DeepRESP
Overview
DeepRESP is a cloud-based AI medical device developed by Nox Medical Ehf, designed to revolutionize sleep diagnostics. It leverages deep learning to accurately score 30-second epochs in sleep states (REM, NREM, wake) and arousal events directly and solely from respiratory inductance plethysmography (RIP) belt signals. This innovative approach eliminates the need for traditional electroencephalography (EEG), electrooculography (EOG), or electromyography (EMG) signals for sleep staging and arousal scoring.
The tool significantly improves the accuracy of the Apnea-Hypopnea Index (AHI) for home sleep tests (HSATs), bringing them closer to the accuracy of polysomnography (PSG) results. DeepRESP is intended to assist medical professionals in the diagnosis and follow-up of various sleep disorders in adults (22 years and above). It can be utilized in diverse clinical environments, including hospitals, ambulatory clinics, and home sleep testing settings.
Beyond automating routine sleep staging, DeepRESP provides a comprehensive scoring solution by analyzing subtle changes in breathing patterns, marking events, and calculating key clinical parameters. Its cloud-based Software as a Medical Device (SaMD) architecture includes a web API for seamless integration with client applications.
Reviewed by Pouyan Golshani, MD — Interventional Radiologist
Key Features
- Cloud-based AI medical device
- Uses deep learning to score sleep states (REM, NREM, wake)
- Scores arousal events directly from RIP belt signals
- Analyzes respiratory inductance plethysmography (RIP) belt signals
- Does not require EEG, EOG, or EMG for sleep staging and arousal scoring
- Improves AHI accuracy for home sleep tests (HSAT)
- Automates sleep staging and arousal scoring in HSAT
- Calculates key clinical parameters like Apnea-Hypopnea Index (AHI)
- Offers a web API for integration with third-party client applications
- Aids in diagnosis and follow-up of sleep disorders in adults
Use Cases
- Diagnosis of various sleep disorders in adults
- Follow-up of sleep disorder treatment
- Enhancing the accuracy and conclusiveness of Home Sleep Apnea Testing (HSAT)
- Automating sleep study analysis and scoring
- Improving detection of sleep apnea, particularly in women, from home testing
- Use in hospitals, ambulatory clinics, and home sleep testing environments
What Physicians Need to Know
DeepRESP is an FDA-cleared and CE-marked software-as-a-medical-device (SaMD) that provides automated analysis of sleep study data. It is crucial for clinicians to review, edit, validate, and finalize all AI analysis results prior to diagnosis. The system provides detailed annotations for sleep stages, arousals, and respiratory events, along with key clinical indices like AHI. DeepRESP's models are not adaptive continuous learning algorithms, ensuring consistent and validated performance. It can be particularly beneficial in improving diagnostic accuracy for populations often underdiagnosed by traditional HSAT, such as women, younger patients, and non-obese individuals.
DeepRESP is an API-centric, cloud-based software system that operates without a graphical user interface (GUI). It integrates with third-party client applications for data processing, visualization, reporting, and result dissemination. It accepts sleep study signals in .ndf and .protobuf formats. The client application is responsible for providing the user interface, managing sleep study and protocol selection, presenting DeepRESP results, and handling exceptions.
Details
| Category | Neurology AI, Wearables & Remote Monitoring (RPM) |
| Pricing | Unknown |
| Deployment | On-premise (devices and software), potentially cloud-connected for data management (Nox Connect, Noxturnal Software System) |
| Data Export | Unknown |
| Training | Unknown |
| Compliance | |
| BAA Available | Unknown |
| HIPAA Compliant | Unknown |
| FDA Status | Cleared AI-estimated |
| Integrations | |
| EHR | Not specified |
| Specialties | Pulmonology, Sleep Medicine |
Support & Reliability
| Training Provided | Unknown |
What the Web Says
DeepRESP by Nox Medical is a cloud-based AI medical device that utilizes deep learning to analyze respiratory inductance plethysmography (RIP) belt signals for scoring sleep states (REM, NREM, wake) and arousal events. It aims to improve the accuracy of Home Sleep Apnea Tests (HSATs) by providing sleep staging and arousal scoring without requiring EEG, EOG, or EMG, which are typically used in polysomnography (PSG). The device is FDA-cleared and intended to assist in the diagnosis and follow-up of sleep disorders in adults across various settings. Physicians maintain full control, able to review underlying signal data, confidence levels, and manually adjust scoring.
Overall: PositiveStrengths
- Improves accuracy of AHI in HSATs by enabling sleep staging and arousal scoring without EEG, EOG, or EMG.
- FDA-cleared and validated with a large, diverse dataset of approximately 5,800 sleep recordings.
- Can identify clinically significant OSA in patients, particularly women, who might be underdiagnosed by traditional HSATs.
- Clinicians retain full control to review, edit, validate, and finalize AI-generated scoring.
- Reduces the need for unnecessary follow-up PSG testing and enables earlier, more confident diagnostic and treatment decisions.
- Offers insights into OSA endotypic traits (loop gain, airway collapsibility, muscle compensation, arousal threshold) and burden parameters.
Limitations
- All AI analysis results require review and verification by a medical professional prior to diagnosis.
- Missing signals can affect DeepRESP's output parameters.
- Patients with rare diseases or rare presentations may be underrepresented in the validation data, requiring careful review of outputs by a medical professional.
- No specific reviews from healthcare IT, tech reviewers, G2, or Capterra for DeepRESP were found, though general platforms were identified.
Based on reviews from: Nox Medical, FDA, Reddit
Last updated: 2026-10-24
Ratings & Reviews
No reviews yet. Be the first to review this tool!
