Search Results

The Sway System Sports Plus is intended for use as a computerized cognitive assessment and management of concussion in individuals ages 18-24. The Sway System Sports Plus is indicated for use when there is a suspected concussion or head injury.

Sway Sports Plus (referred to as Sway Sports+) is a licensed variant of the Device Sway System and is a software solution (SaMD) that provides a computerized cognitive assessment aid for concussion. This prescription device uses an individual's score(s) on a battery of cognitive tasks to provide an indication of the current level of cognitive function in response to a suspected concussion. The computerized cognitive assessment aid for concussion is used only as an assessment aid in the management of concussion to determine cognitive function for patients after a potential concussive event where other diagnostic tools are available and does not identify the presence or absence of concussion. The examiner may have the patient complete the Sway Balance test to provide an additional indicator in their assessment of concussion. The Sway System is not intended as a stand-alone diagnostic device. After the patient completes the Sway Sports+ test(s), the examiner can view the summary statistics for each test such as mean reaction time (speed). Sway Sports+ does not create new information through analysis or interpretation, nor does it offer any diagnostic or treatment recommendations. The Sway Sports+ is not intended to be used as a standalone diagnostic device. Sway Sports+ operates in the Apple iOS or Android operating systems, enabling it to be used on the hardware of Apple iPhone. iPad, and iPod Touch. or Android mobile devices. The internal accelerometer is accessed to analyze motion of the device in response to stimuli during testing. The mobile device hardware is not provided by Sway Medical. Environments where Sway Sports+ may be used may include both clinical and non-clinical settings. Clinical settings may include physician offices, hospitals, physical or occupational therapy clinics, skilled nursing facilities, assisted living centers and other healthcare facilities where cognitive tests may be evaluated. In addition, non-clinical environments may include sports team's locations where the application may be used on the sideline to perform a cognitive assessment, e.g., fitness centers, gymnasiums, or stadiums. It is recommended that the tests be taken in a supervised environment for baseline and post-injury testing. Sway Sports+ consists of the Sway Balance (K121590) and the Sway Computerized Cognitive Assessment Aid. The Sway Computerized Cognitive Assessment Aid (PTY) is exempt from 510(k).

Here's a breakdown of the acceptance criteria and the study information for the Sway System Sports Plus, based on the provided text:

Acceptance Criteria and Device Performance

The provided text details the non-clinical and clinical performance testing results, and how they relate to the substantial equivalence claim rather than explicit "acceptance criteria" in a pass/fail table format. However, we can infer the acceptance criteria from the reported performance results and the comparison to predicate devices.

Table 1: Inferred Acceptance Criteria and Reported Device Performance

Study Details

1. Sample size used for the test set and the data provenance:

   • Validity Studies:
     • VanRavenhorst-Bell et al., 2021: 88 healthy adults aged 18 to 48 years (mean 22.09 ± 4.47 years). Data provenance not explicitly stated, but implies U.S. based on the context of FDA submission. Retrospective as it compares existing tools.
     • Clark & VanRavenhorst-Bell, 2022: 85 adults (mean age 23.1 years; 68% female). Data provenance not explicitly stated, but implies U.S. based on the context of FDA submission. Prospective (all tests administered in one session by trained professionals).
   • Reliability Studies:
     • Van Patten et al., 2021: 55 adults (69% women) with a mean age of 26.69 years (± 9.89; range = 18-58). Data provenance not explicitly stated, but implies U.S. based on the context of FDA submission. Prospective (assessed remotely over 3 subsequent weeks).
     • Caccese et al., 2022: 55 healthy adults (69% women) with a mean age of 26.69 years (SD = 9.89; range = 18-58). Data provenance not explicitly stated, but implies U.S. based on the context of FDA submission. Prospective (assessed remotely over 3 subsequent weeks).
     • Additional study (healthy adults): 97 participants initially recruited, with 56 completing all test and retest sessions (one-week interval). Data provenance not explicitly stated, but implies U.S. based on the context of FDA submission. Prospective.
     • Additional study (concussed individuals): Utilized a dataset derived from NCAA member schools. Data provenance implies U.S. Retrospective (analyzing existing NCAA data for a within-session design).
   • Sensitivity & Specificity Studies:
     • Chikar & Curtiss, 2023: 46 athletes (mean age 19.6, SD=1.4; 59% male) with confirmed concussions. Data from athletic organizations that conducted both ImPACT and Sway assessments. Data provenance implies U.S. and likely retrospective.
     • Additional study (sensitivity/specificity): No explicit sample size given, describes "high sensitivity but moderate specificity." Data provenance not explicitly stated, but implies U.S. based on the context of FDA submission. Likely retrospective analysis of existing data.
     • Additional pilot study (ImPACT vs. Sway): 24 high school and college-aged participants. Data provenance implies U.S. Prospective (underwent baseline and post-injury testing with both tools).

