PathologyGAN: Learning deep representations of cancer tissue
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Histopathological images of tumours contain abundant information about how tumours grow and how they interact with their micro-environment. Better understanding of tissue phenotypes in these images could reveal novel determinants of pathological processes underlying cancer, and in turn improve diagnosis and treatment options. Advances of Deep learning makes it ideal to achieve those goals, however, its application is limited by the cost of high quality labels from patients data. Unsupervised learning, in particular, deep generative models with representation learning properties provides an alternative path to further understand cancer tissue phenotypes, capturing tissue morphologies. In this paper, we develop a framework which allows Generative Adversarial Networks (GANs) to capture key tissue features and uses these characteristics to give structure to its latent space. To this end, we trained our model on two different datasets, an H&E colorectal cancer tissue from the National Center for Tumor diseases (NCT, Germany) and an H&E breast cancer tissue from the Netherlands Cancer Institute (NKI, Netherlands) and Vancouver General Hospital (VGH, Canada). Composed of 86 slide images and 576 tissue micro-arrays (TMAs) respectively. We show that our model generates high quality images, with a Frechet Inception Distance (FID) of 16.65 (breast cancer) and 32.05 (colorectal cancer). We further assess the quality of the images with cancer tissue characteristics (e.g. count of cancer, lymphocytes, or stromal cells), using quantitative information to calculate the FID and showing consistent performance of 9.86. Additionally, the latent space of our model shows an interpretable structure and allows semantic vector operations that translate into tissue feature transformations. Furthermore, ratings from two expert pathologists found no significant difference between our generated tissue images from real ones. The code, generated images, and pretrained model are available at https://github.com/AdalbertoCq/Pathology-GAN
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.003 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it