The influx of innovative technologies has disrupted most of the industries around the globe in the recent past. Healthcare is no different, especially in the case of machine learning, automation, and artificial intelligence (AI). A recent analysis report from CB Insights states that about 86 percent of healthcare provider organizations, healthcare technology vendors, and life science companies are harnessing the potential of AI technology. AI, which works by mimicking human cognitive functions, can bring a paradigm shift in healthcare. This is powered by increasing availability of healthcare data and fast progress of innovative analytics techniques. Some of the most widely adopted AI techniques across healthcare industry include machine learning methods for structured data. This consists of the standard support vector machine, deep learning techniques, and neural network, as well as natural language processing for unstructured data.
Recently, researchers at Google Brain introduced an innovative computer vision for the identification of protein crystallization. It is used to determine the shape of cells and plays a vital role in drug discovery to treat various illnesses. According to the team, the AI works with accuracy rates of over 94 percent. A large number of experiments are conducted for each protein. Though most of the imaging processes are fully automated, the final stage of finding individual protein crystals is through visual inspection, which is prone to human error. Missing any structure can ultimately result in losing the chances for innovative biomedical discoveries. Google trained the AI model in association with Machine Recognition of Crystallization Outcomes (MARCO) initiative, a partnership between pharmaceutical companies and academics. The final resultant solution has been open-sourced and made available on GitHub and detailed in a paper published in the journal PLOS One.
Leading enterprises and startups alike are increasingly stepping into the healthcare space, especially for offering imaging services. Another renowned company, Baidu Research, recently revealed an algorithm that can identify tumor cells in breast tissue. Some of the early results explain that AI has become more reliable than people in the detection of various cancers.
Google researchers also published the results of work with predictive deep learning models in their journal NPJ Digital Medicine. The models are able to determine whether a patient is likely to be readmitted, admission timeline, and the chances of survival. Recently, Google also shared a computer vision model, which detects diabetic retinopathy, a key indicator of cardiovascular disease.