Respirix for Non-Invasive Cardiopulmonary Monitoring: Interview with CEO Eric Kriegstein

For patients with cardiopulmonary illnesses such as chronic obstructive pulmonary disease (COPD or congestive heart failure (CHF), few technologies exist to effectively monitor and manage their conditions in a convenient way.

Pulmonary artery catheters and implants, that can help with monitoring, have successfully reduced heart failure hospitalizations, but these implants are highly invasive and costly.

Respirix is developing a non-invasive device that aims to detect CHF decompensation and COPD exacerbation before patients experience symptoms. Cardiospire works by mining non-invasive physiological data gathered from what looks like a simple spirometer.

The handheld device actually has ECG, SPO2, spirometry, temperature, and airway pressure sensors. After pairing Cardiospire wirelessly with a smartphone, patients are guided through several breath measurements in under five minutes during which time all the data is collected simultaneously. These measurements are securely transmitted from the smartphone to Respirix servers and the patient’s care provider. Readings that suggest potential problems are flagged for attention and further management.

We asked Respirix CEO Eric Kriegstein more about Cardiospire.

Cici Zhou, Medgadget: Tell me about how Respirix started. What was the inspiration behind the idea and product?

Eric Kriegstein: Respirix started by accident. The scientific founders Dan and Evan were undertaking an unrelated large animal study, and they were measuring airway pressure. They noticed that there was a tiny perturbation in the rough airway pressures of the intubated pig that had the same frequency as the ECG; upon further investigation they learned this signal was a phenomenon called a cardiogenic oscillation, which has been shown in research to be caused by pulmonary blood flow to the lungs. Little work had been undertaken to explore the utility of this signal, and Respirix was born from the belief that we could take this overlooked waveform and use it to extract valuable physiological data.

Medgadget: How does the technology work?

Kriegstein: Our platform involves acquiring data non-invasively through commodity sensors and employing proprietary algorithms to interpret these data streams for diagnostic, and prognostic information about patient health. We have developed an elegant handheld device, called the Cardiospire, that enables the acquisition of 6 clinical-grade data streams in a single sitting; these are securely transmitted to our cloud where we look for potentially problematic trend suggestive of patient deterioration. Ultimately, our goal is to demonstrate that we can guide medication changes to patients suffering from cardiopulmonary illness to avert costly and traumatic hospitalizations for indications such as CHF, COPD and pneumonia.

Medgadget: At what stage is the company today? (Number of team members, stage of development and clinical studies, investment funding, etc.)

Kriegstein: We are operating right now with a very small team; we’ve been able to leverage the resources of the TheraNova incubator to drive clinical, engineering, R&D and regulatory development to date. The company has raised $3.3 million in seed investment since inception, and we have been awarded an SBIR Phase I grant from NHLBI, and a CAPCaT grant from UMass M2D2. We have completed two clinical trials, three preclinical trials included a study sponsored by Bayer at their headquarters, and are currently undertaking two new clinical trials at UCSF, KUMC and Narayana Health in Bangalore.

Medgadget: What are some of the biggest challenges currently facing the team?

Kriegstein: As a company, the research we are undertaking is novel and proprietary. What we quickly realized is, we don’t have the luxury of going out and buying large datasets to begin training our ML models, like a startup detecting arrhythmias from ECG can, rather we need to go out and acquire our data in clinical and preclinical studies. The acquisition of our unique set of signals, in conjunction with ground truth measurements often inside a patient or animal’s heart, is a challenge we’ve had to solve as a small startup. But we’ve learned this is also a significant competitive moat, because it is not easy to acquire these datasets let alone build the algorithms from them as we have done.

Medgadget: Looking to the next 5 years, what are some of the major goals for the company?

Kriegstein: A number of very important milestones will be achieved in that timeframe such as FDA clearance of our device, healthcare network customers signed, patients monitoring their daily vitals and sharing those measurements with their care providers, and the completion of large scale, outcomes-based studies demonstrating therapeutic changes guided by our algorithms can interdict the costliest, and most traumatic, hospitalizations experienced by patients in the U.S. every year.