Search Results

Neurophet AQUA is intended for Automatic labeling, visualization and volumetric quantification of segmentable brain structures from a set of MR images. Volumetric data may be compared to reference percentile data.

Neurophet AQUA is a fully automated MR imaging post-processing medical device software that provides automatic labeling, visualization, and volumetric quantification of brain structures from a set of MR images and returns segmented images and morphometric reports. The resulting output is provided in morphometric reports that can be displayed on Picture Archive and Communications Systems (PACS). The high throughput capability makes the software suitable for use in both clinical trial research and routine patient care as a support tool for clinicians in assessment of structural MRIs. Neurophet AQUA provides morphometric measurements based on T1 MRI series. The output of the software includes volumes that have been annotated with color overlays, with each color representing a particular segmented region, and morphometric reports that provide comparison of measured volumes to age and gender-matched reference percentile data. Neurophet AQUA processing architecture includes a proprietary automated internal pipeline that performs segmentation, volume calculation and report generation. The results are displayed in a dedicated graphical user interface, allowing the user to: Browse the segmentations and the measures, Compare the results of segmented brain structures to a reference healthy population, Read and print a PDF report Additionally, automated safety measures include automated quality control functions, such as tissue contrast check, scan protocol verification, which validate that the imaging protocols adhere to system requirements.

Here's a breakdown of the acceptance criteria and the study details for Neurophet AQUA, based on the provided text:

1. Acceptance Criteria and Reported Device Performance

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size for Accuracy Test Set: 64 T1 scans (36 US-based, 40 females, age range 20-90)
• Sample Size for Reproducibility Test Set: 50 repeated T1 scans (31 US-based, 23 females, age range 10-90)
• Data Provenance: Both test sets included retrospective data from various sources:
  • 36 of 64 (56%) scans for accuracy were US-based data.
  • 31 of 50 (62%) scans for reproducibility were US-based data.
  • The test sets included cognitive normal, mild cognitive impairment, and Alzheimer's disease patients.
  • Data was acquired from MR scanners of three main vendors (Siemens, Phillips, and GE).
  • The document explicitly states that "All the testing data was exclusive from the training dataset."

3. Number of Experts Used to Establish Ground Truth for the Test Set and Their Qualifications

• The text states, "Ground-truth data were initially generated using FreeSurfer (General Hospital Corporation, Boston, MA, USA, version 6.0) and verified and corrected by four radiologists."
• Qualifications of Experts: Four radiologists. Specific experience level (e.g., years of experience) is not provided in the document.

4. Adjudication Method for the Test Set

• The ground truth for the training set was initially generated by FreeSurfer and then "verified and corrected by four radiologists." This implies a consensus or expert review process, where the radiologists likely reviewed and refined the FreeSurfer outputs. The specific adjudication method (e.g., 2+1, 3+1) is not explicitly detailed, but it indicates multiple expert review and correction.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

• No, a multi-reader multi-case (MRMC) comparative effectiveness study comparing human readers with and without AI assistance was not reported in this summary. The study focused on the standalone performance of the AI device against expert manual segmentations (ground truth) and reproducibility.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

• Yes, a standalone performance study was conducted. The performance metrics (Dice's coefficient and percentage absolute volume differences) directly evaluate the algorithm's output against the established ground truth without involving human-in-the-loop performance for the reported results. The device "yields reproducible results that are well correlated with expert manual segmentations," indicating an evaluation of the device's output itself.

7. The Type of Ground Truth Used

• The ground truth used was expert consensus / semi-automated expertise. It was "initially generated using FreeSurfer" (a widely-accepted brain segmentation software) and then "verified and corrected by four radiologists."

8. The Sample Size for the Training Set

• 300 T1-weighted MRI scans.

9. How the Ground Truth for the Training Set Was Established

• The ground truth for the training set was established in the same manner as the ground truth for the test set: "Ground-truth data were initially generated using FreeSurfer (General Hospital Corporation, Boston, MA, USA, version 6.0) and verified and corrected by four radiologists."
• These 300 scans were collected from ten different MRI scanner types and included public datasets such as ADNI, IXI, PPMI, HCP, and AIBL.

Ask a specific question about this device