Search Results

The ClearEdge® Balance System is intended for use to assess sway as an indicator of balance. Individual suitability for assessment must be judged on a case by case basis, by a qualified including those certified and/or licensed in their state to prescribe and/or use balance devices such as certified athletic trainers, physical therapists, chiropractors, nurses and physicians. The ClearEdge® Balance System can be used wherever compatible Android mobile operating devices can be used.

The ClearEdge® Balance System is a mobile software system that analyzes balance through measurements of postural sway. The System is comprised of application software, and accompanying hardware consisting of a mobile computing device, proprietary sensor, balance pad, and friction pad. The test data is collected, uploaded to a HIPAA compliant server, analyzed by Motion Intelligence developed algorithms and presented in report form for clinician's review. The reports indicate how a subject's balance score may have changed between testing sessions, and whether the measured change is likely due to measurement error and normal human variation in performance of the selected balance stance or a real change in performance caused by an external factor.

Using proprietary software on an Android tablet and the Edge Sensor, the ClearEdge Balance System tests balance by administering eight select balance stances.

Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided FDA 510(k) summary for the ClearEdge Balance System:

This document focuses on demonstrating substantial equivalence to a predicate device (Sway™ Balance System) rather than absolute performance against pre-defined, numerical acceptance criteria for a novel device. Therefore, the "acceptance criteria" presented are implicitly derived from the comparison to the predicate and statistical demonstrations of equivalence.

1. Table of Acceptance Criteria and Reported Device Performance

The core "acceptance criteria" for demonstrating substantial equivalence are based on demonstrating that the ClearEdge Balance System's performance is equivalent enough to the predicate device (Sway™ Balance System) to not raise new questions of safety or effectiveness. This is shown through a combination of laboratory testing and human performance testing.

The Study Proving the Device Meets Acceptance Criteria

The study described is a comparative study designed to establish substantial equivalence to a predicate device (Sway™ Balance System), rather than a de novo validation against absolute performance metrics.

1. Sample Size Used for the Test Set and Data Provenance:

   • Laboratory Testing: Involved continuous movement data from an "inverted pendulum." The specific sample size in terms of number of data points or test runs is not explicitly stated, but the "Figure 0" implies a variety of frequency and amplitude combinations were tested on the pendulum. The "dots represent the actual frequency & Amp for Sway & CE Score." This is a controlled laboratory environment; data provenance is lab-generated.
   • Human Performance Testing: 34 subjects.
     • Data Provenance: Not explicitly stated (e.g., country of origin, single vs. multi-site), but implies a prospective human study.
     • Retrospective/Prospective: The description "Supplemental clinical testing compared ClearEdge Balance and Sway Balance scores for 34 subjects..." suggests a prospective data collection for this direct comparison.

2. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts:

   • Laboratory Testing: The ground truth for the inverted pendulum test is the mechanical input (frequency and amplitude). No human expert is involved in establishing this ground truth.
   • Human Performance Testing: The "ground truth" in this comparative study is the Sway™ Balance System's score. The study aims to show that ClearEdge's scores are equivalent to Sway's. There are no independent human experts establishing a "true" balance score against which both devices are compared. The assumption is that Sway's measurement is a valid indicator of balance.
   • MDC Development: This involved a test-retest repeatability study with 144 subjects (109 healthy, 35 with neurologic condition). The "ground truth" is derived from statistical analysis of repeated measurements from the ClearEdge system itself, not from expert adjudication of balance.

3. Adjudication Method for the Test Set:

   • Not applicable in the traditional sense for these studies.
   • For the laboratory testing, the "truth" is the known physical properties and movements of the inverted pendulum.
   • For the human performance testing, the comparison is directly between the two devices (ClearEdge and Predicate Sway). There's no third-party adjudication of the "true" balance of the subjects. The statistical method used was a Two One-Sided Test Procedure (TOST) to demonstrate equivalence.

4. If A Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, If So, What Was the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance:

   • No, an MRMC study was NOT done. This device is a balance assessment system, not an AI-assisted diagnostic imaging device that involves human "readers" or interpretation of complex images. The focus is on the device's ability to measure balance similarly to a predicate device.

5. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done:

   • Yes, implicitly. Both the laboratory testing with the inverted pendulum and the human performance testing effectively evaluate the algorithm's performance in producing a score based on sensor input. The device's output (a numeric balance score) is the direct result of its algorithms. There isn't a human "in the loop" modifying the score; the human users apply the device and interpret its output, but the calculation of the score itself is standalone.

