An Inclusive Foundation Model for Generalizable Cytogenetics in Precision Oncology

TL;DR

CHROMA is a large-scale, self-supervised foundation model for cytogenomics that improves generalization and reduces annotation efforts in chromosomal abnormality detection, advancing precision oncology.

Contribution

This paper introduces CHROMA, a pre-trained foundation model that learns generalizable representations of chromosomal abnormalities from extensive unlabeled data, outperforming existing methods.

Findings

Outperforms other methods across all abnormality types.

Requires fewer labeled data for effective training.

Handles imbalanced datasets effectively.

Abstract

Chromosome analysis is vital for diagnosing genetic disorders and guiding cancer therapy decisions through the identification of somatic clonal aberrations. However, developing an AI model are hindered by the overwhelming complexity and diversity of chromosomal abnormalities, requiring extensive annotation efforts, while automated methods remain task-specific and lack generalizability due to the scarcity of comprehensive datasets spanning diverse resource conditions. Here, we introduce CHROMA, a foundation model for cytogenomics, designed to overcome these challenges by learning generalizable representations of chromosomal abnormalities. Pre-trained on over 84,000 specimens (~4 million chromosomal images) via self-supervised learning, CHROMA outperforms other methods across all types of abnormalities, even when trained on fewer labelled data and more imbalanced datasets. By facilitating comprehensive mapping of instability and clonal leisons across various aberration types, CHROMA offers a scalable and generalizable solution for reliable and automated clinical analysis, reducing the annotation workload for experts and advancing precision oncology through the early detection of rare genomic abnormalities, enabling broad clinical AI applications and making advanced genomic analysis more accessible.