BreakHis Breast Cancer Classification
Upload a breast histopathology image to predict the breast cancer subtype. Your image must be at 40X magnification, and ideally between 224x224 and 700x460 resolution. Do not otherwise modify your image.
This demo uses a custom-trained DINOv2 foundation model for pathology images called OpenMidnight with a linear classifier for BreakHis breast cancer classification.
Tumor Types:
- Benign tumors: Tubular Adenoma (TA), Fibroadenoma (F)
- Malignant tumors: Mucinous Carcinoma (MC), Ductal Carcinoma (DC)
These 4 classes were selected from the full BreakHis dataset as they have sufficient patient counts (≥7 patients) for robust evaluation. For this particular demo, images must be one of the sample classes - unsupported classes will yield confusing and/or useless results.
This demonstration is for illustrative purposes only and should not be used for diagnostic/clinical purposes.
Gleason Grading
Upload a prostate cancer image to predict the tumor type. Your image must be at 40X magnification, and ideally between 224x224 and 750x750 resolution. Do not otherwise modify your image.
This demo uses a custom-trained DINOv2 foundation model for pathology images called OpenMidnight with a linear classifier for Gleason grading.
Images are classified as benign, Gleason pattern 3, 4 or 5.
For this particular demo, images must be one of the sample classes - unsupported classes will yield confusing and/or useless results.
This demonstration is for illustrative purposes only and should not be used for diagnostic/clinical purposes.
Colorectal Cancer Tissue Classification
Upload a colorectal cancer image to predict the tissue class. Your image must be at 20X magnification, and ideally at 224x224. Do not otherwise modify your image.
This demo uses a custom-trained DINOv2 foundation model for pathology images called OpenMidnight with a linear classifier for colorectal cancer tissue classification.
The tissue classes are: Adipose (ADI), background (BACK), debris (DEB), lymphocytes (LYM), mucus (MUC), smooth muscle (MUS), normal colon mucosa (NORM), cancer-associated stroma (STR) and colorectal adenocarcinoma epithelium (TUM)
For this particular demo, images must be one of the sample classes - unsupported classes will yield confusing and/or useless results.
This demonstration is for illustrative purposes only and should not be used for diagnostic/clinical purposes.
BACH Breast Cancer Classification
Upload a breast cancer image to predict the tumor type. Your image must be at 20X magnification, and ideally between 224x224 and 1536x2048 resolution. Do not otherwise modify your image.
This demo uses a custom-trained DINOv2 foundation model for pathology images called OpenMidnight with a linear classifier for tumor classification.
Images are classified as benign, normal, invasive, in-situ.
For this particular demo, images must be one of the sample classes - unsupported classes will yield confusing and/or useless results.
This demonstration is for illustrative purposes only and should not be used for diagnostic/clinical purposes.
BRACS Tumor Subtyping
Upload a breast cancer image to predict the tumor type. Your image must be at 40X magnification. Do not otherwise modify your image.
This demo uses a custom-trained DINOv2 foundation model for pathology images called OpenMidnight with a linear classifier for tumor classification.
Images are classified as Normal, Pathological Benign, Usual Ductal Hyperplasia, Flat Epithelial Atypia, Atypical Ductal Hyperplasia, Ductal Carcinoma In Situ, Invasive Carcinoma
For this particular demo, images must be one of the sample classes - unsupported classes will yield confusing and/or useless results.
This demonstration is for illustrative purposes only and should not be used for diagnostic/clinical purposes.