Research progress in artificial intelligence for brain metastases

TL;DR

This review explores how artificial intelligence is being used to improve the diagnosis and treatment of brain metastases through imaging techniques.

Contribution

The paper systematically reviews recent AI applications in brain metastases imaging for segmentation, diagnosis, and prognosis.

Findings

AI improves the automatic detection and segmentation of brain metastases using advanced imaging.

AI helps differentiate brain metastases from other intracranial lesions.

AI shows potential in predicting prognosis and new metastatic developments.

Abstract

As artificial intelligence (AI) continues to evolve, its integration into medical practice is becoming increasingly prominent, particularly in the field of neuro-oncology. This review examines the application of AI—specifically machine learning (ML) and deep learning (DL)—in the imaging evaluation of brain metastases (BM). A systematic search of PubMed was conducted to identify relevant studies published within the past 5 years. The retrieved literature was categorized and analyzed according to three key clinical tasks: segmentation, differential diagnosis, and prognostic prediction. We first outline the capabilities of AI in the automatic detection and segmentation of BM using advanced imaging techniques. Subsequently, we synthesize evidence on how AI aids in distinguishing BM from other intracranial structures and lesions. Finally, we discuss the emerging role of AI in predicting disease prognosis and the development of new metastatic abnormalities. Current evidence suggests that AI not only enhances diagnostic efficiency and reproducibility but also provides clinically meaningful insights that support personalized treatment planning. Importantly, the integration of AI into neuro-oncological imaging remains at a nascent stage, indicating substantial potential for future growth and refinement in both technical performance and clinical applicability.

Figures