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • The document primarily relies on comparisons to established and FDA-cleared devices (ImPACT QT, ImPACT, traditional neurocognitive assessments) and clinical standards for concussion diagnosis (NCAA Concussion Safety Protocol guidelines).
   • For the Clark & VanRavenhorst-Bell, 2022 validity study, "all tests administered by trained professionals." Specific number or qualifications of these professionals (e.g., neuropsychologists, physicians) are not detailed.
   • For the concussed individuals reliability study and the sensitivity/specificity pilot study, the "clinical standards for concussion diagnosis were strictly followed according to NCAA Concussion Safety Protocol guidelines" or involved "athletic organizations that conducted both ImPACT and Sway assessments." This implies diagnosis by medical professionals (e.g., team physicians, athletic trainers) adhering to established protocols, but no specific number or detailed qualifications of these "experts" is provided.

3. Adjudication method for the test set:

   • No explicit adjudication method (e.g., 2+1, 3+1) is described for establishing ground truth within the provided studies. The ground truth often relies on the established scores/diagnoses from predicate devices or clinical protocols.

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 MRMC comparative effectiveness study involving human readers' improvement with AI vs. without AI assistance is described. The device is a "computerized cognitive assessment aid," meaning it provides data, but does not inherently involve human "readers" interpreting images (as would be common in diagnostic imaging MRMC studies). The studies focus on the device's measurement properties and its correlation with other assessment tools.

5. If a standalone (i.e. algorithm only without human-in-the loop performance) was done:

   • Yes, the studies evaluate the Sway Sports+ system (algorithm/software) primarily in a standalone capacity, assessing its internal consistency (reliability), its correlation with established cognitive tests (validity), and its ability to distinguish between concussed and non-concussed states (sensitivity/specificity) based on its output.
   • The document states, "While Sway Sports+ is not intended to be used to classify a patient as having a concussion, it may be useful to create criteria based on the device's output to classify a patient as concussion positive or negative, to allow comparison to predicate devices..." and "Sway Sports+ does not create new information through analysis or interpretation, nor does it offer any diagnostic or treatment recommendations. The Sway Sports+ is not intended to be used as a standalone diagnostic device." However, the performance studies themselves are evaluating the device's output directly. The implication is that a human still needs to interpret these outputs in the broader clinical context. The sensitivity/specificity studies show how well the device's output alone, when used to classify, performs.

6. The type of ground truth used:

   • Comparison to predicate devices: ImPACT QT, ImPACT, traditional neurocognitive assessments (Wechsler Adult Memory Scale-IV (WMS-IV), Wechsler Adult Intelligence Scale-IV (WAIS-IV), Delis-Kaplan Executive Function System (D-KEFS), and California Verbal Learning Test – 3rd Edition (CVLT-III)).
   • Clinical Diagnosis: Concussion diagnosis confirmed by medical professionals following NCAA Concussion Safety Protocol guidelines.
   • Healthy Controls: Participants recruited as "healthy adults" or "healthy controls."

7. The sample size for the training set:

   • Normative Database: Over 126,000 de-identified baseline test data records from users aged 18 to 24 years old were used to establish the normative database. This effectively serves as a large training/reference set for individual score comparison.
   • No specific "training set" size for the AI algorithm itself (if applicable, as it's a "computerized cognitive assessment aid" not explicitly framed as deep learning AI) is provided beyond the normative data.

8. How the ground truth for the training set was established:

   • For the normative database (126,000+ users), the ground truth was established by:
     • Self-reported health status: Users "did not report a concussion within the last 6 months, did not report neurological conditions, did not report learning disability."
     • Behavioral criteria: "Completed a baseline test in English, completed a baseline test in the United States of America, primary language was English, had valid scores on all tests."
     • This establishes a "healthy, non-concussed baseline" for reference rather than a diagnostic ground truth.

Ask a specific question about this device