Seven years ago, health care technology expert Adam Hanina, M.B.A., and his colleagues recognized a major problem in clinical trials and outpatient care settings: Too many individuals take their medications incorrectly or do not take them at all. This can reduce the accuracy of clinical studies, make drug development less efficient and be detrimental to patient health.
To combat this problem, Hanina joined forces with Laura Shafner, M.Sc., a health policy expert with a finance background, and Gordon Kessler, J.D., M.B.A., an intellectual property attorney. They set out to develop an artificial intelligence smartphone application that would visually confirm medication ingestion. In other words, it would help ensure the right patient takes the right medication at the right time. To develop and test the app, the team’s startup company, AiCure, received a Small Business Innovation Research (SBIR) Phase 1 grant in 2011 from the National Center for Research Resources (NCRR), which, under congressional authorization, reorganized into NCATS later that year.
The SBIR and Small Business Technology Transfer (STTR) programs help NCATS speed innovation in translational science. Often, small startup companies lack the early-stage funding to carry an idea through further investment and into commercial markets, where new interventions can be developed and delivered to patients. The NCATS SBIR and STTR programs bridge this translational gap by awarding grants, contracts and technical assistance to small businesses and research organizations focused on advancing translational research and technologies that will improve disease prevention, detection and treatment.
“We had a potentially revolutionary idea to reduce risk in clinical research and improve patient health, and we needed the resources to develop it to the point of commercialization,” Hanina said. “NCATS’ goals align with our company’s mission: to improve translational science and accelerate the drug development process to get high-quality therapies into the hands of more patients more quickly.”
Tackling the Adherence Problem
Hospitalizations and poor health outcomes due to medication nonadherence cost the U.S. approximately $290 billion each year. In addition, poor adherence prevents researchers from properly assessing a drug in clinical trials, which contributes to a high failure rate. Clinical trials can cost $1 billion or more, and they most often fail because researchers cannot prove a therapy’s effectiveness or safety. These pitfalls ultimately can be harmful to patient health and quality of life.
Although researchers already have several options for estimating adherence, such as pill counting and medication containers that record and track administration, these methods cannot confirm actual intake. Direct observation of patients is costly, invasive and impractical, and self-reporting of medication behaviors by patients can be imprecise and unreliable. The AiCure group devised the artificial intelligence adherence technology to address these translational barriers and provide a more efficient and accurate way to measure adherence.
“Most existing adherence monitoring methods are geared toward either clinical research or clinical practice but not both,” Shafner said. “Our technology is designed to fill this gap by providing a gold standard method to measure data in a standardized way across all settings.”
Automating Direct Observation
With the initial SBIR funding, the AiCure team hired an artificial intelligence engineer to develop the monitoring app. Using a mobile device’s camera, software algorithms confirm the identities of the patient and the medication and verify intake. The video to the right describes how the app works. The app sends this information to a cloud-based dashboard that researchers or health care providers can use to monitor adherence and identify issues in real time. Providers can communicate with patients through the dashboard to offer immediate assistance. The application also gives interactive instructions, reminders and suggestions to patients to further increase adherence. For now, the technology monitors pill intake, but it can be adapted to recognize any form of administration, such as injections or oral liquids.
“The AiCure application acts as an intelligent, mobile medical assistant to guide a patient through the health care journey, providing improved quality of care that would have been otherwise inaccessible,” Hanina said.
With the Phase 1 grant, the AiCure team successfully demonstrated that the platform was technically feasible and able to confirm that patients have taken medication. Based on this success, NCATS awarded AiCure a Phase 2 SBIR award in 2013. This funding enabled validation of the technology against blood levels of medications and showed that the app improves adherence rates in schizophrenia and stroke patient populations. The group expects to publish study results later in 2016.
An Expanding Enterprise
“The initial two SBIR awards were instrumental to the development of our company and have led to additional NIH grants,” Hanina said. “Together, the NIH support has enabled our company to attract and leverage an additional $12.25 million in financing from venture capital investors.”
AiCure also has entered contracts with five of the top 12 global pharmaceutical companies to provide the adherence monitoring app for clinical trials of experimental drugs. One of these companies, Takeda Pharmaceuticals U.S.A., Inc., is testing the technology in a clinical trial for patients with psychiatric illnesses.
“Noncompliance is a major issue in clinical trials,” said Atul R. Mahableshwarkar, M.D., senior medical director at Takeda. “The AiCure platform may help improve our ability to detect a drug’s effectiveness as well as help us reduce costs by decreasing the number of patients needed in studies.”
In addition to their work with the drug development industry, the AiCure team is continuing to test the platform in NIH studies of substance abuse and is partnering with government organizations and insurance companies on population health contracts in infectious and cardiovascular disease. Ultimately, the team envisions deploying the technology globally to assist in care for high-risk patients, for whom missed doses could lead to serious outcomes such as hospitalization or death.
“In keeping with NCATS’ disease-agnostic philosophy, the SBIR awards facilitated the development of the AiCure technology and approach for translation across every therapeutic area and route of drug administration,” said Lili Portilla, M.P.A., director of strategic alliances at NCATS and program lead for SBIR and STTR. “In this way, the technology holds promise for generating a multiplier effect of benefits in translational research and for patients.”
Posted August 2016
AiCure’s artificial intelligence smartphone application, described in the story below, was recently evaluated in an NCATS-supported clinical trial of adults with ischemic stroke. The study was designed to test whether using the application — which visually confirms medication ingestion — would improve adherence in patients taking anticoagulation therapy to help prevent blood clots. As published April 6 in Stroke, the researchers found that patients who used the AiCure application had a 50 percent improvement in adherence.