ClosedLoop.ai (formerly Deep Health Technologies), a health care data science startup specializing in AI and automation, today raised $11 million in series A funding led by Greycroft and .406 Ventures with participation from Silicon Valley Bank and Meridian Street Capital. The company plans to put the fresh capital toward marketing and product R&D efforts as it looks to expand its growing customer base.
Health care data will experience a compound annual growth rate of 36% through 2025, according to a 2018 report from IDC. Despite the coming data deluge, health systems remain woefully unprepared. One 2017 study showed 56% of hospitals have no strategies for data governance or analytics.
ClosedLoop was founded to solve that problem in 2017 by Boxer and Ciphent cofounders Andrew Eye and Dave DeCaprio. (VMWare and Optiv acquired Boxer and Ciphent, respectively, for undisclosed amounts.) DeCaprio previously helped to lead the Human Genome Project at MIT’s Broad Institute, an international scientific research project with the goal of determining the base pairs that make up human DNA, and he and Eye were longtime friends who worked together on prior projects including Boxer.
The ClosedLoop platform provides off-the-shelf AI models and automation workflows for health care applications and manual processes involving data science tasks. In addition to preventable hospitalizations, readmissions, risk, chronic disease onset, medication adherence, member churn, and out-of-network utilization, the startup claims its algorithms can match patients with clinical trials, predict product efficacy, and personalize care pathways.
ClosedLoop ostensibly makes it simple to import health care datasets like medical claims, prescriptions, electronic medical records, and custom data including socioeconomic factors, geolocation information, environmental data (like weather), medical device metrics, and wearable data without the need for normalization and cleansing. Hospitals and health insurers can build models in 24 hours with over 2,000 features (e.g., medication adherence, commodities, and frailty) using bespoke tools while at the same time operationalizing models to ensure they’re automatically updated as new data arrives. An API allows advanced users to leverage ClosedLoop’s capabilities within environments like Python and Jupyter notebooks. (Jupyter notebooks are among the most popular apps for creating and sharing machine learning code.)
Bias reporting for age, gender, ethnicity, and socioeconomic status are built directly into the platform and can be evaluated during model training, ClosedLoop says. They’re monitored daily for production deployed models, which is important — partly due to a reticence to release code, datasets, and techniques, much of the data used today to train AI algorithms for diagnosing diseases may perpetuate inequalities, a growing body of research has found.
Eye notes that there were approximately 3.3 million adult 30-day all-cause hospital readmissions in the U.S. in 2011 associated with $41.3 billion in hospital costs. Annually, approximately 2 million patients suffer with health care-associated infections and nearly 90,000 die. Roughly 4.4 million hospital admissions for acute illness or worsening chronic conditions totaled $30.8 billion in hospital costs.
ClosedLoop recently launched the C-19 Index, an open source predictive model that identifies people likely to have a heightened vulnerability to severe complications from COVID-19. While the C-19 Index doesn’t predict who will become infected with COVID-19, it’s intended to help hospitals, public health agencies, and other health care organizations identify, plan for, respond to, and reduce the impact of COVID-19 in their communities. ClosedLoop claims that over 10 million lives across partners including Cerner, Johns Hopkins University, and Ascension have been positively impacted by the C-19 Index.
In October, ClosedLoop — which provides predictions on over 15 million patients in the U.S. — was selected as a finalist among 300 contestants in the U.S. Centers for Medicare & Medicaid Services AI Health Outcomes Challenge. The $1.6 million dollar contest aims to spur the development of AI-driven predictions for health care providers and clinicians, with the goal of demonstrating how AI could be used to predict unplanned hospital and nursing facility admissions and adverse events for Medicare beneficiaries while explaining the predictions to clinicians and patients in ways that would aid in providing appropriate clinical resources.
This latest investment brings Austin, Texas-based ClosedLoop’s total raised to date to $15 million.
Author: Kyle Wiggers
Source: Venturebeat
