AI Helping To Speed Brain Cancer Diagnosis

“As surgeons, we’re limited to acting on what we can see; this technology allows us to see what would otherwise be invisible, to improve speed and accuracy in the OR, and reduce the risk of misdiagnosis,” said senior author Daniel A. Orringer, MD, associate professor of neurosurgery at NYU Grossman School of Medicine, who helped develop SRH and who co-led the study with colleagues at the University of Michigan. “With this imaging technology, cancer operations are safer and more effective than ever before.”

Around 15.2 million people around the globe are diagnosed with cancer every year, among them some 80% will undergo surgery and in many cases part of the tumor will be analyzed during surgery to help provided preliminary diagnosis. In America over 1.1 million biopsy samples are taken every year to be interpreted by pathologists, which is based on hematoxylin and eosin staining of processed tissue, time, resources, and labor, according to the authors.

“The conventional workflow for intraoperative histology, dating back over a century, necessitates tissue transport to a laboratory, specimen processing, slide preparation by highly trained technicians, and interpretation by a pathologist, with each step representing a potential barrier to delivering timely and effective surgical care,” they wrote. 

This new approach takes advantage of advances in optics and artificial intelligence; SRH imaging techniques offer label free, submicrometer resolution imaging of unprocessed biological tissues to reveal tumor infiltration which are then processed and analyzed using a deep learning AI algorithm to provide surgeons brain tumor diagnosis within 3 minutes. Resection using this same technology can then allow surgeons to accurately detect and remove what may be otherwise undetectable tumors.

“SRH utilizes the intrinsic vibrational properties of lipids, proteins, and nucleic acids to generate image contrast, revealing diagnostic microscopic features and histologic findings poorly visualized with hematoxylin and eosin (H&E)-stained images, such as axons and lipid droplets, while eliminating the artifacts inherent in frozen or smear tissue preparations,” write the authors.

For this study a deep convolutional neural network was trained on more than 2.5 million samples from 415 patients to classify tissue into 13 histologic categories representing the most common brain tumors, the CNN was then validated in a clinical trial involving 278 cancer patients undergoing brain tumor resection or epliepsy surgery at 3 medical centers; patient biopsies were split intraoperatively into sister specimens and were randomly assigned to either the experimental or control group. Control specimens received standard practice and were transported to a pathology laboratory for processing, slide preparation, and interpretation taking about a half hour. Experimental specimen analysis was performed intraoperatively from image acquisition and processing to diagnostic prediction via CNN. 

“Notably, the CNN was designed to predict diagnosis independent of clinical or radiographic findings, which were reviewed by study pathologists and are often of central importance in diagnosis,” the scientists wrote.

According to the authors their results showed: “Overall diagnostic accuracy was 93.9% (261/278) for the conventional H&E histology arm and 94.6% (264/278) for the SRH plus CNN arm, exceeding our primary endpoint threshold for noninferiority (>91%).” Data showed that diagnostic errors in the experimental group were distinct from errors in the control group suggesting this technique could possibly achieve close to 100% accuracy. 

“In conclusion, we have demonstrated how combining SRH with deep learning can be employed to rapidly predict intraoperative brain tumor diagnosis,” the authors commented. “Our workflow provides a transparent means of delivering expert-level intraoperative diagnosis where neuropathology resources are scarce, and improving diagnostic accuracy in resource-rich centers. The workflow also allows surgeons to access histologic data in near real-time, enabling more seamless use of histology to inform surgical decision-making based on microscopic tissue features.”

“SRH will revolutionize the field of neuropathology by improving decision-making during surgery and providing expert-level assessment in the hospitals where trained neuropathologists are not available,” noted study co-author Matija Snuderl, MD, associate professor in the department of pathology at NYU Grossman School of Medicine.

“Importantly, our AI-based workflow provides unparalleled access to microscopic tissue diagnosis at the bedside during surgery, facilitating detection of residual tumor, reducing the risk of removing histologically normal tissue adjacent to a lesion, enabling the study of regional histologic and molecular heterogeneity, and minimizing the chance of nondiagnostic biopsy or misdiagnosis due to sampling error.”