Brain tumours can be diagnosed more quickly and potentially more reliably in the operating room by using a portable diagnostic screening technology that distinguishes between cancer and normal tissue, reports a paper published online this week in Nature Biomedical Engineering.
Typically, after a surgeon has removed a primary tumour, preliminary and final diagnoses involve freezing and staining a tumour sample in the pathology lab, followed by assessment of the stained tissue by a pathologist. The accuracy and speed of the process, which can take from tens of minutes to hours, is essential in informing the surgeon about the operation’s success and in providing critical diagnostic information for patient prognosis and management. However, the long turnaround times can delay decision-making in the operating room, the necessary processing of tissue samples can introduce artefacts and rare tumours can be misclassified.
Daniel Orringer, Sandra Camelo-Piragua, and colleagues designed a portable technology that uses Raman spectroscopy - an imaging technique that provides a sample’s molecular fingerprint - to provide fast analysis of fresh brain tumour samples in the operating room, eliminating the need for sample processing. The ‘virtual pathology’ technology produces images that are almost indistinguishable from those of traditionally stained samples, and can be combined with machine learning to classify brain tumours into subtypes, with similar high accuracy (~90%) to that achieved with stained samples.
Although clinical trials will be required to determine its real-world effectiveness, the technology has the potential to assist neurosurgeons and neuropathologists in resecting and diagnosing brain tumours in the operating room.
Machine learning: Model identifies three biomarkers associated with COVID-19 mortalityNature Machine Intelligence
Virology: SARS-CoV-2 capable of infecting intestinal cellsNature Medicine