(90 days)
BrainScope TBI is a multi-modal, multi-parameter assessment indicated for use as an adjunct to standard clinical practice to aid in the evaluation of patients who have sustained a closed head injury within the past 72 hours (3 days), are between the ages of 18-85 years, have a Glasgow Coma Scale (GCS) score of 13-15 (including patients with concussion / mild traumatic brain injury (mTBI), and are being considered for a head CT. BrainScope TBI should not be used as a substitute for a CT scan.
The BrainScope TBI Structural Injury Classification ("SIC") uses brain electrical activity to determine the likelihood of structural brain injury visible on head CT. Negative likely corresponds to those with no structural brain injury visible on head CT. Positive likely corresponds to those with a structural brain injury visible on head CT. Equivocal may correspond to structural brain injury visible on head CT or may indicate the need for further observation or evaluation.
BrainScope TBI provides a measure of brain function Index, (BFI)) for the statistical evaluation of the human electroencephalogram (EEG), aiding in the evaluation of head injury as part of a multi-modal, multi-parameter assessment.
The BrainScope TBI device is intended to record, measure, analyze, and display brain electrical activity utilizing the calculation of standard quantitative EEG (QEEG) parameters from frontal locations on a patient's forehead. The BrainScope TBI calculates and displays raw measures for the following standard QEEG measures: Absolute and Relative Power, Asymmetry, Coherence and Fractal Dimension. These raw measures are intended to be used for post hoc analysis of EEG signals for interpretation by a qualified user.
BrainScope TBI also provides clinicians with quantitative measures of cognitive performance to aid in the assessment of an individual's level of cognitive function. These measures do not interact with any other device measures, and are stand alone.
BrainScope TBI also stores and displays electronic of standardized clinical assessment tools that should be used in accordance with the assessment tools' general instructions. These tools do not interact with any other device measures, and are stand alone.
BrainScope TBI is a portable, non-invasive, non-radiation emitting, point of care device intended to provide results and measures to support clinical assessments and aid in the diagnosis of mild traumatic brain injury (mTBI). It also contains configurable, selectable computerized cognitive performance tests and digitized standard clinical assessment intended to provide a multi-modal panel of measures to support the clinical assessment of concussion / mTBI. BrainScope TBI provides healthcare professionals with a set of validated and clinically accepted library of concussion / mTBI assessments.
1. Acceptance Criteria and Reported Device Performance:
The document does not explicitly state formal acceptance criteria with numerical thresholds. Instead, it focuses on demonstrating substantial equivalence to predicate devices. The performance data presented primarily confirms that the new modifications (cognitive performance tests, PECARN, wireless connectivity) work as specified and do not negatively impact existing functionalities.
Table of Acceptance Criteria (Inferred from Substantial Equivalence Claim) and Reported Device Performance:
| Acceptance Criteria (Inferred/Implicit) | Reported Device Performance |
|---|---|
| For Cognitive Performance Tests: | |
| Functionality of new cognitive tests | Normative data was collected from 707 healthy individuals (age 13-85) to construct databases for the cognitive tests. Test data demonstrated that the modifications (additional cognitive performance tests and PECARN) were implemented as per specifications. The new tests and Reliable Change Index (RCI) output are "well accepted in clinical practice for assessment of Adult and Adolescent patient population." |
| For Standard Clinical Assessments: | |
| Integration of PECARN | Test data demonstrated that the modifications (additional cognitive performance tests and PECARN) were implemented as per specifications. PECARN was added to the existing library of digitized standard clinical assessments. The expanded availability of clinical assessment tools does not affect safety and effectiveness and increases utility. |
| For Wireless Connectivity (OTA): | |
| Functionality of wireless connectivity | Test data demonstrated that the modifications (wireless connectivity) were implemented as per specifications. The BrainScope TBI has wireless connectivity to accept Over the Air (OTA) software upgrades. This provides additional data transfer capabilities. |
| For Existing EEG Algorithms (SIC, BFI, QEEG): | |
| No impact on existing functionality | The new modifications "did not impact existing device functionality including core EEG based algorithms" (e.g., Structural Injury Classification (SIC) and Brain Function Index (BFI)). The device maintains the same technical characteristics as the predicate for EEG parameters (e.g., bandwidth, CMRR, noise floor, ADC resolution, sampling rate, electrode placement, electrode positions, electrode material, real-time EEG display, EEG-based classification algorithm). |
| Basic Safety and EMC Standards: | |
| Conformity to relevant standards | The BrainScope TBI device conforms to "all same basic safety and EMC standards as the predicate." It was also tested to the most recent recognized consensus standard for EMC (IEC 60601-1-2 Ed. 4.0 2014) and other listed standards (e.g., IEC 60601-1/A1:2012, IEC 60601-1-6/A1:2013, IEC 60601-2-26:2012, ANSI/AAMI EC12:2000/(R)2010, ANSI/AAMI/ISO 10993-1:2009, ANSI/AAMI/ISO 10993-5:2009, ANSI/AAMI/ISO 10993-10:2010, MIL-STD-810G, IEC 60529 (2004), ASTM D4169 09). |
| Overall Safety and Effectiveness: | |
| Comparable to predicate devices | Performance data demonstrated that the BrainScope TBI is "as safe and effective as the predicates." |
2. Sample size used for the test set and the data provenance:
- Sample Size: 707 healthy individuals.
