(329 days)
The Osteoboost Belt is indicated to reduce the decline in bone strength and volumetric bone density, as assessed via CT (computed tomography) scans that were analyzed using the O.N. Diagnostics VirtuOst estimate of vertebral bone strength and density in postmenopausal women with osteopenia of the lumbar vertebrae or total hip as diagnosed via dual x-ray absorptiometry with a bone mineral density T-score between -1.0 and -2.49.
- . The clinical effects have only been observed for the duration of the clinical study performed to support the indications for use (1 year).
- Fracture risk was not evaluated in the clinical study to support the indications for use, so . it is not known how the treatment effects correlate with fracture risk.
- . The clinical effects have been demonstrated only for those who used the device as indicated.
The Osteoboost is a wearable belt with an incorporated motor that is designed to transmit lowamplitude, high-frequency (20-40 Hz) vibration to the spine and hips. Osteoboost consists of a vibration pack mounted to a belt made of nylon-covered neoprene. The device is intended to be worn on top of a thin layer of clothing and positioned such that the vibration pack sits tightly over the sacrum. The device includes a pressure sensor under the foam pad to ensure that the device has been fastened with adequate pressure to transmit the vibration correctly. In addition, the belt incorporates an accelerometer positioned at the iliac crest which measures the actual transmission of vibration to the skeleton.
The Osteoboost Belt is a wearable vibration device designed to reduce the decline in bone strength and volumetric bone density in postmenopausal women with osteopenia. The acceptance criteria for this device are established through a combination of clinical and non-clinical performance testing, software verification, and safety assessments.
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria for the Osteoboost Belt are primarily derived from the "SPECIAL CONTROLS" section of the document, indicating the requirements for Class II devices. The reported device performance is extracted from the "SUMMARY OF NONCLINICAL/BENCH STUDIES" and "SUMMARY OF CLINICAL INFORMATION" sections.
| Acceptance Criteria Category | Specific Acceptance Criteria | Reported Device Performance/Study Findings |
|---|---|---|
| Clinical Performance | Reduction in loss of bone strength or bone mineral density. Evaluation of patient risks (pain, dizziness, blurred vision, muscle weakness, difficulty with balance, headache, nausea). Broad range of demographics for at-risk population (minimum 1 year). | Effectiveness: Statistical significance achieved for a 0.48% reduction in bone strength in the active group vs. a 2.84% reduction in the sham group (a difference of 2.36%). A 0.29% volumetric BMD loss in the active group vs. 1.97% loss in the sham group (a difference of 1.68%). This benefit was only observed for subjects who used the device per protocol (at least 3 weekly sessions of 24 minutes each). Duration: 1 year follow-up. Population: 126 postmenopausal women with osteopenia (T-score -1.0 to -2.49). Predominantly Caucasian (95%), with only 5% non-Caucasian subjects. Safety/Risks: No device-related Serious Adverse Events (SAEs). Increased incidence of non-serious adverse effects (back pain, dizziness, muscle weakness) in the active group, reversible upon cessation of use. |
| Non-clinical Performance | Verification and validation of critical performance characteristics (designed outputs, delivered outputs). Accuracy, reliability, and reproducibility of device output. Validation that vibration characteristics are within safe physiologic limits. | Constant Acceleration: Motor frequency remained consistent within approximately ±1% throughout the battery voltage cycle and use-life. Vibration Transmission: Met acceptance criteria; hip acceleration for walking and light chores was larger than when standing still. Accelerometer Accuracy: Measurements from motor and hip accelerometers compared to a calibrated reference accelerometer during varied PWM values met acceptance criteria. Safety of Strong Vibrations: Acceptance criteria for maximum allowable vibration exposure per ISO 2631-1 was met. |
| Biocompatibility | Patient-contacting components demonstrated to be biocompatible. | Not needed, as the device is intended to be worn on top of a thin layer of clothing and does not contact the patient directly. Biocompatibility risk assessment following FDA guidance document "Use of International Standard ISO 10993-1" was conducted. |
| Electrical & Thermal Safety, EMC | Performance data for electrical safety, basic safety, thermal safety, battery safety, and electromagnetic compatibility. | Battery Safety: Compliant with IEC 62133-2:2017. Alarms System: Compliant with IEC 60601-1-8. Electrical Safety/EMC: Compliant with IEC 60601-1:2012, IEC 60601-1-2:2014, and IEC 60601-1-11:2015. Confirmed normal operation in simulated electrosurgical environment, no unintentional radiated emissions, met performance requirements for radiated power/receiver sensitivity/frequency range/channel bandwidth/out-of-band power, immune to electrosurgical equipment. Electrical safety and usability assessed. Test results supported EMC and electrical safety in worst-case running mode. |
| Software | Software verification, validation, and hazard analysis. | Verification Testing: Unit, integration, and system level testing performed. All features pertaining to pressure sensing and modulating vibration speed passed acceptance criteria. Firmware Version: 3.0, C-programming. Hazard Analysis: Performed (implied by the statement "Risk controls were implemented based on the Use Failure Mode Effects Analysis" in the alarms test section). |
| Labeling | Appropriate patient warnings/precautions. Summary of clinical performance testing (outcomes, AEs). Information regarding limitations of clinical significance. Identification of population studied. Instructions on duration/frequency of use, when to discontinue. Instructions for device maintenance/disposal. | Labeling includes: device description, indications for use, instructions for use, warnings, summary of clinical studies, specific population warnings, precautions, charging/cleaning instructions, shelf life/disposal instructions, adverse events, side effects, risks. Meets 21 CFR 801.109 for prescription devices. Includes information on limitations of clinical significance and percentage benefit vs. sham. Specific warnings include: not for osteoporosis; not evaluated in non-Caucasian women, high-risk factors, or BMI > 35; contraindications for certain medical conditions (herniated disc, spinal fusion, joint replacement, active implant); requires thin clothing layer. Details on duration and frequency (at least 3 weekly sessions of 24 minutes each) are provided in the clinical study results and imply inclusion in labeling. |
2. Sample Sizes and Data Provenance
Test Set Sample Size: For the pivotal clinical trial, the sample size was 126 subjects.
