(89 days)
Methinks CTA Stroke is a radiological computer aided triage and notification software, parallel workflow tool for use by hospital networks and trained clinicians to identify and communicate images of specific patients to a specialist, independent of standard of care workflow.
Methinks CTA Stroke uses an artificial intelligence algorithm to analyze images for findings suggestive of a pre-specified clinical condition and to notify an appropriate medical specialist of these findings in parallel to standard of care image interpretation.
Identification of suspected findings is not for diagnostic use beyond notification.
Specifically, the device analyzes CT angiogram images of the brain acquired in the acute setting and sends to PACS and/or notifications to a neurovascular specialist that a suspected large vessel occlusion has been identified and recommends review of those images. Images can be previewed through an image viewer. Methinks CTA Stroke is intended to analyze terminal ICA, MCA-M1 and MCA-M2 vessels for LVOs.
Images that are previewed are for informational purposes only and not intended for diagnostic use beyond notification. Notified clinicians are responsible for viewing non-compressed images on a diagnostic viewer and engaging in appropriate patient evaluation and relevant discussion with a treating physician before making care-related decisions or requests. Methinks CTA Stroke is limited to analysis of imaging data and should not be used in-lieu of full patient evaluation or relied upon to make or confirm diagnosis.
Methinks CTA Stroke is a software-only device which is intended to be used by trained physicians involved in the management of Acute Stroke (AS) patients at emergency settings or other departments across the stroke care pyramid model. They include trained physicians such as emergency physicians, neurologists, general radiologists, neurovascular interventionists, neuroradiologists and any trained stroke professionals.
The target patients (intended patient population) are male and female in the adult population (above 21 years old) with suspected Acute Stroke.
The Methinks CTA Stroke device analyzes Computed Tomography Angiography (CTA) images from the intended patient population to identify suspected Large Vessel Occlusions (LVO). This information is to be used in conjunction with other patient information by a professional to assist with triage/prioritization of medical images.
The input of the software is Computed Tomography Angiography (CTA) in DICOM format from patients suspected of Acute Stroke. The outputs of the software are notifications sent to the trained physicians intended to be used in conjunction with other patient information for professional judgment to assist with triage/prioritization.
Here is a comprehensive breakdown of the acceptance criteria and the study proving the Methinks CTA Stroke device meets those criteria, based on the provided FDA 510(k) clearance letter:
Acceptance Criteria and Study Details for Methinks CTA Stroke
Context: The Methinks CTA Stroke device is a radiological computer-aided triage and notification software that uses an AI algorithm to analyze CT angiogram images for findings suggestive of a Large Vessel Occlusion (LVO) and notifies a neurovascular specialist.
1. Table of Acceptance Criteria and Reported Device Performance
The direct acceptance criteria (pre-specified performance goals) are explicitly stated in the document for Sensitivity and Specificity. The time to notification is also presented as a performance metric.
| Performance Metric | Acceptance Criteria (Pre-specified Goal) | Reported Device Performance (95% CI) |
|---|---|---|
| Sensitivity for LVO | Exceeds (unspecified threshold) | 98.2% (93.6% - 99.8%) |
| Specificity for LVO | Exceeds (unspecified threshold) | 91.6% (87.2% - 94.9%) |
| Time to Notification | Not explicitly stated as an acceptance criteria threshold, but documented. | Mean: 3.30 minutes (3.23 - 3.36 minutes) |
Note: While the document states "Sensitivity and specificity exceed the pre-specified performance goals for LVO," the specific numerical thresholds for these goals are not provided in the extract.
2. Sample Size Used for the Test Set and Data Provenance
- Test Set Sample Size: 336 cases
- LVO Positive: 110 cases
- LVO Negative: 226 cases
- Data Provenance: Retrospective, blinded, multicenter, multinational study. Institutions included in the validation study were different from institutions included in training, ensuring separation and representativity. This was verified by checking countries, states, and ZIP codes. The specific countries are not mentioned beyond "multinational."
