It’s hard not to get excited about the Tricorder X prize if (like me) you are interested in new things in general, and in ways innovation can be applied to reengineering the US healthcare system in particular. I spent some time recently reading carefully through the draft guidelines which describe just what this “Tricorder” needs to do in order to win the $7M first prize. What strikes me most, apart from the sheer ambitiousness of the goals, is the extent to which success is going to require us health industry types to approach the world differently than we have done in the past.
Tricorder target requirements
“The winner(s) of the $10M Qualcomm Tricorder X PRIZE will be the best‐performing solution in its ability to diagnose a set of 15 distinct conditions in a pool of people within three days, while providing a strong consumer experience. In addition, the winning solutions must:
- Meet minimum scores for both diagnosis and consumer experience
- Continuously monitor five vital signs over the course of the consumer testing period and log this data to the cloud
- Have a maximum mass of no more 5 than pounds
This diagnosis and monitoring must be performed in the hands of a consumer, independently of a healthcare worker or facility.”
Breaking down silos
The 15 diseases (see bottom of post) that need to be diagnosed look like they have been selected with prevalence and importance in mind. But looking at them from a technical point of view, what is striking is how diverse are the parameter sets you need to measure to be able to diagnose all 15 diseases.
Broadly speaking, there seem to me to be at least four silos of expertise that need to be brought together for this project, and mostly these groups of people do not intersect much.
- There are chemical-type measurements (e.g. blood sugar, or hemoglobin concentration in blood; or markers of infection such as leukocyte esterase in urine).
- And there are electro-mechanical measurements (e.g. EKG patterns for atrial fibrillation; or EEG signals for sleep apnea; or image pattern recognition for melanoma detection).
- Then there is some type of AI diagnostic algorithm set that takes these various physical and chemical measurements and turns them into differential diagnoses.
- Not to mention the need for real time monitoring and storage in the cloud of 5 vital signs.
Traditionally, participants in the healthcare ecosystem have stuck pretty much to one of these four silos. Doctors and nurses have worried about integrating various tests and measurements to make diagnoses. Health IT people are busy worrying about EHR’s and moving data around in the cloud.
Then in the medical device world, one group of people have been systematically moving forward point-of-care chemical analysis for the last few decades, creating along the way companies like Abaxis, iStat, and Lifescan to name a few. In parallel, a different set of individuals have been advancing the state of the art in “patient monitoring”, measuring EKGs, blood pressure, oximetry and other parameters using optical and electrical techniques, and creating companies like Nellcor, Masimo, and Critikon.
To be successful at measuring the 15 diseases that are the focus of the Xprize, a team is going to have to take existing technology in each of these different silos. Shrink it down in many cases, and then integrate it all. And I suspect some of the necessary tests are not available as yet outside the laboratory (e.g. thyroid testing), and so they need to be converted to point-of-care formats.
To me it feels a bit like a new Moon shot. And not really something that a traditional startup or venture capitalist would have tackled. It’s really a perfect example of how a “prize” can capture the imagination and focus people on a quite audacious step forward.
I’m not sure yet whether or not a real Tricorder will emerge. But I am pretty sure some amazing things will.
Tricorder X prize criteria
Diagnosing 15 diseases
- Urinary tract infection
- Atrial fibrillation
- Strep throat
- Sleep apnea
- Melanoma screen
- Chronic obstructive pulmonary disease (COPD)
- Abnormalities in a comprehensive metabolic panel
- Absence of disease
Monitoring 5 Vital Signs
- Arterial blood pressure
- Respiratory rate
- Pulse oximetry