6. The Type of Ground Truth Used:

   • Laboratory Testing: Mechanical parameters of an inverted pendulum as a model for human sway.
   • Human Performance Testing: The scores generated by the legally marketed predicate device (Sway™ Balance System) are used as the reference "truth" for comparison, demonstrating equivalence.
   • MDC Development: Statistical derivation from repeated measurements of the device under test (ClearEdge) itself on human subjects to determine variability.

7. The Sample Size for the Training Set:

   • Not explicitly stated for a distinct "training set." This submission details validation studies for a device rather than a deep learning AI model that would typically have a separate, large training set. The device appears to rely on established biomechanical models and signal processing rather than machine learning that requires supervised training.
   • The term "developed algorithms" for Motion Intelligence suggests proprietary methods, but details on how these were "trained" (if they involve machine learning) are absent. The MDC development study (144 subjects) could be considered a form of internal validation used to characterize the device's measurement properties.

8. How the Ground Truth for the Training Set Was Established:

   • As a distinct "training set" is not explicitly mentioned for algorithm development in line with deep learning, the method for establishing its "ground truth" is also not detailed.
   • For the MDC development, the "truth" for calculating measurement error and normal variation is derived from repeat measurements of the same individuals, assuming that changes between short-interval tests beyond a certain threshold represent actual change rather than random variability.

Ask a specific question about this device

The Sway™ Balance System is intended for use to assess sway as an indicator of balance. Individual suitability for assessment must be judged on a case by case basis, by a qualified individual including those certified and/or licensed in their state to prescribe and/or use balance devices such as certified athletic trainers and coaches, physical therapists, nurses and physicians. Conditions affecting postural sway include nausea, headache, orthopedic injury, ear infection, medications, head injury, dehydration and fatigue. The Sway™ Balance System can be used wherever an iOS mobile operating device is available.

The Sway Balance™ System is a mobile measurement system that analyzes balance through thoracic sway, using the built in accelerometer of a mobile device. The Sway Balance™ System is a stand-alone mobile operating system software application that does not include any peripheral hardware add-ons.

The provided text is a 510(k) summary for the Sway Balance™ System, a mobile application designed to assess balance. Based on this document, here's an analysis of the acceptance criteria and the study proving the device meets them:

1. Acceptance Criteria and Reported Device Performance

2. Sample Size and Data Provenance (Test Set)

• Sample Size: The document does not explicitly state the sample size for the clinical testing (which serves as the "test set" for performance evaluation). It mentions "studies" (plural) but no numbers of participants.
• Data Provenance: Not specified. It's unclear if the data was retrospective or prospective, or the country of origin.

3. Number of Experts and Qualifications (Ground Truth for Test Set)

• This information is not provided in the document. The ground truth (force platform data) is established by a device, not human experts in this context.

4. Adjudication Method (Test Set)

• This is not applicable for this type of device comparison study. The ground truth is established by a quantitative measurement (force platform), not through expert consensus requiring adjudication.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• No, an MRMC study was not done. This device is a quantitative measurement tool, not an AI-assisted diagnostic tool that would typically involve human readers interpreting images. The study focuses on comparing the device's output to a gold standard measurement device, not on how it assists human interpretation.

6. Standalone (Algorithm Only) Performance

• Yes, standalone performance was assessed. The entire testing described (bench testing and clinical testing comparing to a force platform) is an assessment of the algorithm's standalone performance, i.e., how accurately the Sway Balance™ System itself measures sway. There is no human element in the loop of this specific performance evaluation.

7. Type of Ground Truth Used

• The ground truth used was comparative device data, specifically data obtained from "force platform technology." Force platforms are established gold standards for measuring balance and sway.

8. Sample Size for the Training Set

• The document does not mention a training set or its size. As a software application utilizing a built-in accelerometer to measure physical motion, it's possible that a traditional "training set" for machine learning algorithms (as understood in the context of image recognition or complex pattern detection) was not explicitly used or documented in this summary. The development likely involved calibration and verification against physical principles and established balance measurement techniques rather than training on a large dataset of patient balance assessments.

9. How Ground Truth for the Training Set Was Established

• Since a distinct "training set" is not mentioned, the method for establishing its ground truth is not provided. If implied, the "training" (or development and calibration) would be based on the known physics of accelerometry and validation against physical movements and potentially against force platform data during the development phase.

Ask a specific question about this device