Brain Cancer Diagnosed in Minutes: AI System Hits 88% Accuracy

The Hetairos AI System

According to НВ — Техно: An artificial intelligence system called Hetairos can identify over 100 molecular subtypes of central nervous system tumors by analyzing digitized histological slides. This breakthrough technology delivers results within minutes. Developed by teams led by Moritz Gerstung at the German Cancer Research Center (DKFZ) and Felix Zamm at Heidelberg University’s Medical Faculty, the system demonstrates high diagnostic precision. For context, traditional DNA methylation analysis-the current gold standard for brain tumor classification-can take up to two weeks to return results.

Training and Validation of Hetairos

Hetairos was trained and validated using more than 11,000 digitized tissue samples collected from 9,606 patients across 11 medical centers on four continents. The system can distinguish between 102 distinct molecular tumor subtypes, and in 50-70% of cases, it makes predictions with high confidence. When operating at high confidence, its accuracy reaches approximately 87-88%. In a direct comparison with expert neuropathologists analyzing 210 cases, Hetairos achieved 68% accuracy, while the specialists reached only 30%.

The study shows that artificial intelligence can extract molecular information directly from standard tissue sections, fundamentally changing cancer diagnostics, said lead study author Darui Jin.

However, Moritz Gerstung noted that diagnosing very rare tumor types remains a serious challenge for Hetairos, though the system’s performance is expected to improve with larger and more diverse datasets.

Felix Zamm emphasized that Hetairos is designed as a diagnostic support tool, not a replacement for molecular analysis. He highlighted the technology’s importance, especially in resource-limited countries, since it relies on standard tissue sections that are available worldwide. This underscores Hetairos’s potential to improve medical diagnostics and expand access to quality healthcare.

DNA methylation analysis is considered the gold standard for classifying brain tumors, but results often take about two weeks. Hetairos, in contrast, can dramatically speed up the diagnostic process-a critical advantage for patients requiring urgent treatment.

The advancement of AI technologies in medicine, particularly in cancer diagnostics, opens new possibilities for improving the speed and accuracy of diagnostic workflows. Hetairos could become a vital tool for physicians, especially in resource-limited settings, as its ability to rapidly process histological slides may significantly impact patient outcomes. Continued research and refinement of this system promise even greater accuracy and accessibility for cancer diagnosis worldwide.

As advancements in cancer diagnostics continue to emerge, the potential for rapid detection is becoming a focal point in medical research. For instance, a recent development by Canadian researchers involves a patch that can identify skin cancer within just four days. This innovative approach highlights the growing trend of utilizing technology to expedite cancer diagnosis, paralleling the breakthroughs seen with AI systems like Hetairos. To learn more about this cutting-edge skin cancer detection method, visit this article.