Learning from Lost Provenance: Multiple Instance Learning for Cancer Registry Tumor Group Classification

2026-07-03

AI Summary

This study addresses the challenge in cancer registry automation posed by the absence of fine-grained, pathology-report-aligned annotations, as only patient-level labels are typically available. To overcome this limitation, the authors propose an attention-based multiple instance learning (ABMIL) framework that implicitly infers report-level pseudo-labels from patient-level supervision. By leveraging label distillation, the method constructs a high-quality training set without requiring additional manual annotation or extensive computational resources, enabling efficient fine-tuning of tumor grouping classifiers. Evaluated on British Columbia cancer registry data, the approach achieves a macro F1-score of 0.83 and significantly outperforms existing baselines across most tumor categories, offering a practical and scalable solution for automating cancer registry workflows.

Abstract

Modernizing cancer registries with deep learning is opening new opportunities to automate labor-intensive tasks such as the coding of pathology reports. However, progress is constrained by the scarcity of report-level human-annotated training data. Cancer registries generate substantial volumes of expert-assigned labels as a routine product of their operations, but these exist at the patient level and are not linked to the individual pathology reports that informed them, limiting their direct use for training models. We develop an efficient framework for training deep learning classifiers by leveraging these operationally-generated labels without requiring per-report human annotation, demonstrated for tumor group classification at the BC Cancer Registry. We use Attention-Based Multiple Instance Learning (ABMIL) to recover the lost link between patient-level labels and the reports that informed them, leveraging the attention the model places on each report to distil a large, noisily-labeled corpus into a compact, high-quality per-report training dataset. A classifier fine-tuned on a distilled dataset achieved a macro F1 of 0.83, outperforming established baselines across most tumor groups. By turning routine operational labels into high-quality training data without additional annotation or large-scale computing infrastructure, ABMIL offers a practical and accessible route to automating cancer registry workflows.

Problem

• Multiple Instance Learning
• Cancer Registry
• Tumor Group Classification
• Lost Provenance
• Pathology Report Coding

Innovation

• Attention-Based Multiple Instance Learning
• cancer registry automation
• tumor group classification
• label distillation
• pathology report coding