.16527/) (2022)
- [Virchow2: Scaling Self-Supervised Mixed Magnification Models in Pathology](https://mlsanity.com/paper/2408.00738/) (2024)
- [CellViT: Vision Transformers for Precise Cell Segmentation and Classification](https://mlsanity.com/paper/2306.15350/) (2023)
- [A Clinical Benchmark of Public Self-Supervised Pathology Foundation Models](https://mlsanity.com/paper/2407.06508/) (2024)
- [Predicting breast tumor proliferation from whole-slide images: the TUPAC16 challenge](https://mlsanity.com/paper/1807.08284/) (2018)
- [Benchmarking foundation models as feature extractors for weakly-supervised computational pathology](https://mlsanity.com/paper/2408.15823/) (2024)
- [Domain generalization across tumor types, laboratories, and species -- insights from the 2022 edition of the Mitosis Domain Generalization Challenge](https://mlsanity.com/paper/2309.15589/) (2023)
- [CellViT++: Energy-Efficient and Adaptive Cell Segmentation and Classification Using Foundation Models](https://mlsanity.com/paper/2501.05269/) (2025)
- [Are pathologist-defined labels reproducible? Comparison of the TUPAC16 mitotic figure dataset with an alternative set of labels](https://mlsanity.com/paper/2007.05351/) (2020)
- [Benchmarking Deep Learning and Vision Foundation Models for Atypical vs. Normal Mitosis Classification with Cross-Dataset Evaluation](https://mlsanity.com/paper/2506.21444/) (2025)
- [Benchmarking Foundation Models for Mitotic Figure Classification](https://mlsanity.com/paper/2508.04441/) (2025)
- [To What Extent Do Token-Level Representations from Pathology Foundation Models Improve Dense Prediction?](https://mlsanity.com/paper/2602.03887/) (2026)
- [Mitosis Detection in the Wild: Multi-Tumor and Context-Aware Generalization in the MIDOG 2025 Challenge](https://mlsanity.com/paper/2606.07368/) (2026)

## Abstract (arXiv)

Pathology foundation models (FMs) are models trained on vast amounts of typically unlabeled data and have been shown to yield regularized latent spaces that can be used effectively in downstream classification tasks. This is also true for the classification of mitotic figures vs. other cells. However, it is so far unclear if the latent space of current FMs provides features that are discriminant and spatially suitably resolved to also serve as a backbone for dense object detection paradigms. In this work, we investigate this question for common current pathology FMs (UNI, UNI2-h, Virchow, Virchow2, H-optimus-0, H-optimus-1) and compare their performance against a fully end-to-end trained baseline based on a ResNet50 architecture. We combine FM backbones with representatives of single stage, dual stage and self-attention-based detectors (RetinaNet, Faster R-CNN, Deformable DETR respectively) on the multi-domain MIDOG++ dataset, and on the TUPAC16 dataset as an out-of-domain case. We show that the H-optimus-0 and Virchow models yielded competitive performance, indicating that the latent spaces of current FMs, all trained on image-level self-supervision, are suitable for direct mitotic figure detection and may be slightly more robust on our out-of-domain test case. All code is made available publicly at https://github.com/DeepMicroscopy/FM4MFdet.