(182 days)
The Viz HDS device is intended for automatic labeling, visualization, and quantification of segmentable brain structures from a set of Non-Contrast CT (NCCT) head scans. The software is intended to automate the current manual process of identifying, labeling, and quantifying the volume of segmentable brain structures identified on NCCT images. Viz HDS provides volumes from NCCT scans acquired at a single time point. The Viz HDS software is indicated for use in the analysis of the following structures: Intracranial Hyperdensities, Lateral Ventricles and Midline Shift. The device output should be reviewed along with patient's original images by a physician.
Viz HDS is a software-only device that uses a locked artificial intelligence machine learning (AI/ML) algorithm to processes non-contrast head CT scans to outline intracranial hyperdensity areas, lateral ventricles (right and left), midline shift, and then quantify the volume of intracranial hyperdensity(ies), volume of lateral ventricle asymmetry ratio and distance of midline shift.
Viz HDS analyzes the head NCCT series in DICOM format and produces a summary series and a segmentation series in DICOM format. The summary series is a two-slice output: a single slice from the NCCT series with segmented areas overlaid on it, and a summary table providing the calculated measurements. The segmentation series shows an RGB overlay, on each slice of the input series, of the lateral ventricles and hyperdensity(ies) segmentation masks and a midline shift. For slices including hyperdensities or ventricle/ventricles, its volume would be mentioned in a color legend that is also overlaid on the slice. The colors are only for visual differentiation between the segmented regions, the colors don't have a meaning on their own. The device output is exported in DICOM format, which is sent to a pre-configured PACS destination together with the original NCCT series for review by a physician to aid in the assessment of measuring intracranial hyperdensity(ies), lateral ventricles, and midline shift.
Here's a breakdown of the acceptance criteria and the study details for the Viz HDS device, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
Acceptance Criteria and Device Performance for Viz HDS
| Measurement Trait | Acceptance Criteria (Upper 95% CI Bound) | Reported Device Performance (Upper 95% CI Bound) |
|---|---|---|
| Hyperdensities Total Volume (MAE) | ≤ 7.5 mL | 70% (Lower CI Bound) |
| Both Lateral Ventricles (DICE Score) | > 70% (Lower CI Bound) | > 70% (Lower CI Bound) |
Note: The text explicitly states "less than 7.5 mL" and "greater than 70%", confirming the device met the specified criteria.
2. Sample Size and Data Provenance for Test Set
- Sample Size for Test Set: Not explicitly stated in the provided text.
- Data Provenance: Not explicitly stated in the provided text. The text mentions "clinical site" for stratification, but not the origin of the data itself (e.g., country, specific hospitals).
- Retrospective or Prospective: Not explicitly stated.
3. Number and Qualifications of Experts for Ground Truth (Test Set)
- Number of Experts: Not explicitly stated. The text mentions "trained radiologists" was involved in establishing the ground truth.
- Qualifications of Experts: The experts were "trained radiologists." No further details on their experience (e.g., "10 years of experience") are provided.
4. Adjudication Method (Test Set)
- Adjudication Method: Not explicitly stated. The text only mentions that "ground truth as established by trained radiologists." It does not detail how disagreements among radiologists, if any, were resolved.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- MRMC Study: No, an MRMC comparative effectiveness study was not explicitly mentioned as being performed. The study described compares the device's output to "ground truth as established by trained radiologists" and does not describe a scenario where human readers' performance with and without AI assistance was measured.
- Effect Size of Human Readers Improvement: Not applicable, as an MRMC study was not performed.
6. Standalone Performance (Algorithm Only)
- Standalone Performance: Yes, a standalone performance study was done. The study compares the Viz HDS's output directly to the established ground truth. This is a measure of the algorithm's performance without direct human intervention in the measurement process itself, although the output is intended for physician review.
7. Type of Ground Truth Used
- Type of Ground Truth: The ground truth was established by "expert consensus" from trained radiologists. The text states, "ground truth as established by trained radiologists."
8. Sample Size for Training Set
- Sample Size for Training Set: Not explicitly stated in the provided text.
9. How Ground Truth for Training Set Was Established
- Ground Truth for Training Set: Not explicitly stated in the provided text.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).