Open, reproducible computational pathology research
The Computational Pathology Lab develops open-source methods, software, and reproducible workflows for turning complex pathology and spatial imaging data into quantitative biological insight.
Our current public-facing research focus includes vascular spatial analysis, digital pathology, imaging mass cytometry (IMC), tumor microenvironment analysis, and translational cancer research.
🌐 Website: https://computationalpathologylab.github.io/
🐙 GitHub: https://github.com/ComputationalPathologyLab
📍 Italy — IRCCS Humanitas Research Hospital, Pathology Service
We develop computational approaches for vessel segmentation, morphometric quantification, spatial vessel metrics, and downstream biological analysis in histological images.
A central project is VeSpA (Vessel Spatial Analysis), which integrates vessel segmentation and measurement workflows into the QuPath environment.
Our tooling supports analysis of whole-slide and high-resolution pathology images, with an emphasis on workflows that are practical, reproducible, and extensible within QuPath.
We build workflows for Imaging Mass Cytometry (IMC) data processing, including channel preparation, segmentation, cell-level feature extraction, phenotyping, and spatial analysis.
We use modern research-software practices and workflow technologies such as Python, R, Nextflow, containers, and HPC/Slurm to make computational pathology analyses easier to reproduce and scale.
Our repository ecosystem also explores AI-assisted analysis and research tooling, including structured research templates and agentic interfaces for IMC analysis.
| Project | What it provides |
|---|---|
| VeSpA | QuPath extension for annotation-based vessel segmentation, lumen-aware processing, morphometric measurements, and reconstruction in QuPath. |
| IMC-Data-Analysis | R/Python workflows for Imaging Mass Cytometry, including cell phenotyping and spatial biology analysis. |
| imc-nextflow-pipeline | Reproducible Nextflow pipeline for IMC channel stacking, panel generation, Steinbock-based segmentation, feature extraction, and graph export. |
| qupath-vessel-segmentation-plugin | QuPath plugin for vessel segmentation and annotation generation. |
| VeSpA_benchmarking | Benchmarking of VeSpA against ground truth and other available tools. |
| AI_pathology_research_kit | Reusable structure and guidance for rigorous, traceable AI-assisted pathology research. |
| agentimc | IMC analysis interface and copilot tooling covering segmentation, quantification, phenotyping, and spatial analysis. |
| hpc-slurm-job-generator | GUI for generating self-contained Slurm jobs for IMC/Steinbock workloads on HPC systems. |
Our repositories span Python, R, Java/QuPath extensions, Nextflow, Jupyter notebooks, and lightweight web tooling.
We aim to make research workflows transparent and reusable by combining:
- version-controlled analysis code
- explicit input/output conventions
- containerized or environment-aware execution where appropriate
- workflow orchestration for multi-step analyses
- machine-readable configurations and metadata
- benchmark and validation workflows
- documentation designed for both researchers and developers
Where human or patient-derived data are involved, raw data should remain under the appropriate institutional governance and should not be committed to public repositories unless explicitly permitted.
The lab also maintains resources covering foundational skills in Python, GitHub, Linux shell, and high-performance computing.
For the broader research context and software catalogue, visit the lab website.
We are interested in collaborations around:
- computational and digital pathology
- vascular biology and spatial tissue analysis
- imaging mass cytometry
- tumor microenvironment research
- pathology image analysis and segmentation
- reproducible biomedical workflows
- open-source research software
For research collaborations, software questions, or technical discussions, contact the lab at computationalpathologylab@humanitas.it.
The Computational Pathology Lab is based within the Pathology Service at IRCCS Humanitas Research Hospital in the Milan area, Italy.
The lab’s public research identity emphasizes reproducible computational pathology for vascular biology, tumor microenvironment analysis, imaging mass cytometry, and translational cancer research.
- 🌐 Lab website
- 🐙 GitHub organization
- 🧑🔬 Lab members
- 🔬 Projects
- 💻 Software
- 📄 Publications
- 📚 Resources
- ✉️ Contact
Licensing is defined at the individual repository level. Please check the LICENSE file in each project before reusing code, models, datasets, or other project materials.
- Ricerca Finalizzata 2021 by Italian Ministry of Health—Giovani Ricercatori (GR)— “Change promoting”, GR-2021-12373209
- AIRC (Associazione Italiana Ricerca contro il Cancro) My First Grant AIRC 2025, 32683
- BANDO DI RICERCA COLLABORATIVA UNDER 40 2025 Fondazione Regionale per la Ricerca Biomedica (FRRB) Regione Lombardia, 022024R0052
Computational Pathology Lab · Open science · Reproducible research · Digital pathology
