The Pathology AI Library

The library

Browse

Filter by what you actually have to work with. Every combination is a shareable link β€” send a colleague the exact view you are looking at.

125 entries

Differential diagnosis outlines organised by organ system and bench.

  • Free tier
  • Runs on any laptop
  • Stays on your machine

Pathology image-caption dataset drawn from textbooks and articles.

  • Free
  • Runs on any laptop
  • Stays on your machine

ArteraAI Breast

FDA 510(k)

Prognostic AI test using digital pathology images for breast cancer.

  • Paid

ArteraAI Prostate

FDA De Novo

Answers a different question from a detection tool: not 'is there cancer' but 'how is this likely to behave'.

  • Paid

A no-frills slide viewer. Opens big scanner files quickly and lets you draw regions.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 682β˜…

Research-focused. Useful if you want the academic papers rather than usable tools.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 12β˜…

Lymph node metastasis detection challenge dataset.

  • Free
  • Runs on any laptop
  • Stays on your machine

The reason synoptic reporting exists: every element you fill in becomes a field a registry or a trial can actually use.

  • Free
  • Runs on any laptop
  • Stays on your machine

Usable from QuPath via an extension if you would rather not touch Python.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 2,293β˜…

Vision-transformer nucleus segmentation and classification.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 391β˜…

Checklist for artificial intelligence in medical imaging.

  • Free
  • Runs on any laptop
  • Stays on your machine

Attention-based multiple-instance learning for weakly supervised whole-slide classification.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 1,720β˜…

Worth knowing about because it takes seriously a problem most benchmarks ignore: whether a model can score well without actually looking at the slide.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 1β˜…

Vision-language foundation model for pathology image-text tasks.

  • Free for academic use
  • Needs a consumer GPU
  • Stays on your machine
  • 520β˜…

Trial reporting and protocol guidelines for AI interventions.

  • Free
  • Runs on any laptop
  • Stays on your machine

Unified multimodal model spanning patch and whole-slide analysis.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 31β˜…

Agent that navigates a slide the way a pathologist moves a microscope.

  • Free
  • Needs a workstation GPU
  • Stays on your machine

Runs on a server your institution controls, so several people can annotate the same slides in a browser. Needs IT support to set up.

  • Free
  • Needs a workstation GPU
  • Stays on your machine

The stage most pathology AI has never been through. Worth asking a vendor whether their product has.

  • Free
  • Runs on any laptop
  • Stays on your machine

Uses spatial context from surrounding tissue to improve expression prediction.

  • Free
  • Needs a consumer GPU
  • Stays on your machine

The reason a slide from one scanner can be opened by another vendor's software. Ask about DICOM support before buying anything.

  • Free
  • Runs on any laptop
  • Stays on your machine

Ask an AI vendor how their results are stored. If the answer is a bespoke database, your results are locked in.

  • Free
  • Runs on any laptop
  • Stays on your machine

A good starting point if your department is considering going digital and wants peer experience rather than vendor material.

  • Free
  • Runs on any laptop
  • Stays on your machine

Server platform for managing, annotating and analysing large slide collections.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 139β˜…

Instruction tuning grounded in symbolic clinical reasoning.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 4β˜…

Preference alignment to reduce hallucination in pathology VLMs.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 2β˜…

'CE marked' under the old directive and under IVDR are not the same claim. Check which one a vendor means.

  • Free
  • Runs on any laptop
  • Stays on your machine

Single-cell resolution spatial gene expression from histology.

  • Free
  • Needs a consumer GPU
  • Stays on your machine

Run this before trusting any result computed on a batch of scanned slides. Most 'model failure' turns out to be slide quality.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 113β˜…

A realistic entry point for a department that cannot justify a bulk scanner. Small enough to sit on a desk.

  • Paid
  • Runs on any laptop
  • Stays on your machine

Practical rather than clever β€” the sort of thing that saves ten minutes a day.

  • Free
  • Runs on any laptop
  • Stays on your machine

Paired histology and spatial transcriptomics dataset and benchmark.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 425β˜…

Transformer and graph approaches to expression prediction from histology.

  • Free
  • Needs a consumer GPU
  • Stays on your machine

Python toolkit for histology image analysis and feature extraction.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 480β˜…

Checks a batch of scanned slides for problems like blur or pen marks before anyone spends time on them.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 329β˜…

If a vendor cannot tell you how their output reaches your LIS, this is the vocabulary in which to ask the question.

