Medical tech innovations are much in the news these days. While many of these seem like distant, pie-in-the-sky ideas, experts tell DecisionHealth you can expect see some widely used in health care over the next few years.
 
Natural language processing (NLP), for example, is emerging as a way to clean up unstructured data. Retail giant Amazon’s AWS (Amazon Web Services) division announced Nov. 27 that it was launching Comprehend Medical – “a natural language processing service that makes it easy to use machine learning to extract relevant medical information from unstructured text … [such as] doctors’ notes, clinical trial reports and patient health records.” It’s currently available only to customers of AWS.
 
An example of a possible use for Comprehend Medical given on the AWS home page is that it can “extract ‘methicillin-resistant Staphylococcus aureus,’ often input as ‘MRSA,’ and provide context, such as whether a patient has tested positive or negative, to make the extracted term meaningful.” Amazon also suggests applications for population health, identifying optimal candidates for clinical trials and more.
 
Comprehend Medical is not the only entity promising to parse raw medical data with technology. A spokesman for the Israeli firm MedAware tells us its product “utilizes AI [artificial intelligence] technology to eliminate prescription errors” by learning “the prescribing patterns of physicians, patients and institutions to identify those prescriptions that are outliers to learned pattern, catching the most unexpected errors [such as] a healthy patient prescribed chemo [or] an 80-year-old woman prescribed birth control.”
 
It’s all connected
  • Internet of things. Another hot topic is patient devices connected to, and communicating with, electronic health records (EHR) on the “internet of things” (IoT) model that has given us Alexa and home maintenance apps like Nest. Though much of the discussion of these devices have centered on the Apple Watch and its potential as a health-monitoring source for providers, experts to whom DecisionHealth spoke say this trend will develop from many directions. Interactive health care technology for patients and their care partners will take off as those products become more seamless and reliable, expects Liddy Manson, director of the AgingWell Hub at Georgetown University's Global Social Enterprise Initiative. That happened in the case of her daughter, a Type 1 diabetic who resisted using blood glucose trackers into which she had to enter data manually until they were fully automatic – like the Dexcom unit she eventually adopted. Manson predicts fast adoption for products that provide end-to-end service – as a hypothetical (for now) example, an automated pill-dispensing technology that delivers prescriptions “blister-packed from CVS with a HIPAA-compliant Alexa telling you to take your medication. That’s what’s going to make the seesaw tip: making solutions as an additive to technology that people are already using.” Some companies are already heading in that direction, she says, such as Withings with its Bluetooth-enabled products such as a scale that also reads BMI, muscle and bone mass and heart rate. Taqee Khaled, director of innovation and strategy at digital business consultancy The Nerdery in Chicago, knows several companies working on funding for and testing IoT ideas for hospitals. 
  • Apps for opioids. The opioid crisis and the federal government's commitment to fighting it provide fertile ground for tech product development. In May 2018, the U.S. Food & Drug Administration (FDA) launched an "innovation challenge" to get tech companies to develop "novel solutions to detecting, treating and preventing addiction, addressing diversion and treating pain." In November 2018, the FDA selected eight winners of the challenge, including a pain therapy device from Avanos and a rapid dry screen from Algomet Rx, which products FDA promised expedited review and other special considerations. Meanwhile the FDA continues to advance other opioid treatment helpers; on Dec. 10, 2018, for example, the agency cleared the reSET-O mobile app "to help increase retention (the amount of time a patient participates) in an outpatient treatment program for individuals with opioid use disorder (OUD)." And private companies are cooking up their own solutions: health care services company KAMMCO, for example, has just launched a dashboard for providers in their HIE that "allows clinicians to identify individuals in their patient population who received at least one prescription/administration of opioids/controlled substances, by facility and date range up to 12 months." This provides more complete data than even typical prescription drug monitoring programs (PDMPs), the company claims, because in addition to filled prescription data, it also includes medications administered to patients in health care facilities.