Search Results

StrokeSENS ASPECTS is a computer-aided diagnosis (CADx) software device used to assist the clinician in the assessment and characterization of brain tissue abnormalities using CT image data.

The Software automatically registers images and uses an Atlas to segment and analyze ASPECTS Regions. StrokeSENS ASPECTS extracts image data from individual voxels in the image to provide analysis and computer analytics and relates the analysis to the atlas defined ASPECTS regions. The imaging features are then synthesized by an artificial intelligence algorithm into a single ASPECT (Alberta Stroke Program Early CT) Score.

StrokeSENS ASPECTS is indicated for evaluation of patients presenting for diagnostic imaging workup with known MCA or ICA occlusion, for evaluation of extent of disease. Extent of disease refers to the number of ASPECTS regions affected which is reflected in the total score. StrokeSENS ASPECTS provides information that may be useful in the characterization of ischemic brain tissue injury during image interpretation (within 12 hours from time last known well).

StrokeSENS ASPECTS provides a comparative analysis to the ASPECTS standard of care radiologist assessment by providing highlighted ASPECTS regions and an automated editable ASPECTS score for clinician review. StrokeSENS ASPECTS presents the original and annotated images for concurrent reads. StrokeSENS ASPECTS additionally provides a visualization of the voxels contributing to the automated ASPECTS score.

Limitations:

1. StrokeSENS ASPECTS is not intended for primary interpretation of CT images. It is used to assist physician evaluation.
2. StrokeSENS ASPECTS has been validated in patients with known MCA or ICA occlusion prior to ASPECTS scoring.
3. Use of StrokeSENS ASPECTS in clinical settings other than brain ischemia within 12 hours from time last known well, caused by known ICA or MCA occlusions, has not been tested.
4. StrokeSENS ASPECTS has only been validated and is intended to be used in patient populations aged over 21.

Contraindications:

• StrokeSENS ASPECTS is contraindicated for use on brain scans displaying neurological pathologies other than acute ischemic stroke, such as tumors or abscesses, hemorrhagic transformation, and hematoma.

Cautions:

• Patient Motion: Excessive patient motion leading to artifacts that make the scan technically inadequate.

StrokeSENS ASPECTS is a stand-alone software device that uses machine learning algorithms to automatically process NCCT (non-contrast computed tomography) brain image data to provide an output ASPECTS score based on the Alberta Stroke Program Early CT Score (ASPECTS) guidelines.

The post-processing image results and ASPECTS score are identified based on regional imaging features and overlayed onto brain scan images. StrokeSENS ASPECTS provides an automated ASPECTS score based on the input CT data for the physician. The score includes which ASPECTS regions are identified based on regional imaging features derived from non-contrast computed tomography (NCCT) brain image data. The results are generated based on the Alberta Stroke Program Early CT Score (ASPECTS) guidelines and provided to the clinician for review and verification. At the discretion of the clinician, the scores may be adjusted based on the clinician's judgement.

StrokeSENS ASPECTS can connect with other DICOM-compliant devices, to transfer NCCT scans for software processing.

Results and images can be sent to a PACS via DICOM transfer and can be viewed on a PACS workstation or via the StrokeSENS UI or other DICOM-compatible radiological viewer.

StrokeSENS ASPECTS provides an automated workflow which will automatically process image data received by the system in accordance with pre-configured user DICOM routing preferences.

StrokeSENS ASPECTS principal workflow for NCCT includes the following key steps:

• Receive NCCT DICOM Image
• Automated image analysis and processing to identify and visualize the voxels which have been included in the ASPECTS score (Also referred to as a 'heat map' or 'VCTA; Voxels Contributing to ASPECTS Score').
• Automated image analysis and processing to register the subject image to an atlas to segment and highlight ASPECTS regions and to display whether or not each region is qualified as contributing to the ASPECTS score.
• Generation of auto-generated results for review and analysis by users.
• Generation of verified/modified result summary for archiving, once the user verifies or modifies the results.

Once the auto-generated ASPECTS score results are available, the physician is asked to confirm that the case in question is for an ICA or MCA occlusion and is able to modify/verify the ASPECTS regional score. The ASPECTS auto-generated results, including the ASPECTS score, indication of affected side, affected ASPECTS regions and voxel-wise analysis (shown as a heatmap of voxels 'contributing to ASPECTS score'), along with the user-verified/modified result summary can be sent to the Picture Archiving and Communications System (PACS).

