Remote biometric monitoring technology company Biotricity recently announced securing National Institutes of Health funding from the National Heart, Blood and Lung Institute with plans to launch a study of its Bioflux-AI technology.

Bioflux-AI combines an FDA-approved, high-precision, small mobile cardiac telemetry device with AI-driven algorithms specifically trained for the prediction of stroke in stage 4 and stage 5 chronic kidney disease patients.

The vendor continues to deliver remote patient monitoring technologies to physicians and cardiac patients with a suite of wearable, real-time active cardiac monitoring devices. In recent 1Q23 financial results, the company noted expansion of coverage across the U.S. with 2,000 physicians across 29 states integrating the company's remote patient monitoring devices in their practice.

Healthcare IT News sat down with CEO Dr. Waqaas Al-Siddiq to better understand stroke prediction in CKD patients and advances in remote cardiac monitoring.

Q. Please describe what remote cardiac monitoring entails.

A. Remote cardiac monitoring focuses on non-invasive portable devices worn by patients to continuously collect personal ECG data (the electrical activity of the cardiovascular system) that can help diagnose arrhythmias.

Remote cardiac monitoring is available in multiple forms: Holter monitoring, extended Holter monitoring and mobile cardiac telemetry. The concept of active monitoring, where data is being reviewed in real time, is only possible with mobile cardiac telemetry.

Neither type of Holter monitoring is capable of active monitoring. This can be misleading to consumers because the word "monitor" is in the title, but these devices lack connectivity and only record a patient's ECG for analysis later.

Holter technology was originally designed to capture a person's ECG for 24 to 48 hours. Upon completion of the patient's study, the device was returned, and the data was downloaded and sent for analysis.

Extended Holter technology, developed recently, allows for up to 21 days of recording with most extended Holters being used between seven and 10 days. Extended Holter technology was created because heart issues are often intermittent, making one to two days of recording insufficient to catch these intermittent anomalies.

However, for higher risk patients, recording for longer periods of time carries a risk because it will take far longer for the patient to receive a diagnosis. Consequently, mobile cardiac telemetry was created to deliver active monitoring for longer periods with no patient risk.

Cardiac telemetry is the most beneficial of remote cardiac monitoring because it is real-time cardiac monitoring. For this form of monitoring, the devices are smart, have cellular connectivity, and not only record a patient's ECG, but also analyze it in real time, looking for outliers.

When an outlier is detected, the device transmits this information to a call center along with the data in question for intervention, if applicable. This real-time data transmission and analysis reduces the time to diagnosis and improves patient care, enabling early intervention that can be lifesaving. There is no question that an at-risk patient should be on cardiac telemetry when it comes to remote cardiac monitoring.

Q. Your company recently secured NIH funding from the National Heart, Blood and Lung Institute with plans to launch a study of your Bioflux-AI. Please elaborate on what Bioflux-AI is and what the study will cover and try to prove.

A. Bioflux-AI is an AI engine that we are currently developing to enhance our algorithm-based arrhythmia detection and build out predictive capabilities. Currently, we use device-based algorithms that we have developed to perform arrhythmia detection.

To move into predictable monitoring, we are developing Bioflux-AI to further analyze data, post-device-level analysis. To date, we have tracked more than 220 billion heartbeats with 2 billion heartbeats annotated and analyzed by specialists for atrial fibrillation (AFib).

Because AFib can be difficult to diagnose, our focus has been on detection, with a specific emphasis on AFib in combination with other chronic conditions such as kidney disease or sleep apnea. Our first goal with Bioflux-AI is to develop a complete and predictive AFib detection capability across different chronic diseases, after which we will begin focusing on other arrythmias.

The focus of this study is to look at prediction and detection of AFib in kidney patients. More than half of dialysis and chronic kidney disease patients die of heart disease as opposed to kidney disease. Often it is a result of undiagnosed arrhythmias such as AFib.

Our focus in this study is to collect data from kidney patients to develop custom algorithms for the detection of AFib in these patients. The goal is to develop a predictive algorithm that can identify arrhythmias, or the potential occurrence of an arrhythmia, early enough to support early intervention.

Q. How important is AI to this type of telemedicine/remote patient monitoring?

A. AI is incredibly important and a necessity when it comes to the longer monitoring periods. Shorter periods of monitoring like one to five days are manageable without AI because the data stream is smaller.

Longer studies entail more data, which makes it much more difficult to sort through and analyze without the implementation of AI. I imagine a future where our biometrics are tracked not for days, but for months and years. For studies that long, there would be no choice but to use AI to sift through and identify the areas that are important for review.

Without AI capable of sifting through data to identify areas to focus on, long-term monitoring will become too cumbersome for healthcare professionals, making adoption highly unlikely. Traditionally, Holter technology was restricted to two days because of the issue of too much data.

With extended Holter technology and mobile cardiac telemetry, the increased duration was supported by service providers or smart devices, respectively, to analyze and review data for healthcare professionals. Without this support, adoption would have been incredibly difficult.

AI can take cardiac monitoring one step further. Because heart issues are the leading cause of death for both men and women and are intermittent by nature, the longer the data stream, the better the diagnostic yield.

Most important, for long-term patient support, continuous long-term heart rhythm recording is needed to enable patient engagement and management. Just like with diabetes, where continuous glucose monitoring supports patients in managing their condition, continuous heart rhythm recording is needed to support cardiac patients in managing their condition.

This level of data requires AI or deep data algorithms for data analysis otherwise the amount of data will be impossible for healthcare professionals to manage. This issue is unique to the electrical activity of the heart as the data for analysis is different when compared to glucose monitoring or blood pressure monitoring.

The latter are high-level numbers that have ranges for normal versus abnormal. Heart rate would be the equivalent parameter to these. In this case, we are talking about electrical heart activity, which is unique to patients and the numbers or measurements that are being analyzed relate to multiple measurements related to the signal, how that signal is visually represented, and how often they occur.

With electrical heart activity, one type of event can be considered benign, but multiple occurrences of the same event in succession is considered very high risk. This complexity and multivariate analysis requires AI when it comes to electrical heart rhythm recording.

Similarly, as telemedicine and remote patient monitoring expand to include multiple biometrics and data streams, the complexity of analysis will increase. Cross referencing data streams will require deeper analysis and risk will vary patient to patient, depending on their underlying condition or conditions.

As the number of variables increase and the access to data improves, the need for AI and the importance of AI will increase. Without it, healthcare professionals will simply be unable to sift through the data.

Q. Why is remote cardiac monitoring important to the future of healthcare delivery?

A. There are three critical reasons that underscore the importance of remote cardiac monitoring. First, heart disease is the number one killer in the world, and in order to help prevent many of these deaths it is necessary to remotely monitor patients to support earlier diagnosis.

Second, heart issues are intermittent and often have no symptoms, requiring data collection over a long period of time to understand what is occurring. And third, chronic disease is the No. 1 expense for the healthcare system because patients are diagnosed late and their condition requires them to be in and out of the hospital often, as opposed to the lower cost alternative of remote diagnostics.

Remote cardiac monitoring is instrumental in all these scenarios.

Additionally, many other chronic conditions increase a patient's cardiac risk, further necessitating the need for remote cardiac monitoring. Increased heart disease risk presents itself in a variety of areas such as, but not limited to, patients with other conditions and patients on different medications.

Over time and as lifespans increase, so does cardiac risk, driving a growing need for remote cardiac monitoring. Remote cardiac monitoring supports remote care and delivers insights that support better management and early intervention. Uniquely, it is a technology that delivers value and multiple advantages to both patients and providers.

Twitter: @SiwickiHealthIT
Email the writer: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication.