Noah Labs Vox AI Detects Heart Failure From a 5-Second Voice Clip
A five-second voice recording on a smartphone could soon alert doctors that a heart failure patient is deteriorating – weeks before they’d otherwise end up in the hospital. On March 25, 2026, the U.S. Food and Drug Administration granted Breakthrough Device Designation to Noah Labs Vox, a software-only AI tool that analyzes subtle acoustic changes in a patient’s voice to detect worsening heart failure up to three weeks before hospitalization becomes necessary.
The designation places Vox on an expedited regulatory pathway, accelerating its clearance and eventual commercialization in the United States. For the estimated 64 million people worldwide living with heart failure – including roughly six million Americans – this represents a potential shift from invasive, hardware-dependent monitoring toward something as simple as speaking into a phone once a day.
What makes Vox particularly notable isn’t just the AI itself but the scale of data behind it: the algorithm was trained on more than three million heart failure patient voice recordings, described as the world’s largest exclusive dataset of its kind. That foundation, combined with validation across more than five multicentre clinical trials at some of the world’s most respected institutions, has positioned this Berlin-based startup at the forefront of voice-based cardiac diagnostics.
How Vox Works: Voice as a Biomarker
Noah Labs Vox operates on a deceptively simple premise. Patients record a brief, standardized five-second voice clip using any commercially available smartphone or tablet. The recording doesn’t depend on what the patient says – the AI isn’t listening to words. Instead, it uses proprietary machine learning to extract and analyze acoustic features tied to physiological changes, specifically pulmonary congestion and fluid overload.
When fluid builds up around the heart and lungs, it produces subtle but measurable changes in how a person’s voice sounds. These shifts are imperceptible to the human ear but detectable by an algorithm trained on millions of samples. Vox processes each clip and generates what’s been described as a “wetness” score, flagging patients at high risk of decompensation so clinicians can intervene before a crisis unfolds.
No additional hardware is required. No implants. No external readers. Just a phone and five seconds of the patient’s time each day.
The Clinical Validation Behind the Technology
Vox’s credibility rests heavily on its clinical pedigree. The algorithm has been validated in more than five multicentre clinical trials conducted in partnership with institutions including Mayo Clinic, the University of California San Francisco (UCSF), Charité Berlin, Maastricht University Medical Center, and Hospital Clínic de Barcelona. These trials have demonstrated what the company describes as exceptional sensitivity in detecting clinical worsening weeks ahead of hospitalization.
One key supporting study is the PRE-DETECT-HF trial (registered as NCT07443969), which has provided critical data for both the FDA pathway and Noah Labs’ parallel pursuit of European Union Medical Device Regulation (EU MDR) certification. The company anticipates receiving EU approval by mid-2026 based on this trial’s results.
While specific sensitivity percentages and readmission reduction rates have not been publicly detailed, the breadth and caliber of the trial network – spanning three continents and five leading academic medical centers – lends substantial weight to the technology’s clinical promise.
What FDA Breakthrough Device Designation Actually Means
The FDA’s Breakthrough Device Designation program is reserved for technologies that address unmet medical needs in life-threatening or irreversibly debilitating conditions. It doesn’t mean the device is approved for sale. What it does is open a faster, more interactive regulatory pathway.
For Noah Labs, the designation means several concrete things: more frequent communication with the FDA during the review process, priority review of submissions, and an expedited timeline toward clearance. The company has confirmed that an FDA pivotal trial is set to launch soon, a critical step toward full commercial authorization in the United States.
Oliver Piepenstock, CEO and co-founder of Noah Labs, framed the milestone clearly: “Voice is a powerful predictor for worsening heart failure. This Breakthrough Device Designation is a pivotal milestone in Noah Labs Vox’s path to driving early access to life-saving interventions for patients with life-threatening diseases.”
The Heart Failure Monitoring Landscape
To appreciate what Vox brings to the table, it helps to understand what currently exists. The dominant approach to remote heart failure monitoring relies on hemodynamic sensors – most notably Abbott’s CardioMEMS HF System, which received FDA clearance back in 2014. CardioMEMS is a wireless sensor surgically implanted into the pulmonary artery. It tracks blood pressure and heart rate, transmitting data remotely so clinicians can spot early warning signs of cardiac exacerbations.
Abbott’s next-generation CardioMEMS Hero received FDA clearance in March 2026, and the UK’s National Institute for Health and Care Excellence (NICE) formally recommended the original CardioMEMS system for use within the NHS that same month. These are proven technologies with demonstrated outcomes in reducing readmissions.
But they require surgery. They require hardware. And they inherently limit access for the vast majority of the 64 million global heart failure patients who live in settings where invasive procedures aren’t readily available.
| Feature | Noah Labs Vox | Abbott CardioMEMS / Hero |
|---|---|---|
| Detection Method | AI analysis of 5-second voice clips for acoustic biomarkers | Implanted pulmonary artery sensor tracking blood pressure and heart rate |
| Invasiveness | Non-invasive, software-only | Requires surgical implant procedure |
| Hardware Required | Any smartphone or tablet | Implanted sensor plus external reader device |
| Accessibility | Suitable for rural and resource-limited settings | Requires clinical infrastructure for implantation |
| Regulatory Status (US) | FDA Breakthrough Device Designation; pivotal trial launching soon | FDA cleared (2014 original; 2026 Hero) |
| Head-to-Head Comparison Data | No direct comparative studies exist between the two approaches | |
It’s worth noting that no head-to-head studies have compared Vox to implant-based systems. The two approaches target different aspects of the monitoring challenge: CardioMEMS focuses on hemodynamic pressure data, while Vox analyzes vocal biomarkers associated with fluid overload. They may ultimately prove complementary rather than competitive.
