FIRST LOOK: A Machine Learning Algorithm to Reduce Costs and Increase Quality


Diagnostic tests are the lynchpin of medical care. From a health point of view, bad testing can lead to missed or incorrect diagnoses. From a cost point of view, bad testing is a major source of unneeded expenditures—the majority of the care labeled as “low value” by the Choosing Wisely campaign were diagnostic tests.

Click here to watch Dr. Obermeyer’s First Look presentation.

And yet testing decisions are notoriously hard – especially in a world where every patient is a “big data” challenge. Electronic health records hold exponentially increasing amounts of data as patients age, medical complexity increases, and technology expands rapidly. Doctors must absorb all this information in deciding which, if any, tests to deploy.

My team is building algorithms to aid physicians in these testing decisions – we have one completed prototype and several in the works. Importantly we are finding that these algorithms add value in two different ways. First, they can save costs – by dramatically reducing low-value testing of patients in whom nothing is found. Second, they can actually improve health outcomes – by identifying large pools of “misses”: untested patients who actually could have been caught, had they been tested in time. I will discuss what we might need to take these prototypes to scale as well as the kind of partnerships needed to build and apply these algorithms.

For more information about Dr. Obermeyer’s research, please contact Partners HealthCare Innovation by clicking here.

Most Recent Posts:

Blog Chen Molecule Unlocking Glaucoma image

Mass Eye and Ear Scientist Identifies Molecule Unlocking Glaucoma

Potential Cure May Impact Other Neurodegenerative Disease Therapies   There is no cure (yet). Glaucoma often robs…

Read More

First Brigham Ignite Awards of the Year Announced

The Two Projects Focus on Robotic AI-Guided Intubation and Gamma Delta T Cell Therapy for Solid Tumors…

Read More
Blog Barriers to Innovation image

‘Barriers to Innovation’ Survey Shows Improvement in Underrepresented Faculty’s Perception as Innovators

Findings inform educational programs like CILP, next slated for May 15   Progress is being made in…

Read More