Data Provenance: The study was conducted at a single U.S. site. It was a prospective randomized, sham-controlled, triple-blinded clinical trial.
3. Number of Experts and Qualifications for Ground Truth
The document does not explicitly state the number of experts used to establish the ground truth for the test set or their specific qualifications (e.g., radiologist with 10 years of experience).
However, the ground truth for the primary effectiveness endpoint relied on:
- CT scans analyzed using the O.N. Diagnostics VirtuOst estimate of vertebral bone strength and density. This implies that the VirtuOst software/methodology serves as the "expert system" or a validated analytical approach for deriving the ground truth metrics (bone strength and volumetric bone density) from CT data.
- DEXA T-score for initial diagnosis of osteopenia, which would typically be assessed by qualified medical professionals.
4. Adjudication Method for the Test Set
The document does not specify an adjudication method (e.g., 2+1, 3+1) for the test set. Given that the primary endpoint involves quantitative measurements from CT scans analyzed by a specific software (VirtuOst) and DEXA T-scores, it is less likely that a human reader adjudication process (like those used for image interpretation in qualitative studies) would be explicitly mentioned for these endpoints. The "ground truth" here is derived from the analysis of medical imaging data using a specified, presumably validated, method.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
There is no indication that a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was done. The clinical study was a randomized, sham-controlled trial comparing the device's effect on bone strength and density against a sham device, rather than a study comparing human readers with and without AI assistance for diagnosis or interpretation. The device itself delivers a physical therapy, not an AI-powered diagnostic output to a human reader.
6. Standalone (Algorithm Only Without Human-in-the-Loop Performance)
The device is a physical therapy device (wearable vibration belt), not a diagnostic algorithm. Therefore, no standalone (algorithm-only) performance was done in the context of diagnostic accuracy. The "performance" being evaluated is the physiological effect of the device on bone health in human subjects.
However, the software embedded in the device (firmware) controls its operation. Verification testing for this embedded software ensured that it functions as designed, including controlling the motor, collecting data from sensors, and monitoring device functionality. This can be considered a "standalone" evaluation of the internal control system's performance, but not in the sense of a standalone diagnostic algorithm.
7. Type of Ground Truth Used
The type of ground truth used for the clinical study was based on:
- Quantitative Imaging Biomarkers: Measurements of vertebral bone strength and volumetric bone density, assessed via CT scans analyzed using the O.N. Diagnostics VirtuOst estimate.
- Clinical Diagnosis: Diagnosis of osteopenia via Dual X-ray Absorptiometry (DXA) with a bone mineral density T-score between -1.0 and -2.49.
- Adverse Events: Self-reported or clinically observed adverse events/side effects established the safety ground truth.
These are objective quantitative measurements from validated diagnostic methods, rather than expert consensus on subjective interpretations, pathology, or solely outcomes data (like fracture rates, which were explicitly stated as not evaluated as a primary endpoint).
8. Sample Size for the Training Set
The document does not provide information on a "training set" for an AI algorithm. The Osteoboost Belt is a medical device that delivers a physical vibration, not an AI-driven diagnostic or predictive model that would require a large training dataset. The software within the device is firmware for control and data collection, not a learning algorithm.
9. How the Ground Truth for the Training Set Was Established
As there is no mention of an AI-based training set for a learning algorithm, the concept of establishing ground truth for a training set is not applicable to the information provided about the Osteoboost Belt. The software (firmware) verification focused on "that the system functions as designed," implying functional validation against predefined specifications, not learning from data.
N/A