# Multiple Instance Learning for the Detection of Lymph Node and Omental Metastases in Carcinoma of the Ovaries, Fallopian Tubes and Peritoneum

## TL;DR

This study shows that AI can accurately detect ovarian cancer metastases in lymph nodes and omentum, potentially saving time for pathologists.

## Contribution

The study introduces an AI model using attention-based multiple-instance learning for metastasis detection in ovarian cancer.

## Key findings

- The model achieved an AUROC of 0.998 and 100% balanced accuracy for lymph node metastases.
- For omental metastases, the model reached an AUROC of 0.963 and 98% balanced accuracy.
- The model could pre-screen slides, reducing workload and improving diagnostic efficiency.

## Abstract

Ovarian cancer staging hinges on the histopathological evaluation of large amounts of non-primary tumour-related tissue (e.g., lymph nodes and omentum) for the presence of metastatic disease. This study aimed to determine whether artificial intelligence could effectively identify nodal and omental metastatic cancer deposits using attention-based multiple-instance learning to classify whole-slide images (WSIs) as either containing tumour cells or not. Training and validation were conducted with a total of 855 WSIs of surgical specimens from 404 patients. All objective measures of accuracy demonstrated the model’s great potential in identifying metastatic disease. In the clinical setting, this model could potentially pre-screen WSIs prior to histopathologist review, offering significant time-saving benefits and streamlining clinical diagnostic workflows.

Background/Objectives: Surgical pathology of tubo-ovarian and peritoneal cancer carries a well-recognised diagnostic workload, partly due to the large amount of non-primary tumour-related tissue requiring assessment for the presence of metastatic disease. The lymph nodes and omentum are almost universally included in such resection cases and contribute considerably to this burden, principally due to volume rather than task complexity. To date, artificial intelligence (AI)-based studies have reported good success rates in identifying nodal spread in other malignancies, but the development of such time-saving assistive digital solutions has been neglected in ovarian cancer. This study aimed to detect the presence or absence of metastatic ovarian carcinoma in the lymph nodes and omentum. Methods: We used attention-based multiple-instance learning (ABMIL) with a vision-transformer foundation model to classify whole-slide images (WSIs) as either containing ovarian carcinoma metastases or not. Training and validation were conducted with a total of 855 WSIs of surgical resection specimens collected from 404 patients at Leeds Teaching Hospitals NHS Trust. Results: Ensembled classification from hold-out testing reached an AUROC of 0.998 (0.985–1.0) and a balanced accuracy of 100% (100.0–100.0%) in the lymph node set, and an AUROC of 0.963 (0.911–0.999) and a balanced accuracy of 98.0% (94.8–100.0%) in the omentum set. Conclusions: This model shows great potential in the identification of ovarian carcinoma nodal and omental metastases, and could provide clinical utility through its ability to pre-screen WSIs prior to histopathologist review. In turn, this could offer significant time-saving benefits and streamline clinical diagnostic workflows, helping to address the chronic staffing shortages in histopathology.