Foundation models for fast, label-free detection of glioma infiltration

• 成果类型: Article
• 署名作者: Kondepudi, Akhil; Pekmezci, Melike; Hou, Xinhai; Scotford, Katie; Jiang, Cheng; Rao, Akshay; Harake, Edward S.; Chowdury, Asadur; Al-Holou, Wajd; Wang, Lin; Pandey, Aditya; Lowenstein, Pedro R.; Castro, Maria G.; Koerner, Lisa Irina; Roetzer-Pejrimovsky, Thomas; Widhalm, Georg; Camelo-Piragua, Sandra; Movahed-Ezazi, Misha; Orringer, Daniel A.; Lee, Honglak; Freudiger, Christian; Berger, Mitchel; Hervey-Jumper, Shawn; Hollon, Todd
• 署名单位: University of Michigan System; University of Michigan; University of Michigan System; University of Michigan; University of California System; University of California San Francisco; University of California System; University of California San Francisco; University of Michigan System; University of Michigan; Medical University of Vienna; Medical University of Vienna; University of Michigan System; University of Michigan; Rutgers University System; Rutgers University New Brunswick; New York University; University of Michigan System; University of Michigan; Medical University of Vienna
• 刊物名称: Nature
• ISSN/ISSBN: 0028-3562
• DOI: 10.1038/s41586-024-08169-3
• 发表日期: 2025-01-09
• 页码: 439-+
• 关键词: raman-scattering microscopy; 5-aminolevulinic acid; brain-tumors; resection; surgery; extent; robust; TRIAL

摘要

A critical challenge in glioma treatment is detecting tumour infiltration during surgery to achieve safe maximal resection(1-3). Unfortunately, safely resectable residual tumour is found in the majority of patients with glioma after surgery, causing early recurrence and decreased survival(4-6). Here we present FastGlioma, a visual foundation model for fast (<10 s) and accurate detection of glioma infiltration in fresh, unprocessed surgical tissue. FastGlioma was pretrained using large-scale self-supervision (around 4 million images) on rapid, label-free optical microscopy, and fine-tuned to output a normalized score that indicates the degree of tumour infiltration within whole-slide optical images. In a prospective, multicentre, international testing cohort of patients with diffuse glioma (n = 220), FastGlioma was able to detect and quantify the degree of tumour infiltration with an average area under the receiver operating characteristic curve of 92.1 +/- 0.9%. FastGlioma outperformed image-guided and fluorescence-guided adjuncts for detecting tumour infiltration during surgery by a wide margin in a head-to-head, prospective study (n = 129). The performance of FastGlioma remained high across diverse patient demographics, medical centres and diffuse glioma molecular subtypes as defined by the World Health Organization. FastGlioma shows zero-shot generalization to other adult and paediatric brain tumour diagnoses, demonstrating the potential for our foundation model to be used as a general-purpose adjunct for guiding brain tumour surgeries. These findings represent the transformative potential of medical foundation models to unlock the role of artificial intelligence in the care of patients with cancer.