- Data Provenance: The document states that normative data was collected from "707 healthy individuals" for the cognitive tests. It does not specify the country of origin, but given the context of FDA submission, it is typically expected to be from a US-based population or a population generalizable to the US. It is a prospective collection for establishing normative data for the new cognitive tests.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
The document does not explicitly state the number or qualifications of experts used to establish ground truth for the test set of the new cognitive performance tests. It refers to the new tests and RCI output being "well accepted in clinical practice," which implies expert consensus in the field, but no specific quantification of experts is provided within this document. For the EEG-based Structural Injury Classification (SIC), the ground truth is based on the visible injury on head CT, which is a clinical standard.
4. Adjudication method for the test set:
Not explicitly stated for the cognitive performance tests or the overall assessment of device modifications. Given the nature of normative data collection for cognitive tests, the "ground truth" is often statistical (e.g., population averages, standard deviations) rather than expert adjudication on individual cases for classification. For the SIC, the ground truth is based on head CT findings, which are objective imaging results.
5. 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:
The document does not mention an MRMC comparative effectiveness study where human readers' performance with and without AI assistance was evaluated. The device is described as an "adjunct to standard clinical practice" and "should not be used as a substitute for a CT scan," suggesting it provides additional information rather than directly assisting in CT interpretation by a reader.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
Yes, the Structural Injury Classification (SIC) and Brain Function Index (BFI) are described as "EEG based algorithms" which "use brain electrical activity to determine the likelihood of structural brain injury visible on head CT." This implies a standalone algorithmic assessment of the EEG data, providing "Negative, Equivocal and Positive outputs" for SIC and a "statistical evaluation of the human EEG" for BFI.
7. The type of ground truth used:
- For Structural Injury Classification (SIC): Visible structural brain injury on head CT. This is a form of outcomes data or a clinical standard.
- For Cognitive Performance Tests: Normative data derived from a population of healthy individuals. This establishes a baseline for cognitive function.
- For QEEG parameters: Standards related to the accurate recording, measuring, analyzing, and displaying of brain electrical activity, as well as comparison to the predicate device's performance characteristics.
8. The sample size for the training set:
The document does not specify the sample size for the training set. It only mentions the "normative data collected from 707 healthy individuals" used to construct databases for the cognitive tests, which likely serves as a reference/validation set for these specific components. The EEG-based algorithms (SIC, BFI) were likely developed and trained on separate, larger datasets that are not detailed in this particular summary, as these algorithms are shared with the predicate device (BrainScope One).
9. How the ground truth for the training set was established:
The document does not provide details on how the ground truth for the training set was established for the core EEG algorithms (SIC, BFI) as these algorithms were already established and cleared under the predicate device (BrainScope One). For the new cognitive performance tests and PECARN, the ground truth for the "normative data" was established by collecting data from "healthy individuals," implying a healthy control group without known head injury or cognitive impairment.
§ 882.1450 Brain injury adjunctive interpretive electroencephalograph assessment aid.
(a)
Identification. A brain injury adjunctive interpretive electroencephalograph assessment aid is a prescription device that uses a patient's electroencephalograph (EEG) to provide an interpretation of the structural condition of the patient's brain in the setting of trauma. A brain injury adjunctive interpretive EEG assessment aid is for use as an adjunct to standard clinical practice only as an assessment aid for a medical condition for which there exists other valid methods of diagnosis.(b)
Classification. Class II (special controls). The special controls for this device are:(1) The technical parameters of the device, hardware and software, must be fully characterized and include the following information:
(i) Hardware specifications must be provided. Appropriate verification, validation, and hazard analysis must be performed.
(ii) Software, including any proprietary algorithm(s) used by the device to arrive at its interpretation of the patient's condition, must be described in detail in the software requirements specification (SRS) and software design specification (SDS). Appropriate software verification, validation, and hazard analysis must be performed.
(2) The device parts that contact the patient must be demonstrated to be biocompatible.
(3) The device must be designed and tested for electrical safety, electromagnetic compatibility (EMC), thermal, and mechanical safety.
(4) Clinical performance testing must demonstrate the accuracy, precision-repeatability and reproducibility, of determining the EEG-based interpretation, including any specified equivocal zones (cutoffs).
(5) Clinical performance testing must demonstrate the ability of the device to function as an assessment aid for the medical condition for which the device is indicated. Performance measures must demonstrate device performance characteristics per the intended use in the intended use environment. Performance measurements must include sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) with respect to the study prevalence per the device intended use.
(6) The device design must include safeguards to ensure appropriate clinical interpretation of the device output (
e.g., use in appropriate patient population, or for appropriate clinical decision).(7) The labeling and training information must include:
(i) A warning that the device is not to be used as a stand-alone diagnostic.
(ii) A detailed summary of the clinical performance testing, including any adverse events and complications.
(iii) The intended use population and the intended use environment.
(iv) Any instructions technicians should convey to patients regarding the collection of EEG data.
(v) Information allowing clinicians to gauge clinical risk associated with integrating the EEG interpretive assessment aid into their diagnostic pathway.
(vi) Information allowing clinicians to understand how to integrate the device output into their diagnostic pathway when the device is unable to provide a classification or final result.