Pathologists agreed just three-quarters of the time when diagnosing breast cancer from biopsy specimens, according to a recent study. The difficult, time-consuming process of analyzing tissue slides is why pathology is one of the most expensive departments in any hospital. Faisal Mahmood, assistant professor of pathology at Harvard Medical School and the Brigham and Women's Hospital, leads a team developing deep learning tools that combine a variety of sources -- digital whole slide histopathology data, molecular information, and genomics -- to aid pathologists and improve the accuracy of cancer diagnosis. Mahmood, who heads his eponymous Mahmood Lab in the Division of Computational Pathology at Brigham and Women's Hospital, spoke this week about this research at GTC DC, the Washington edition of our GPU Technology Conference. The variability in pathologists' diagnosis "can have dire consequences, because an uncertain determination can lead to more biopsies and unnecessary interventional procedures," he said in a recent interview.
Machine learning-guided virtual reality simulators can help neurosurgeons develop the skills they need before they step in the operating room, according to a recent study. Research from the Neurosurgical Simulation and Artificial Intelligence Learning Centre at the Montreal Neurological Institute and Hospital (The Neuro) and McGill University shows that machine learning algorithms can accurately assess the capabilities of neurosurgeons during virtual surgery, demonstrating that virtual reality simulators using artificial intelligence can be powerful tools in surgeon training. Fifty participants were recruited from four stages of neurosurgical training; neurosurgeons, fellows and senior residents, junior residents, and medical students. They performed 250 complex tumour resections using NeuroVR, a virtual reality surgical simulator developed by the National Research Council of Canada and distributed by CAE, which recorded all instrument movements in 20 millisecond intervals. Using this raw data, a machine learning algorithm developed performance measures such as instrument position and force applied, as well as outcomes such as amount of tumour removed and blood loss, which could predict the level of expertise of each participant with 90 per-cent accuracy.
A group of researchers from Beth Israel Deaconess Medical Center (BIDMC) and Harvard Medical School (HMS) have developed a way to train artificial intelligence to read and interpret pathology images. Scientists tested the artificial intelligence (AI) during a competition at the annual International Symposium of Biomedical Imaging, where it was tasked to look for breast cancer in images of lymph nodes. It turns out it can detect breast cancer accurately 92 percent of the time and won in two separate categories during the contest. Andrew Beck from BIDMC says they used the deep learning method, which is commonly used to train AI to recognize speech, images and objects. They fed the machine with hundreds of slides marked to indicate which parts have cancerous cells and which have normal ones.
"These are patterns that even the most sophisticated scientist couldn't detect by eye," said Lawrence A. David, Ph.D., a senior author of the study and assistant professor of molecular genetics and microbiology at Duke School of Medicine. "While some people are warning about artificial intelligence leading to killer robots, we are showing the positive impact of AI in its potential to overcome disease." The research, published this week in the Journal of Infectious Diseases, suggests that a focus on gut microbes may be important for developing improved vaccines and preventive approaches for cholera and other infectious diseases. "Our study found that this'predictive microbiota' is as good at predicting who gets ill with cholera as the clinical risk factors that we've known about for decades," said Regina C. LaRocque, M.D., MPH, of the Massachusetts General Hospital Division of Infectious Diseases, a senior author of the study and assistant professor of medicine at Harvard Medical School. "We've essentially identified a whole new component of cholera risk that we did not know about before."
You are free to share this article under the Attribution 4.0 International license. Machine learning algorithms can accurately assess the capabilities of neurosurgeons during virtual surgery before they step into an actual operating room, a new study shows. Researchers recruited fifty participants from four stages of neurosurgical training; neurosurgeons, fellows and senior residents, junior residents, and medical students. The participants performed 250 complex tumor resections using NeuroVR, a virtual reality surgical simulator. The National Research Council of Canada developed the system; CAE recorded all instrument movements in 20 millisecond intervals.