Amazon Buys Health Navigator, Founded by Chicago ER Doctor David Thompson

Amazon continues to scale up its healthcare investments and their AI innovations. The company has recently acquired Health Navigator, a startup that creates APIs for online health services. Health Navigator is designated to become part of Amazon Care, its pilot health care service program for employees, Amazon Care, which will serve as a medical benefit providing employee health care. The service will provide virtual house calls via video, with home visits for escalated care needs.

Amazon has been investing in healthcare 

Amazon has a clear interest in healthcare. With one of the largest workforces in the world, and with major financial backing capabilities, the e-commerce giant is creating a series of innovative healthcare solutions and has already invested heavily in healthcare.

This is the second health startup acquired by Amazon, the third-largest retailer on the planet following Walmart and CVS, and its achieving enviable scale. The company first bought online pharmacy PillPack, completing the purchase in 2018, for a little under $1 billion. Amazon has integrated PillPack’s services into Amazon Care, which offers deliveries of prescriptions with remotely communicated treatment plans. Known for its ongoing interest in revolutionizing healthcare, in late 2018 Amazon launched Comprehend Medical, a machine learning service for health information. An early adoptee of Amazon’s AI system was Change Healthcare, which processes health claims for pharmacies and uses Comprehend Medical to anticipate whether an insurance claim will be denied.

Health Navigator’s goals and how it works

Health Navigator’s platform was designed to be integrated into online health services, including telemedicine and medical call centers. The final aim is to standardize the process of working with patients. Its platform works with pure language processing-based instruments in order to document health complaints and provide care suggestions and is built-in into apps with APIs.

On its website, Health Navigator describes how it partners with organizations such as Microsoft to provide patients and medical staff with symptom-checking tools that create opportunities for remote diagnosis and ease its use as a medical tool. The platform also helps with triage, assisting patients in deciding whether to stay at home, see a doctor, or go directly to the emergency room. Also used by a growing number of telemedicine companies, Health Navigator offers virtual home exams and apps for doctors to connect with patients.

The platform has comprehensive human language interpretation capabilities, which help to specify symptoms even when patients use atypical language to describe them. For example, patients may describe symptoms with brief descriptors like “headache” or longer phrases such as “I have bad diarrhea and stomach pain.” Some patients may use slang terms like puking, throwing up, or barfing to describe vomiting. But the system’s language platform parses through a constellation of terms to find the most accurate for conveying symptomology to medical professionals.

A ‘translator’ from patient to doctor language

The Health Navigator NLP engine works with both US and UK English free-text. It basically translates symptoms patients describe into probable diagnosis.

• The Health Navigator NLP engine recognizes sentences, phrases, slang, and even understands common misspellings that patients use to describe their medical issues. It is also able to recognize the scientific terms the medical staff uses to document symptoms, diseases, and conditions.
• The NLP engine translates narrative free-text into a usable and intelligent medical terminology using Health Navigator’s proprietary Coded Chief Complaints vocabulary. The Coded Chief Complaints clinical vocabulary covers over 99% of the reasons for patient visits.
• The NLP engine can translate common self-diagnosis language that a patient may enter such as “bladder infection.” 
• The NLP engine has an overall accuracy of 95-99% in correctly identifying the reason for the visit.

The Health Navigator Diagnosis Engine generates a list of possible causes (pre-diagnoses) for a specific symptom or problem. This can be used in digital health assistants (health bots, diagnosis symptom checkers) and electronic health records.

The Health Navigator Diagnosis Engine is a proprietary knowledge-based inference engine. It accepts as inputs: age, gender, pregnancy status, symptoms, known medical problems, and other clinical factors. It uses Bayesian logic to output a list of causes sorted by likelihood. This expert system is a structured and evolving database of curated content. It is the result of nearly two decades of subject matter expert input, literature review, data mining and analytics, and testing against large clinical data sets.

 Health Navigator’s founder, an ER doctor, and teacher

The startup now acquired by Amazon was founded in 2014 by physician David Thompson. Thompson is well known in the medical sector for developing a set of protocols, called Schmitt-Thompson, which has become a standard scale used by nurses and other clinicians to guide patients to medically appropriate care, typically via call centers. According to his LinkedIn profile, Thompson is an emergency medicine doctor and a part-time faculty member at Northwestern Memorial Hospital in Chicago, Illinois.