Google’s Voice Recognition Project Could Revolutionize Electronic Health Records

Google is taking steps to incorporating its artificial intelligence (AI) software into the medical field. Medical Digital Assist is the title of their research project, and its main goal is to facilitate doctor’s note taking process via voice recognition. The project stems from the healthcare side of their large-scale AI program, Google Brain, and is likely to incorporate existing technologies from Google’s Home, Translate, and Assistant product lines.  

Alongside Stanford Medicine, the Google team tackled a digital scribe study in 2017 using speech recognition tools to assist doctors in composing electronic health records (EHRs) from interactions with patients. Filling out EHRs can be a very tedious tasks for physicians, deviating a good chunk of their time from focusing on patient care.  

The greatest challenge to overcome in creating this technology is that it must be able to extract important information from interactions between the doctor and patient and form coherent notes on it. Though it contains much dialect pertaining to important medical issues, the interaction between patient and doctor also consists of normal small talk.  The “How was your vacation?”, or “That was a great game last night!” types of conversation that occur in the clinical setting would have to be carefully sifted through by this voice recognition technology. This is a difficult challenge, but if solved, would save physicians hours of note taking.  

Google Brain will conclude its first phase of study this August, and according to the head of the Stanford research team, the two will continue to collaborate throughout the second phase.  

Technology in Practice

Many other tech companies are taking a stab at incorporating AI into healthcare. Augmedix has been equipping physicians with Google Glass smart glasses to record clinical interactions. The devices camera will stream to remote scribes, who can document what they see into medical records. In this manner, the notes will be taken by a human being who can filter out extraneous conversation, as well as pick up on non-verbal cues. Additionally, Microsoft has been working with Pittsburgh Medical Center to create an AI scribe system. Amazon is attempting to implement AI into healthcare as well, using medical records to recognize poor diagnoses.  

Source: CNBC 

Jack is a Pre-Med Student at Penn State University with a keen interest in how new medical technologies are changing the future of healthcare. Reach out to Jack if you have a compelling story idea or with feedback about past articles.