AI Transforming Brain Tumour Detection: Rapid and Accurate Diagnosis

In a groundbreaking development, artificial intelligence (AI) has revolutionized the diagnosis of brain tumours. A recent study in Nature Medicine showcased the superior speed and precision of AI compared to conventional pathology methods in cancer detection.

As reported by The Independent, this advanced machine-learning technology outperformed pathologists by a minute 1% in accuracy, but its significant achievement lay in its remarkable efficiency, delivering results in just under two and a half minutes, a far cry from the 20 to 30 minutes typically required by pathologists.

In this innovative research, experts from New York University’s School of Medicine and Langone Hospital utilized laser-scattered light to identify cancerous tissue. They employed a neural network, trained with over 2.5 million samples from 415 patients across 13 categories, including common brain tumour types like malignant glioma, lymphoma, metastatic tumours, and meningioma.

During brain surgeries involving 278 patients across three medical centers, tissue specimens were divided for examination. The traditional pathology process was used for some, while others underwent experimental AI assessment during the surgery. Notably, the errors made by both groups varied, suggesting the possibility of achieving 100% accuracy by combining pathologist expertise with AI analysis.

Dr. Daniel Orringer, an associate professor of neurosurgery at NYU’s Grossman School of Medicine and a senior author of the study, emphasized the groundbreaking advancements. He stated, “As surgeons, our actions are limited to what is visually identified; this technology opens our eyes to otherwise hidden attributes, enhancing speed and accuracy in the operating theatre and reducing the risk of misdiagnosis. This imaging technology is reshaping cancer operations, making them safer and more effective than ever before.”

The pioneering imaging technology, stimulated Raman histology (SRH), developed by Dr. Orringer, illuminates features typically unseen in tissue samples, allowing for easier identification of tumour tissues.