  • Free
  • Runs on any laptop
  • Stays on your machine

Simultaneous nucleus segmentation and classification in histology.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 734β˜…

Faster successor to HoVer-Net for nucleus segmentation and classification.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 51β˜…

Runs after you have signed out, and raises a flag if it disagrees. Designed as a safety net, not a first opinion.

  • Paid

Topography and morphology coding for cancer registries.

  • Free
  • Runs on any laptop
  • Stays on your machine

International standard for medical device software lifecycle processes.

  • Paid
  • Runs on any laptop
  • Stays on your machine

Embedding-based cell segmentation that runs acceptably on a CPU.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 233β˜…

Risk management standard for medical devices.

  • Paid
  • Runs on any laptop
  • Stays on your machine

The idea worth knowing: morphology carries enough signal to interpolate molecular data between measured points β€” but it is inference, not measurement.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 169β˜…

Read this before trusting any performance figure. Most published accuracy is measured under conditions that do not resemble a working laboratory.

  • Free
  • Runs on any laptop
  • Stays on your machine

The scanner itself, not an AI tool β€” it produces the digital slides everything else works on.

  • Paid

Biomedical vision-language assistant trained rapidly from PubMed figures.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 2,225β˜…

Universal codes for laboratory observations and report sections.

  • Free
  • Runs on any laptop
  • Stays on your machine

Mitotic count drives grading in several tumours, and it is one of the least reproducible things pathologists do β€” which is exactly why this benchmark measures generalisation rather than headline accuracy.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 14β˜…

Vision-language foundation model for precision oncology.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 241β˜…

A good starting dataset if you are learning β€” small enough to work with on a laptop.

  • Free
  • Runs on any laptop
  • Stays on your machine

Matters if slides are going to the cloud β€” proprietary scanner formats were not designed for that.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 168β˜…

Image data management server for microscopy, including whole-slide images.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 219β˜…

The genuinely low-cost option. You print the body, add optics and a Raspberry Pi, and get automated scanning for telepathology or teaching.

  • Free
  • Runs on any laptop
  • Stays on your machine

C library with Python bindings for reading proprietary whole-slide formats.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 510β˜…

Low-cost high-throughput whole slide imaging from open-source hardware.

  • Free
  • Runs on any laptop
  • Stays on your machine

Notable as an early example of pathology analysis reachable from a general assistant rather than a dedicated application.

  • Paid
  • Cloud service only
  • Uploads your data

Pathology language-vision assistant built on PLIP with two-stage training.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 65β˜…

Paige FullFocus

FDA 510(k)

Whole-slide image viewer for primary diagnosis.

  • Paid

Paige Prostate

FDA De Novo

An assistive second look at prostate biopsies. It flags suspicious areas β€” it does not make the diagnosis, and you remain responsible for the report.

  • Paid

Prostate biopsy Gleason grading challenge dataset.

  • Free
  • Runs on any laptop
  • Stays on your machine

Pan-cancer nucleus instance segmentation and classification dataset.

  • Free
  • Runs on any laptop
  • Stays on your machine

Training-free LLM agent that explores whole slides and reasons over findings.

  • Free
  • Cloud service only
  • Uploads your data
  • 44β˜…

PathAI AISight Dx

FDA 510(k)

Digital pathology image management and viewing platform.

  • Paid

Generative pathology assistant with a domain-adapted CLIP encoder.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 136β˜…

Pathology image-caption corpus released with PathAsst.

  • Free
  • Runs on any laptop
  • Stays on your machine

Multimodal generative copilot for human pathology.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 131β˜…

Practice cases and self-testing for pathologists and trainees.

  • Free tier
  • Runs on any laptop
  • Stays on your machine

Fine-grained language-image pretraining grounding captions to regions.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 8β˜…

The most useful part is the structure, not the individual prompts: it shows how to build a prompt that can be evaluated rather than just admired.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 1β˜…

Pathology instruction-tuning dataset released with PathAsst.

  • Free
  • Runs on any laptop
  • Stays on your machine

Python library for computational pathology preprocessing and modelling.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 460β˜…

Expert-validated multimodal pathology understanding benchmark.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 39β˜…

The interesting idea here is grounding: answers point back to a textbook page instead of being asserted from nowhere.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 65β˜…

Standardised evaluation suite for pathology slide-level foundation models.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 152β˜…

Reinforcement-learning pathology reasoner trained on chain-of-thought data.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 101β˜…

Not an AI tool, but the reference most pathologists already use β€” included because AI outputs should always be checked against a trusted source.

