Loading
Scanning microscope slides and letting software measure things a pathologist cannot see, including predictions of who will benefit from a treatment.
Whole-slide imaging, scanning glass slides into gigapixel digital images, is now routine in large pathology laboratories. FDA-cleared tools: Paige Prostate (detection), ArteraAI Prostate (2025, first prognostic and predictive AI) and ArteraAI Breast (May 2026, risk stratification in early HR+/HER2- breast cancer). Foundation models (Virchow, UNI, CONCH, Prov-GigaPath) predict molecular status (MSI, HRD, HER2) from H&E alone.
Gigapixel whole-slide images; tile-level self-supervised encoders aggregated to slide-level predictions.
Dependencies are what this technology cannot be delivered without: manufacturing steps, instruments, software, upstream methods. See its full chain on the map.
An FDA-cleared AI test (May 2026) that reads breast cancer slides to estimate recurrence risk in early hormone-positive disease.
The first AI tool cleared by the FDA to predict both prognosis and treatment benefit from a routine biopsy slide, in prostate cancer.
The first AI for reading biopsy slides authorised by the FDA, which points pathologists to prostate cancer they might otherwise miss.
The evidence behind the stage I de-escalation statement on the triple-negative page: an ordinary haematoxylin and eosin slide identifies a fifth of patients whose outcome without chemotherapy matches treated cohorts. A prospective trial of chemotherapy omission is the missing step.
HER2-low is a drug eligibility label, not a biological subtype, in triple-negative disease; because a third of patients qualify for trastuzumab deruxtecan on a score pathologists disagree about, re-scoring and digital assistance for HER2 0 versus 1+ is a practical gap.
It explains why a PD-L1 score alone predicts imperfectly: PD-L1 on stromal cells with excluded T cells is a poor-outcome pattern, and B7-H4 marks the cold tumours now being targeted by antibody-drug conjugates.
Triple-negative is a laboratory definition; this guideline wrote the oestrogen and progesterone half of it, and the 1 to 10 percent low-positive band it created is still argued over.
Query for this technology: (TITLE:"digital pathology" OR ABSTRACT:"digital pathology" OR TITLE:"computational pathology" OR ABSTRACT:"computational pathology" OR TITLE:"whole slide image" OR ABSTRACT:"whole slide image") AND (deep learning OR artificial intelligence). Results are unfiltered search hits about Digital pathology & AI, not a curated reading list.
Shares HistoWiz, Analytical cellular pathology (Amsterdam), Digistain, H&E staining (haematoxylin and eosin).
Shares A dedicated fund for randomised trials of cancer AI with patient outcomes, Continuous prospective validation for every oncology AI tool after deployment, Oncology workforce, Federated training of pathology and radiology models across hospitals.
Shares Nucleai, Map which tumour clones sit next to which immune cells before choosing therapy, Standards for spatial and multiplex tissue biomarkers before they reach the clinic, Clear the suppressive neutrophils out of pancreatic tumours first.
Shares Calibrated reference slides so every lab scores HER2-low the same way, Clinical, pathological, and PAM50 gene expression features of HER2-low breast cancer, Reflex re-scoring of HER2 0 versus 1+ with digital assistance so every eligible triple-negative patient reaches trastuzumab deruxtecan, AI quantification of HER2-low and HER2-ultralow.
Shares Calibrated reference slides so every lab scores HER2-low the same way, Clinical, pathological, and PAM50 gene expression features of HER2-low breast cancer, Reflex re-scoring of HER2 0 versus 1+ with digital assistance so every eligible triple-negative patient reaches trastuzumab deruxtecan, AI quantification of HER2-low and HER2-ultralow.
Shares Ataraxis AI, Ibex Medical Analytics, Valar Labs, Imagene AI.
Open-source projects that implement or serve this technology, from OnCo's own catalogue: licence and last activity as the repository reported them on the day of the fetch. Listing is not endorsement; check the licence before reuse and the validation before clinical use.
Clustering-constrained attention multiple instance learning: the weakly supervised whole-slide classification pipeline from the Mahmood lab that many pathology AI papers build on.
The open bioimage analysis software for whole-slide images: cell detection, classifiers, tumour region annotation and scripting; the everyday tool of computational pathology labs.
Star-convex object detection for nuclei in fluorescence and brightfield images, integrated in QuPath.
Simultaneous nuclear segmentation and classification in histology, the Warwick method many tools reuse.
Radboud's Automated Slide Analysis Platform for viewing and annotating whole-slide images.
The Warwick Tissue Image Analytics centre's end-to-end toolbox for computational pathology, with pretrained models.
Dartmouth's sliding-window whole-slide classification code.
Image analysis algorithms for histopathology: colour deconvolution, nuclear segmentation, features.
Commercial and regulated products that serve this technology. Each card says what is behind it: a regulator's database, the literature, a public body's list, or only the company's own words. Listing is not endorsement, and a clearance is a regulatory fact, not a clinical one.
A high-throughput whole-slide scanner and its viewing software, sold for primary diagnosis in histopathology.
A whole-slide scanning and viewing system for histopathology, the first such system cleared in the United States for primary diagnosis.
A platform for training and running deep-learning models on whole-slide images, sold to laboratories that want to build their own scoring algorithms.
A digital pathology image management platform that laboratories run as the layer between scanners and the algorithms they buy.
A digital pathology case management and image analysis platform used in diagnostic laboratories and in research.
Slide viewing and case management inside the same workstation a hospital already uses for radiology, which is how several health systems went digital in pathology.
Open-source software, hardware and data projects catalogued by a third party, the Open Medical Registry, that bear on this technology. Listing is not endorsement; check each project's own licence and validation before clinical use.
CellViT: Vision Transformers for Precise Cell Segmentation and Classification
Open source tools for computational pathology - Nature BME
DeepSpot: Deep learning model for predicting spatial transcriptomics from H&E histopathology images.
Machine learning tool for analysis of lung adenocarcinoma tumors
Histomic Prognostic Signature (HiPS): A population-level computational histologic signature for invasive breast cancer prognosis
A panoptic segmentation approach for tumor-infiltrating lymphocyte assessment: development of the MuTILs model and PanopTILs dataset.
Tools for computational pathology
CNN ensemble for prostate cancer Gleason grading
QuPath - Open-source bioimage analysis for research
Solid Tumor Associative Modeling in Pathology
SurvivMIL: A multimodal, Multiple Instance Learning pipeline for survival outcome of Neuroblastoma Patients
Computational Pathology Toolbox developed by TIA Centre, University of Warwick.
From the Open Medical Registry (openmedical.sh), an MIT-licensed catalogue of open-source medicine. Blurbs are one line from each registry record; every project keeps its own licence.