Here's an analysis of the acceptance criteria and study that proves the device meets those criteria, based on the provided FDA 510(k) Clearance Letter.

Acceptance Criteria and Device Performance

The provided text details two primary performance studies: Standalone Performance and Clinical Validation (MRMC study), along with a Clinical Validation of Voxels Contributing to ASPECTS (VCTA). The acceptance criteria are implicitly derived from the reported performance benchmarks for these studies.

1. Table of Acceptance Criteria and Reported Device Performance

2. Sample Size and Data Provenance

Standalone Performance Test Set:

• Sample Size: 200 non-contrast CT scans.
• Data Provenance: Patients were from multiple clinical sites, including 77 from Canada, 59 from the United States, 51 from Europe, and 13 from Asia-Australia. The cases are implied to be retrospective, as they are being used for validation after data collection.

MRMC Clinical Validation Test Set:

• Sample Size: 100 non-contrast CT scans.
• Data Provenance: 50% of the patients were from 11 sites in Canada, and the other 50% were from 12 sites in the United States. Implied retrospective, used for validation.

VCTA Clinical Validation Test Set:

• Sample Size: Not explicitly stated, but derived from "the proportion of cases." It would be part of the same test sets or a subset thereof.

3. Number of Experts and Qualifications for Ground Truth

• Standalone Performance: Not explicitly stated how many experts established the ground truth, but it refers to an "expert consensus reference standard" for the primary standalone performance assessment. The types of experts are implied to be medical professionals capable of scoring ASPECTS.
• MRMC Clinical Validation: For the reader study, the ground truth was established through a "reference standard." The text states that "the results showed statistically significant improvements in the agreement between the readers and a reference standard." The number and qualifications of experts establishing this specific reference standard are not detailed, though it's likely a panel of expert radiologists or neurologists.
• VCTA Clinical Validation: "Expert neuroradiologist assessment of ischemic tissue." The number of neuroradiologists involved is not specified, but the term "consensus score" implies more than one. No specific years of experience or board certifications are mentioned.

4. Adjudication Method for the Test Set

• The term "expert consensus reference standard" or "consensus score" (for VCTA) is used, implying an adjudication process to arrive at the ground truth. However, the exact method (e.g., 2+1, 3+1, majority vote, etc.) is not explicitly described for any of the studies.

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

• Yes, an MRMC study was done.
• Effect Size of Human Reader Improvement with AI vs. without AI assistance:
  • AUC Improvement: A statistically significant improvement of 5.7% from 68.6% (unaided) to 74.3% (aided) (p-value<0.001).
  • Sensitivity Improvement: A statistically significant improvement of 9.7% from 41.3% (unaided) to 51.0% (aided) (p-value<0.001).
  • Specificity Improvement: A statistically significant improvement of 1.6% from 95.9% (unaided) to 97.5% (aided) (p-value<001).
  • Overall Accuracy Improvement: Improved by 2.6% from 89.5% (unaided) to 92.0% (aided) (p-value<001).
  • Inter-reader Consistency (Fleiss's Kappa): Increased by 28.5% from 32.3% (unaided) to 60.8% (aided).
  • Reduction in performance variation: The range in AUC between users was narrower with StrokeSENS ASPECTS than unassisted.

6. Standalone (Algorithm Only) Performance Study

• Yes, a standalone performance study was done.
• The results are detailed in the "Standalone Performance" section, including AUC-ROC (90.9%), accuracy (90.6%), sensitivity (70.6%), and specificity (93.9%).

7. Type of Ground Truth Used

• Expert Consensus: The primary ground truth for the standalone performance and MRMC studies was based on an "expert consensus reference standard."
• Neuroradiologist Assessment: For the VCTA validation, ground truth was derived from "expert neuroradiologist assessment of ischemic tissue."
• The ground truth in all cases relates to the ASPECTS score and identification of affected ASPECTS regions, which is a radiological assessment. It does not mention pathological or long-term outcomes data as the primary ground truth.

8. Sample Size for the Training Set

• Not explicitly stated in the provided text. The document focuses on the performance and validation datasets. The number of cases used to train the "machine learning algorithms" or "artificial intelligence algorithm" is not disclosed in this summary.

9. How Ground Truth for Training Set Was Established

• Not explicitly stated in the provided text. Given it's an AI/ML algorithm, the training set would also require labeled data (ground truth). However, the method for establishing this ground truth (e.g., expert consensus, single expert, automated labeling) is not described in this document.

Ask a specific question about this device