Medibyt has received U.S. Food and Drug Administration 510(k) clearance for CPETWise, allowing the Israeli digital-health company to begin U.S. commercialization of software designed to automate and standardize cardiopulmonary exercise test analysis and reporting. FDA records identify the platform under K260242, with the agency reaching its substantial-equivalence decision on August 20 after receiving the submission in January. The device is classified within pulmonary-function predicted-value calculator software under the anesthesiology review panel, and the FDA database confirms that no Predetermined Change Control Plan was authorized as part of this clearance.
The commercial argument is centered on a bottleneck rather than a new physiological measurement. Cardiopulmonary exercise testing can simultaneously capture oxygen consumption, carbon dioxide production, ventilation, heart-rate response and other exercise physiology, but interpretation requires specialized expertise and can be time-consuming. Medibyt says a 100-case proof-of-concept evaluation at the University of Washington produced an average user-satisfaction score of 4.5 out of 5 and reduced interpretation time by more than 50%, while CPETWise preserved clinician access to raw data and standardized graphical displays.
Why can cardiopulmonary exercise testing reveal problems that resting tests may miss?
Many cardiovascular and pulmonary conditions become physiologically apparent only when demand increases. A patient may have relatively normal oxygen saturation, heart rate and ventilation while sitting in a clinic but develop abnormal responses when muscles require much greater oxygen delivery during exercise. CPET measures several systems simultaneously while workload rises, helping clinicians understand whether exercise intolerance is driven primarily by cardiovascular limitation, pulmonary mechanics, impaired oxygen extraction, deconditioning or another physiological pattern.
This is one reason CPET can be valuable in heart failure, pulmonary disease, unexplained shortness of breath, preoperative assessment and exercise medicine. The challenge is that interpretation is multidimensional. Clinicians must assess relationships among oxygen uptake, workload, ventilation, carbon dioxide production, anaerobic threshold, heart-rate response and numerous derived variables rather than relying on one simple positive-or-negative laboratory result.
What does CPETWise automate without taking the clinician out of the interpretation loop?
Medibyt describes CPETWise as a workflow and analysis platform that ingests complex physiological data, performs standardized calculations and produces graphical panels and structured reports. The company says the system can identify the principal physiological factor limiting performance while allowing clinicians to review the raw underlying measurements rather than presenting an unexplained black-box conclusion.
That visibility matters in regulated medical software. Automated analysis can reduce repetitive calculation and formatting work, but a physician still needs the ability to examine whether an algorithm’s conclusion makes sense in the context of the patient, test quality and clinical history. Medibyt’s emphasis on retaining the underlying data therefore positions the platform as interpretation support and workflow standardization rather than an autonomous replacement for cardiologists, pulmonologists or exercise physiologists.
What evidence did Medibyt report before moving into U.S. commercialization?
The company says it completed a 100-case proof-of-concept study at the University of Washington in which users reported average satisfaction of 4.5 out of 5 and CPET interpretation time fell by more than 50%. Medibyt also says the workflow provides immediate access to raw data, standardized graphical panels and automated identification of the primary limiting factor. These are encouraging operational findings but should be recognized as company-reported proof-of-concept results rather than a randomized outcomes trial showing improved patient survival or diagnostic accuracy.
Demonstration environments have also been deployed at Columbia University, the University of Washington, UCLA, Sheba Medical Center and Ichilov Medical Center, among other institutions. The company has begun a commercial pilot with a U.S. CPET business to integrate data transfer between existing testing equipment and CPETWise. Those deployments can provide valuable workflow experience as the product moves from regulatory clearance toward routine use.
Why could standardization matter almost as much as interpretation speed?
A sophisticated diagnostic test loses value if interpretation differs substantially according to who reads it. CPET is particularly vulnerable because many measurements interact, and expert physicians may weigh patterns differently depending on training and experience. A structured software workflow can potentially reduce variation by ensuring the same calculations, reference values and graphical relationships are considered consistently.
Standardization could also make remote interpretation easier. A smaller hospital might be capable of performing a technically adequate CPET but lack an on-site specialist comfortable analyzing every component. A connected workflow could allow data to be reviewed by experts elsewhere while the local center retains test capability, potentially expanding access without requiring every facility to build a full specialist program from scratch.
Does FDA clearance mean CPETWise’s future machine-learning functions are also cleared?
No. The FDA record applies to the cleared CPETWise configuration reviewed under K260242. Medibyt separately says it is developing next-generation machine-learning capabilities using access to approximately 4,000 retrospective CPET records interpreted by expert clinicians, but future functionality must remain consistent with the cleared device or undergo whatever additional regulatory process is required. The FDA database specifically indicates that a Predetermined Change Control Plan was not authorized for K260242.
This distinction is increasingly important in medical AI because companies often discuss future algorithmic capabilities alongside a current FDA clearance. The fact that one software version is cleared does not automatically authorize every subsequent AI model, new diagnostic classification or expansion into a different type of test. Medibyt has also discussed future work in spirometry and BodyBox pulmonary-function testing, but those plans should be treated as development ambitions rather than part of the current clearance.
Could faster interpretation make CPET more commercially viable for hospitals?
Potentially. Diagnostic departments have to consider not only the clinical value of a test but how much specialist time it consumes and how many patients can move through the service. If interpretation falls from a long expert workflow to a substantially shorter clinician-reviewed process, hospitals may be able to increase throughput without proportionally increasing specialist staffing.
Medibyt’s company website promotes an even more aggressive efficiency vision, describing automated analysis and reporting within minutes and compatibility with data outputs from multiple metabolic testing systems. Those claims will need to be evaluated in real-world deployment, where difficult cases, poor data quality and complex disease combinations can require more manual review than ideal demonstrations suggest.
What are the main risks when automating interpretation of exercise physiology?
The first is overconfidence. Exercise physiology can be affected by medications, effort, equipment calibration, anemia, obesity, musculoskeletal limitations and multiple simultaneous cardiopulmonary conditions. A system that assigns one “primary limiting factor” may simplify a case that is genuinely multifactorial, which is why clinicians must retain visibility into raw and derived measurements.
The second risk is dependence on standardized reference equations that may perform differently across populations. Software must also handle technically inadequate tests and unusual physiological patterns without generating authoritative-looking reports that exceed the quality of the source data. FDA clearance establishes a regulatory basis for U.S. marketing, but post-commercial experience will reveal how frequently clinicians need to override or edit the platform’s automated interpretation.
What should hospitals watch as CPETWise enters the U.S. market?
Real-world efficiency will be the most immediate test. A greater-than-50% reduction in interpretation time could be meaningful if reproduced across institutions without increasing missed abnormalities or inappropriate conclusions. Integration with existing CPET hardware and hospital information systems will also be important because workflow software becomes much less attractive if staff must manually transfer files or duplicate documentation.
The broader significance is that CPETWise targets a category of clinical AI that may be easier to adopt than fully autonomous diagnosis. It takes a test physicians already trust, automates repetitive analytical work and keeps the clinician in control of the final interpretation. If Medibyt can reproduce its reported time savings at scale, the company may demonstrate that some of the most commercially useful medical AI products will not be those that replace physicians, but those that make specialist expertise easier to deploy.
