A group of Google researchers has developed a machine learning and Augmented Reality -powered microscope which may help in the real-time detection of cancer and also save millions of lives. From the yearly meeting of the American Association for Cancer Research at Chicago, Illinois on Monday, Google explained a prototype augmented reality microscope platform which may help hasten and democratise the adoption of learning tools to pathologists around the globe. The stage consists of a light microscope that enables real time image analysis and demonstration results of ML algorithms directly in the field of view. The ARM can be installed in existing light microscopes around the globe, utilizing cost effective, easily available components, and with no need for entire drag numerical versions of a tissue being analysed. In principle, the ARM can offer a broad range of visual feedback, such as text, arrows, shapes, heatmaps or cartoons, and also is capable of running many kinds of machine learning algorithms targeted at solving various problems such as object detection, quantification or classification, Martin Stumpe, technical manager and Craig Mermel, Product Manager, Google Brain Team, wrote in a blog post.
Applications of learning to medical areas such as ophthalmology, dermatology, radiology, and pathology have shown excellent promise. The Google team configured ARM to conduct two distinct cancer detection algorithms, one which finds breast cancer metastases in lymph node specimens and another that finds cancer of the prostate at prostatectomy specimens. We believe that the MRA has the potential for a massive effect on global health, particularly to the identification of infectious diseases, including tuberculosis and malaria, from the developing countries, Google noted.