Jeremy Howard, the former president and chief scientist of predictive-modeling platform Kaggle, is back with a new startup he thinks can revolutionize medical diagnostics using deep learning. There’s a lot of work to be done, but a lot of reason to be optimistic.
Researchers at the University of Washington, Tacoma, have built a machine learning system capable of predicting readmission risks for congestive heart failure patients. It has shown good results in a pilot deployment, and now the team hopes to commercialize the technology.
Apixio has raised a $13.5 million series C round of venture capital from Bain Capital Ventures and several angel investors. The company’s technology extracts data from numerous types of medical records, including patient charts and doctors’ notes, in order to help hospitals and insurance companies get a better sense of patients’ health for risk-assessment purposes. Apixio has raised more than $22 million since launching in 2009. It first launched publicly in 2011, then focusing on its semantic-analysis engine for identifying latent connections between diagnoses and symptoms.
A startup called BaseHealth launched on Tuesday with a mission to deliver personalized wellness plans while keeping doctors very much in the picture. The company’s platform combines genetic data, lifestyle data and medical records to determine patients’ risks and how they can mitigate them.
With $15 million from Kleiner Perkins and Jafco Ventures, Zephyr Health thinks it’s well-suited to help pharmaceutical companies and medical device makers navigate tougher financial times by making better use of the data around them.
IBM has shared some details about a new project called WatsonPaths that lets doctors actually interact with the system to understand how it came to its conclusions, and to tell it whether its “thinking” was right. This type of interaction is critical in any type of machine learning system where speed isn’t the primary objective, because it lets humans see things they might not have and also train the machines to be more accurate in the future. WatsonPaths is a GUI-based tool and is being developed along with doctors at the Cleveland Clinic.
Two new research partnerships whose participants range from pharmaceutical companies to IT vendors are taking aim at improving disease treatment via data analysis. They’re targeting a handful of diseases specifically — heart disease and cancer among them — but they point toward a data-driven health care future.
It’s easy enough to quantify our lives with data and score people against the models we build, but sometimes — like in the case of organ transplants — deciding what scores matter is a matter of life and death.
GE Healthcare is pushing a system called Corvix for doing agent-based simulations on complex problems. In India, the technology simulated a population of 80 million people in order to determine the best places to build medical facilities.
A group of researchers from Columbia and Stanford have created a method for turning complex cellular datasets into visualizations that map the similarities between tens of thousands of cells within a tissue sample.