3. Number of Experts Used to Establish Ground Truth and Qualifications
- Number of Experts: Two primary readers, with a third expert for adjudication. (Total of 3 experts involved in establishing ground truth for any given case of disagreement)
- Qualifications of Experts: US board-certified neuroradiologists. (No years of experience are specified).
4. Adjudication Method for the Test Set
- Method: Majority vote (2+1 adjudication). Ground truth was established by two US board-certified neuroradiologists. If they disagreed regarding LVO findings, a third ground truther established the final ground truth based on the majority vote.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- The provided document does not indicate that an MRMC comparative effectiveness study was done looking at how human readers improve with AI vs. without AI assistance. The study focuses solely on the standalone performance of the AI algorithm.
6. Standalone Performance (Algorithm Only)
- Yes, a standalone performance study was done. The reported Sensitivity and Specificity values (98.2% and 91.6% respectively) represent the performance of the AI algorithm itself in identifying LVOs, without human-in-the-loop assistance for the core performance metrics.
7. Type of Ground Truth Used
- Ground Truth Type: Expert consensus. Specifically, it was established by two US board-certified neuroradiologists, with a third neuroradiologist for adjudication in case of disagreement.
8. Sample Size for the Training Set
- The document does not explicitly state the sample size used for the training set. It only mentions that "Institutions included in the validation study were different from institutions included in training," but the training dataset size is not provided.
9. How Ground Truth for the Training Set Was Established
- The document does not explicitly describe how ground truth for the training set was established. It only details the ground truth establishment process for the test set. It is implied that similar expert review would have been used, but no specific methodology or number of readers are provided for the training data.
§ 892.2080 Radiological computer aided triage and notification software.
(a)
Identification. Radiological computer aided triage and notification software is an image processing prescription device intended to aid in prioritization and triage of radiological medical images. The device notifies a designated list of clinicians of the availability of time sensitive radiological medical images for review based on computer aided image analysis of those images performed by the device. The device does not mark, highlight, or direct users' attention to a specific location in the original image. The device does not remove cases from a reading queue. The device operates in parallel with the standard of care, which remains the default option for all cases.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Design verification and validation must include:
(i) A detailed description of the notification and triage algorithms and all underlying image analysis algorithms including, but not limited to, a detailed description of the algorithm inputs and outputs, each major component or block, how the algorithm affects or relates to clinical practice or patient care, and any algorithm limitations.
(ii) A detailed description of pre-specified performance testing protocols and dataset(s) used to assess whether the device will provide effective triage (
e.g., improved time to review of prioritized images for pre-specified clinicians).(iii) Results from performance testing that demonstrate that the device will provide effective triage. The performance assessment must be based on an appropriate measure to estimate the clinical effectiveness. The test dataset must contain sufficient numbers of cases from important cohorts (
e.g., subsets defined by clinically relevant confounders, effect modifiers, associated diseases, and subsets defined by image acquisition characteristics) such that the performance estimates and confidence intervals for these individual subsets can be characterized with the device for the intended use population and imaging equipment.(iv) Stand-alone performance testing protocols and results of the device.
(v) Appropriate software documentation (
e.g., device hazard analysis; software requirements specification document; software design specification document; traceability analysis; description of verification and validation activities including system level test protocol, pass/fail criteria, and results).(2) Labeling must include the following:
(i) A detailed description of the patient population for which the device is indicated for use;
(ii) A detailed description of the intended user and user training that addresses appropriate use protocols for the device;
(iii) Discussion of warnings, precautions, and limitations must include situations in which the device may fail or may not operate at its expected performance level (
e.g., poor image quality for certain subpopulations), as applicable;(iv) A detailed description of compatible imaging hardware, imaging protocols, and requirements for input images;
(v) Device operating instructions; and
(vi) A detailed summary of the performance testing, including: test methods, dataset characteristics, triage effectiveness (
e.g., improved time to review of prioritized images for pre-specified clinicians), diagnostic accuracy of algorithms informing triage decision, and results with associated statistical uncertainty (e.g., confidence intervals), including a summary of subanalyses on case distributions stratified by relevant confounders, such as lesion and organ characteristics, disease stages, and imaging equipment.