Posts by Collection

Big Brother Project (2017 - 2018)

Published: October 01, 2017

Real-time face detection, tracking, and identification from video.

Deep Football Analysis — Player & Referee Detection (2018)

Published: January 01, 2018

Detecting and tracking players and referees in football match videos with possession metrics.

Kaggle Competition: Pneumothorax Segmentation — Top 7% Bronze Medal (2019)

Published: January 01, 2019

U-Net with EfficientNet-B4 for pneumothorax segmentation on chest X-rays; ranked 98/1475.

Predicting Poverty from Satellite Imagery (2020)

Published: January 02, 2019

Predicting wealth levels across African regions from satellite imagery and nighttime light data.

Implementation of A-LAMP: Adaptive Layout-Aware Multi-Patch Deep CNN for Photo Aesthetic Assessment (2021)

Published: January 01, 2021

Implementation of A-LAMP for photo aesthetic quality assessment on the AVA dataset.

EFTD: Efficient Fine-Tuning of DINOv3 for Atypical Mitotic Figure Classification (2025)

Published: January 01, 2025

Winning Solution of Task 2 MIDOG 2025 (MICCAI 2025) - We finetune DINOv3-H+ model with LoRA for atypical mitotic figure classification across scanner and tissue domains

MIPHEI-ViT: Multiplex Immunofluorescence Prediction from H&E Images (2025)

Published: January 02, 2025

Predicting multiplex immunofluorescence marker expression from H&E whole slide images using ViT foundation models.

Automatic grading of cervical biopsies by combining full and self-supervision

Published in ECCV 2022 Workshop on AI for Medicine and Intelligent Analysis (AIMIA), 2022

Combining full supervision and self-supervision for automatic grading of cervical biopsies from histological images.

Recommended citation: Lubrano di Scandalea Mélanie, Lazard Tristan, Balezo Guillaume, Berlemont Sylvain, Walter Thomas (2022). Automatic grading of cervical biopsies by combining full and self-supervision. ECCV 2022 AIMIA Workshop.
Download Paper

Interpretable HER2 scoring by evaluating clinical guidelines through a weakly supervised, constrained deep learning approach

Published in Computerized Medical Imaging and Graphics, 2023

Weakly supervised, constrained deep learning approach for interpretable HER2 scoring aligned with clinical guidelines.

Recommended citation: Pham Manh-Dan, Balezo Guillaume, Tilmant Cyprien, Petit Stephanie, Salmon Isabelle, Hadj Sama Ben, Fick Rutger HJ (2023). Interpretable HER2 scoring by evaluating clinical guidelines through a weakly supervised, constrained deep learning approach. Computerized Medical Imaging and Graphics.
Download Paper

Enabling collagen quantification on HE-stained slides through stain deconvolution and restained HE-HES

Published in IEEE International Symposium on Biomedical Imaging (ISBI 2023), 2023

Deep learning method for collagen quantification on H&E slides via stain deconvolution using HE-HES restained datasets.

Recommended citation: Balezo Guillaume, Bertram Christof A., Tilmant Cyprien, Petit Stéphanie, Ben Hadj Saima, Fick Rutger H.J. (2023). Enabling collagen quantification on HE-stained slides through stain deconvolution and restained HE-HES. ISBI 2023.
Download Paper

Robust Pan-Cancer Mitotic Figure Detection with YOLOv12

Published in MICCAI 2025 - MIDOG Workshop, 2025

YOLOv12-based mitotic figure detection across domains — 2nd place (Task 1), MIDOG 2025.

Recommended citation: Bourgade Raphaël, Balezo Guillaume, Feki Hana, Monier Lily, Blons Matthieu, Blondel Alice, Loussouarn Delphine, Vincent-Salomon Anne, Walter Thomas (2025). Robust Pan-Cancer Mitotic Figure Detection with YOLOv12. MIDOG 2025 Challenge (MICCAI).
Download Paper

Efficient Fine-Tuning of DINOv3 Pretrained on Natural Images for Atypical Mitotic Figure Classification in MIDOG 2025

Published in MICCAI 2025 - MIDOG Workshop, 2025

LoRA fine-tuning of DINOv3 for atypical mitotic figure classification across domains — 1st place (Task 2), MIDOG 2025.

Recommended citation: Balezo Guillaume, Bourgade Raphaël, Walter Thomas (2025). Efficient Fine-Tuning of DINOv3 Pretrained on Natural Images for Atypical Mitotic Figure Classification in MIDOG 2025. MIDOG 2025 Challenge (MICCAI).
Download Paper

MIPHEI-ViT: Multiplex Immunofluorescence Prediction from H&E Images using ViT Foundation Models

Published in Computers in Biology and Medicine, 2026

Predicting multiplex immunofluorescence marker expression directly from H&E whole slide images using ViT foundation models.

Recommended citation: Balezo Guillaume, Trullo Roger, Pla Planas Albert, Decencière Etienne, Walter Thomas (2026). MIPHEI-ViT: Multiplex immunofluorescence prediction from H&E images using ViT foundation models. Computers in Biology and Medicine, 206, 111564.
Download Paper

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Teaching experience 2

Workshop, University 1, Department, 2015

This is a description of a teaching experience. You can use markdown like any other post.