Search Results

Rapid OH is a radiological computer aided triage and notification software indicated for suspicion of Obstructive Hydrocephalus (OH) in non-enhanced CT head images of adult patients. The device is intended to assist trained clinicians in workflow prioritization triage by providing notification of suspected findings in head CT images.

Rapid OH uses an artificial intelligence algorithm to analyze images and highlight cases with suspected OH on a server or standalone desktop application in parallel to the ongoing standard of care image interpretation. The user is presented with notifications for cases with suspected OH findings. Notifications include compressed preview images, that are meant for informational purposes only and not intended for diagnostic use beyond notification. The device does not alter the original medical image and is not intended to be used as a diagnostic device.

The results of Rapid OH are intended to be used in conjunction with other patient information and based on professional judgment, to assist with triage/prioritization of medical images. Notified clinicians are responsible for viewing full images per the standard of care.

Contraindications/Limitations/Exclusions:

• Rapid OH is intended for use for adult patients.
• Input data image series containing excessive patient motion or metal implants may impact module analysis accuracy, robustness and quality.
• Ventriculoperitoneal shunts are contraindicated

Exclusions:

• Series with missing slices or improperly ordered slices
• data acquired at x-ray tube voltage < 100kVp or > 140kVp.
• data not representing human head or head/neck anatomical regions

Rapid OH software device is a radiological computer-aided triage and notification software device using AI/ML. The Rapid OH device is a non-contrast CT (NCCT) processing module which operates within the integrated Rapid Platform to provide a notification of suspected findings of obstructive hydrocephalus (OH). The Rapid OH device is SaMD which analyzes input NCCT images that are provided in DICOM format for notification of suspected findings for workflow prioritization.

Here's a breakdown of the acceptance criteria and study details for the Rapid OH device, based on the provided FDA 510(k) clearance letter:

1. Table of Acceptance Criteria and Reported Device Performance

Note: The document states "Standalone performance primary endpoint passed with sensitivity (Se) of 89.5% (95% CI:0.837-0.935) and specificity (Sp) of 97.6% (95% CI:0.940-0.991)". While explicit numerical acceptance criteria for sensitivity and specificity are not provided, the "passed" statement implies that the reported performance fell within pre-defined acceptable ranges or met a statistical hypothesis.

2. Sample Size for the Test Set and Data Provenance

• Sample Size for Test Set: 320 cases
• Data Provenance: The document mentions "diversity amongst demographics (M: 45%, F: 54%); Sites (and manufacturers (GE, Philips, Siemens, Toshiba) and confounders (ICH, Ischemic Stroke, Tumor, Cyst, Aqueductal stenosis, Mass effect, Brain atrophy and Communicating hydrocephalus)". While specific countries of origin are not explicitly stated, the mention of multiple manufacturers (Siemens, GE, Toshiba, Philips) and multiple sites (74 sites for algorithm development, and "Sites" for the validation set) suggests a diverse, likely multi-site, and potentially multi-country dataset, although this is not definitively confirmed for the test set itself. The dataset appears to be retrospective, as it's used for algorithm development and validation based on existing cases.

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

• Number of Experts: 3 experts (implied from "Truthing was established using 2:3 experts.")
• Qualifications of Experts: Not explicitly stated in the provided text. They are referred to as "experts." In regulatory contexts, these would typically be radiologists or neuro-radiologists with significant experience in interpreting head CTs.

4. Adjudication Method for the Test Set

• Adjudication Method: "2:3 experts." This means that ground truth was established by agreement from at least 2 out of 3 experts.

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

• MRMC Study Done: No, an MRMC comparative effectiveness study was not explicitly mentioned for this device. The study described is a standalone performance validation of the algorithm.

6. Standalone Performance (Algorithm Only without Human-in-the-Loop) Done

• Standalone Performance Done: Yes, "Final device validation included standalone performance validation. This performance validation testing demonstrated the Rapid OH device provides accurate representation of key processing parameters under a range of clinically relevant conditions associated with the intended use of the software." The reported sensitivity and specificity values are for this standalone performance.

7. Type of Ground Truth Used

• Type of Ground Truth: Expert consensus ("Truthing was established using 2:3 experts.")

8. Sample Size for the Training Set

• Sample Size for Training Set: 3340 cases (This refers to "Algorithm development" which encompasses training and likely internal validation/development sets).

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

• How Ground Truth Was Established (Training Set): The document states "Algorithm development was performed using 3340 cases... Truthing was established using 2:3 experts." This implies that the same expert consensus method (2 out of 3 experts) used for the test set was also used to establish ground truth for the cases used in algorithm development (which includes the training set).

Ask a specific question about this device