The Scale of the Problem Vox Aims to Solve
Heart failure remains the leading cause of hospital admission globally. In the United States alone, approximately six million people live with the condition, and prevalence is projected to increase by 46% by 2030. Direct medical costs are expected to reach $53 billion over that same period.
Current monitoring gaps are significant. Many patients lack access to reliable early warning systems, either because invasive tools aren’t available in their region or because regular in-person visits aren’t feasible. The result is a cycle of preventable hospitalizations – patients deteriorate at home without anyone knowing until they arrive at the emergency department in acute decompensation.
Vox’s software-only design directly targets this gap. A patient in a rural community with a smartphone has the same access to the technology as one at a major academic medical center. That scalability is central to the device’s value proposition and a key reason the FDA’s Breakthrough program applies – the technology addresses an unmet need for a life-threatening condition.
Practical Use and Best Practices for Clinicians and Patients
While Vox is not yet commercially available in the United States, its operational framework has been outlined through trial protocols and company disclosures. For clinicians evaluating the technology and patients who may eventually use it, several practical considerations stand out.
- Recording protocol: Patients speak a brief, standardized phrase for exactly five seconds. Clips that are shorter or longer can skew the acoustic feature extraction, so adherence to the precise duration is essential.
- Daily consistency: Recordings should be submitted once per day, ideally at the same time each morning. Consistent timing helps the algorithm detect subtle trends that might otherwise be masked by normal daily vocal variation.
- Audio quality: Background noise can obscure the biomarkers the AI is looking for. Patients should record in a quiet room, holding the device approximately six to eight inches from their mouth.
- Compliance matters: Skipping days interrupts the monitoring pattern. The algorithm’s ability to detect deterioration up to three weeks before hospitalization depends on continuous daily data.
The system integrates with Noah Labs’ existing Ark platform, which is already deployed in European clinical sites for broader heart failure management. Vox functions as an upgrade within that ecosystem, adding voice-based predictive capability to existing remote monitoring workflows.
Expert Perspectives on Clinical Impact
The clinical community has responded to the FDA designation with measured optimism. Prof. Felix Hohendanner, MD, PhD, from Charité Berlin – one of the trial partner institutions – offered a pointed assessment: “The FDA Breakthrough Device Designation for Noah Labs Vox is a significant achievement for heart failure care. Early detection of clinical worsening is a central challenge in managing these patients – the ability to identify deterioration weeks before hospitalization, using nothing more than a patient’s voice, has the potential to fundamentally change how we intervene and prevent adverse outcomes.”
That phrase – “using nothing more than a patient’s voice” – captures the essential disruption here. The barrier to entry for monitoring drops from a surgical procedure to a daily habit that takes less time than brushing your teeth.
Noah Labs itself occupies a growing institutional footprint. The company is a member of the American Heart Association’s Centre for Health Technology and Innovation, and it has secured up to €11 million from the European Innovation Council (EIC) Accelerator for voice-based diagnostics. Additional EU and Brandenburg state funding supports the company’s “Pre-Sense” project, which analyzes multiparametric data for early cardiac decompensation detection.
Market Context and What Comes Next
The global patient monitoring market is valued at approximately $21.4 billion in 2025 and is projected to reach $25.9 billion by 2035, growing at a compound annual growth rate of 1.9%. AI-driven remote monitoring tools like Vox are a significant driver of that growth, particularly as healthcare systems worldwide push toward value-based care models that penalize preventable readmissions.
For Noah Labs, the immediate roadmap includes launching the FDA pivotal trial, completing EU MDR certification by mid-2026, and scaling commercial deployment. In the United States, the Breakthrough Device Designation provides meaningful acceleration, but full clearance still requires rigorous trial data and regulatory review.
The broader trajectory is clear: non-invasive, AI-powered diagnostics are moving from research curiosity to clinical reality. Voice-based biomarkers represent a frontier that extends well beyond heart failure – the same underlying principle that fluid overload changes vocal acoustics could potentially apply to other conditions affecting respiratory and cardiac function. For now, though, the focus is singular: keeping heart failure patients out of the hospital by catching deterioration early, one five-second voice clip at a time.
Key Takeaways
- Noah Labs Vox received FDA Breakthrough Device Designation on March 25, 2026, for detecting worsening heart failure up to three weeks before hospitalization using a daily five-second voice recording.
- The AI was trained on over three million heart failure patient voice samples and validated in more than five multicentre trials with partners including Mayo Clinic, UCSF, and Charité Berlin.
- Vox is entirely software-based – no implants, no hardware beyond a smartphone or tablet – making it accessible in rural and resource-limited settings.
- An FDA pivotal trial is launching soon, with EU MDR approval anticipated by mid-2026.
- Heart failure affects 64 million people globally and six million Americans, with U.S. direct medical costs projected to hit $53 billion by 2030.
- No head-to-head studies currently compare Vox to implant-based systems like Abbott’s CardioMEMS, though the approaches may prove complementary.