  • Free
  • Runs on any laptop
  • Stays on your machine

Many pathologists already know the teaching platform; the clinical viewer is a separate, cleared product.

  • Paid

Knowledge-guided structured reasoning for whole-slide pathology.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 28β˜…

GRPO-trained reasoning model for pathology visual questions.

  • Free
  • Needs a consumer GPU
  • Stays on your machine

Pathology visual question answering benchmark.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 15β˜…

Notable because it is genuinely open β€” no access request, no non-commercial restriction.

  • Free
  • Needs a consumer GPU
  • Stays on your machine

Pathology CLIP model trained on image-text pairs mined from social media.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 382β˜…

Slide-level multimodal model generating reports from tile embeddings.

  • Free
  • Needs a consumer GPU
  • Stays on your machine

Slide-level foundation model with a tile encoder and long-context aggregator.

  • Free for academic use
  • Needs a workstation GPU
  • Stays on your machine
  • 626β˜…

Histopathology image-text pairs mined from educational videos.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 186β˜…

Instruction-tuned pathology assistant trained from educational video narration.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 89β˜…

Free, installs on an ordinary laptop, and does not upload your slides anywhere. The usual first stop if you have never used digital pathology software.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 1,411β˜…

Another automatic cell-detection option inside QuPath. Needs a one-time Python setup.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 94β˜…

Newer segmentation option that runs without a graphics card.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 29β˜…

Adds automatic nucleus detection to QuPath. Runs on a normal laptop, no coding needed.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 48β˜…

Worth knowing about if your hospital already uses Sectra for radiology β€” the pathology module reuses that infrastructure.

  • Paid

Its practical value in pathology today is speeding up manual annotation, not replacing it.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 54,633β˜…

The most important cautionary result in computational pathology. It explains why a model reporting excellent accuracy on TCGA may fail completely on your slides.

  • Free
  • Runs on any laptop
  • Stays on your machine

Vision-language assistant operating on whole slides rather than patches.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 126β˜…

End-to-end deep learning pipeline for whole-slide images.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 369β˜…

Slide-level prompt learning for few-shot MIL.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 4β˜…

Mixture-of-experts reasoner unifying ROI and whole-slide tasks.

  • Free
  • Needs a workstation GPU
  • Stays on your machine

The difference between a report a computer can count and one it can only store.

  • Free for academic use
  • Runs on any laptop
  • Stays on your machine

Early demonstration that gene expression can be predicted from H&E alone.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 148β˜…

Reporting guideline for AI diagnostic accuracy studies.

  • Free
  • Runs on any laptop
  • Stays on your machine

Star-convex polygon nucleus detection.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 1,247β˜…

Pan-cancer whole-slide and molecular data archive.

  • Free
  • Runs on any laptop
  • Stays on your machine

Python toolbox for computational pathology pipelines.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 542β˜…

Multimodal whole-slide foundation model producing slide-level embeddings.

  • Free for academic use
  • Needs a workstation GPU
  • Stays on your machine
  • 360β˜…

Stain normalisation as a differentiable, GPU-capable operation.

  • Free
  • Runs on any laptop
  • Stays on your machine
  • 188β˜…

Toolkit for whole-slide preprocessing and foundation-model feature extraction.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 603β˜…

A quick way to judge a paper: if it does not report these items, you cannot tell whether the model would work in your laboratory.

  • Free
  • Runs on any laptop
  • Stays on your machine

UNI

Self-supervised vision foundation model for pathology tile encoding.

  • Free for academic use
  • Needs a consumer GPU
  • Stays on your machine
  • 763β˜…

Dual-scale vision-language MIL for whole-slide classification.

  • Free
  • Needs a workstation GPU
  • Stays on your machine
  • 103β˜…

Large pathology vision foundation model trained on a very large slide corpus.

  • Free for academic use
  • Needs a workstation GPU
  • Stays on your machine

Not AI, but included deliberately β€” AI output should always be checked against a trusted image reference.

  • Free
  • Runs on any laptop
  • Stays on your machine

Whole-slide VQA benchmark organised around morphological description.

  • Free
  • Runs on any laptop
  • Stays on your machine

Multimodal model for whole-slide images with a morphology-aware benchmark.

  • Free
  • Needs a workstation GPU
  • Stays on your machine

Runs pretrained patch classification models across whole slides, with a QuPath front end.

  • Free
  • Needs a consumer GPU
  • Stays on your machine
  • 86β˜