FDA to issue leapfrog guidance on brain-computer interface devices

The Food and Drug Administration (FDA) will issue a guidance document on May 20, 2021, regarding non-clinical testing and clinical considerations regarding brain-computer interface (BCI) devices. Implanted BCI devices are neuroprostheses that interface with the patient's nervous system to restore lost motor and/or sensory capabilities for patients with paralysis or amputation. The guidance document is a leapfrog guidance, intended to share initial thoughts regarding emerging technologies early in product development, and is nonbinding.

 

The guidance document describes information that should be included in submission for an investigational device exemption (IDE). Along with detailed technical specifications, the FDA recommends that the applicant submit risk analysis detailing potential hazards of the device from the perspective of the user; description of the software and hardware supporting operation of the device, including information on cybersecurity aspects; and potential use-related hazards that could result from the user interface design.

 

The guidance document also includes comment on clinical study considerations. It states that implanted BCI devices are likely to be significant risk devices subject to all requirements of the IDE regulation. The guidance suggests design components which should be considered when developing the clinical study protocol, including implications relating to studying BCI devices in realistic home use environments that may require assessing caregiver safety and what training caregivers will require to assist the patient. The FDA recommends a long-term follow-up period of at least a year, due to the current lack of data on long-term effectiveness of implanted electrodes. Also, in addition to safety endpoints and effectiveness endpoints, patient input may be an important factor in evaluating the device. This would include patient preference information (such as whether the device is comfortable and user friendly) and patient-reported outcomes, such as pain reduction.

AHRQ announces funding opportunity for AI, digital health and clinical decision support

The Agency for Healthcare Research and Quality (AHRQ) announced a funding opportunity for research projects that test digital healthcare interventions aimed at improving quality at the point of care. The digital healthcare solutions tested should use advanced analytics; patient-centered clinician and patient facing digital technologies; or clinical decision-making tools. Applications are due June 16, 2021. For the first two-year phase, the project budget may not exceed $275,000. Projects continued into the second three-year phase may have a total project budget not exceeding $750,000.

 

For the research area of interest involving advanced analytics, the AHRQ suggested examples such as projects integrating AI during the provision of healthcare services and assessing AI's impact on practice workflow and quality of care, or projects applying machine learning against large health data sets to improve quality at the point of care. Research involving patient-centered digital healthcare technologies could include projects testing innovative technologies that facilitate information sharing between patients and providers, or the use of novel digital solutions for patients with multiple chronic conditions. Examples of projects focused on clinical decision-making support would be projects evaluating a digital point of care solution combining natural language processing with a decision support tool, or projects testing decision-making tools that incorporate the use of patient-generated data and patient-reported outcomes at the point of care.

Cleveland Clinic study finds smartphone ECG can be useful for monitoring AF

While availability of direct-to-consumer devices that enable the user to record heart rhythms continues to grow, use of the devices in clinical practice remains low. One reason for limited clinical use is that most devices allow the user to share an electrocardiogram (ECG) recording with their healthcare provider as an email attachment. It is impractical for busy clinicians to incorporate this raw data into normal workflows.

A small Cleveland Clinic study evaluated the use of the Kardia Mobile smartphone monitor coupled with the Kardia Pro (KP) platform for follow-up of patients following successful atrial fibrillation (AF) ablation. The KP platform utilizes artificial intelligence to triage preliminary interpretation of ECG tracings and route suspected abnormal recordings to the electrophysiologist’s in-basket. The study randomized 100 patients who presented three to four months after successful AF ablation into a self-monitoring group using the Kardia Mobile monitor, and a control group following normal standard of care. Healthcare utilization and anxiety were similar for the two groups, but more patients in the control group required additional ECGs or cardiac monitors compared to the self-monitoring group.

The study authors concluded that the KP platform can be effectively incorporated into the care of patients to assist in detection of AF recurrences. They noted:

The synergistic relationship between instantaneous interpretation via the automated algorithm, digital platform, patient, and healthcare provider is key for successful adoption of digital technology into busy clinical practices and will turn the technology into an asset rather than a burden and transform the relationship between the AF patient and the physician into a partnership rather than a unidirectional process.

Stanford research study: Apple Watch useful in assessing frailty in CV patients

A recent Stanford University study compared the effectiveness of the Apple Watch in assessing frailty in cardiovascular patients, in comparison with an in-clinic six-minute walk test (6MWT). The study enrolled 110 patients who were scheduled for vascular or cardiac procedures at a Veterans Administration facility, supplied them with an iPhone and Apple Watch running the VascTrack research application, and followed them for six months. Periodic in-clinic 6MWT results were compared with in-home 6MWT performed using the app, and passive step information collected by the Apple Watch. The tests assessed frailty, defined as walking fewer than 300 meters on an in-clinic 6MWT. The Apple Watch was nearly as effective in assessing frailty via the 6MWT in the home setting as in the clinic setting (83% sensitive and 60% specific in assessing frailty in the home setting, versus 90% sensitive and 85% sensitivity in the clinic setting). Passive data collected at home was nearly as accurate as the 6MWT in-home assessment.

ONC announces finalists for synthetic data research

The Office of the National Coordinator for Health Information Technology (ONC) has announced the Phase I finalists for development of innovative models using Synthea, an open-source synthetic patient generator that models the medical history of synthetic patients from publicly available sources, such as health statistics. The finalists will have until July 13, 2021 to submit their prototypes. ONC had invited proposals on enhancements to Synthea, and novel uses of synthetic data generated using Synthea. The nine finalists are working on a variety of solutions, including incorporation of social determinants of health (SDOH) data to predict diabetes progression and big data analysis of the opioid epidemic in Illinois.

APIs for clinical research: future promise, constraints on current utility

The Office of the National Coordinator for Health Information Technology (ONC) released a report in March 2021 summarizing perspectives of researchers on the use of application programming interfaces (APIs) for clinical research. Currently, extracting data required for research from the electronic health record (EHR) or data warehouse, or curating data to conform to a common data model (CDM) across multiple sites, are complex and resource-intensive activities. The ONC interviewed researchers to evaluate whether standardized APIs may reduce dependence on custom data extracts and complex data curation. While increased use of standardized APIs may be useful to streamline research efforts across organizations and support multi-site trials, there are several barriers to broader use of APIs in clinical research studies at the present time.

Read more ...

AHRQ seeks public comment on clinical algorithms influencing bias

The Agency for Healthcare Research and Quality (AHRQ) is seeking public comment on clinical algorithms in medical practice that could have the effect of introducing bias or influencing access to care, quality of care or health outcomes for minorities or socioeconomically disadvantaged individuals. Comments can be entered directly on the AHRQ’s website up to May 4, 2021.

The AHRQ lists 11 questions inviting comments on topics such as algorithms that use race or ethnicity as a variable; use of social determinants of health (SDOH) in clinical algorithms; which clinical algorithms have evidence indicating contribution to health disparities; and what are developing standards on how to develop, validate or update standards to avoid bias.

View LinkedIn Profile