In a talk given today at the American Association for Cancer Research’s annual meeting, Google researchers described a prototype of an augmented reality microscope which might be utilized to help doctors diagnose patients. When pathologists are examining biological tissue to see if there are indications of cancer, and if so, how much and what kind, the procedure can be very time-consuming. And it is a practice that Google thinks might benefit from deep learning tools. However, in most places, adopting AI technology is not feasible. The organization believes this microscope could allow groups with restricted funds, like little labs and labs, or developing nations to benefit from all of these tools in an easy, easy-to usage manner.

Google says the scope might perhaps help accelerate and democratize the adoption of learning tools for pathologists around the globe.” . The microscope is a regular light microscope, the kind utilized by pathologists worldwide. Google just tweaked it a bit in order to present AI technology and augmented reality. Neural networks are trained to find cancer cells in images of human tissue. Then, following a slide with human tissue is placed under the microscope, the same picture an individual sees throughout scope’s eyepieces is fed into a computer. AI algorithms then find cancer cells in the tissue, which the system then outlines in the picture seen throughout the eyepieces.

It is all done in real time and works fast enough that it is still effective when a pathologist moves a slide to look at a fresh section of tissue. Thus far, the Google group has demonstrated the scope’s capabilities with breast and prostate cancer, showing the system is accurate in its assessments. However, Google says such a set up might be utilized to detect other cancers as well as infectious diseases such as tuberculosis and malaria. It might also be utilized to display other data or annotations to the view. Of course, microscopes have proven useful in industries other than pathology, and we believe the Augmented Reality Microscope could be adapted for a wide range of applications across healthcare, life sciences research and material science, says Google. This evidence of concept analysis is also described in a newspaper that is currently under inspection. Google says that a more in-depth study is required for robust assessments of the system’s performance and shortcomings. It concluded We are excited to continue to research how an Augmented Reality Microscope might help accelerate the adoption of machine learning for favorable impact around the globe.”.


Source: Google