Rapid ICH is a radiological computer aided triage and notification software in the analysis of non-enhanced head CT images. The device is intended to assist hospital networks and trained radiologists in workflow triage by flagging and communication of suspected positive findings of pathologies in head CT images, for IPH, IVH, SAH, and SDH Intracranial Hemorrhages (CH).

Rapid ICH uses an artificial intelligence algorithm to analyze images and highlight cases with detected ICH 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 ICH findings. Notifications include compressed preview images, which 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 is a a diagnostic device.

The results of Rapid ICH are intended to be used in conjunction and based on professional judgment, to assist with trage /prioritization of medical images. Notified radiologists are responsible for viewing full images per the standard of care.

Rapid ICH is a radiological computer-assisted triage and notification software device. The Rapid ICH module is a non-enhanced CT (NCCT) processing module which operates within the integrated Rapid Platform to provide triage and notification of suspected intracranial hemorrhage. The Rapid ICH module is an AI/ML module. The output of the module is a priority notification to clinicians indicating the suspicion of ICH based on positive findings. The Rapid ICH module uses the basic services supplied by the Rapid Platform including DICOM processing, job management, imaging module execution and imaging output including the notification and compressed image.

Here's a breakdown of the acceptance criteria and study details for the Rapid ICH device, based on the provided text:

Acceptance Criteria and Device Performance

The primary performance goals for Rapid ICH were defined by sensitivity and specificity thresholds.
Acceptance Criteria Table and Reported Device Performance:

Study Details

2. Sample Size and Data Provenance:

• Test Set Sample Size: 314 cases (148 ICH positive, 166 ICH negative).
• Data Provenance: Retrospective, multicenter, multinational study. Specific countries are not detailed, but "multinational" implies diverse geographical origins.

3. Number of Experts and Qualifications for Ground Truth:

• Number of Experts: Not explicitly stated how many individual experts established the ground truth. The document mentions "expert reader truthing of the data," suggesting one or more experts.
• Qualifications of Experts: The document states "trained radiologists" are intended users and mentions "expert reader truthing." However, specific qualifications such as years of experience, board certification, or subspecialty are not provided for the ground truth experts.

4. Adjudication Method for the Test Set:

• The document implies ground truth was established by "expert reader truthing of the data," but does not specify an adjudication method (e.g., 2+1, 3+1, consensus review process if multiple readers were involved).

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

• No, an MRMC comparative effectiveness study was NOT mentioned for evaluating human readers' improvement with AI assistance. The study focused on the standalone performance of the AI algorithm (accuracy) and the time-to-notification benefit.

6. Standalone Performance (Algorithm Only):

• Yes, a standalone performance study was done. The reported sensitivity, specificity, and AUC values directly reflect the algorithm's performance in identifying ICH presence. The study evaluated the software's performance in identifying abnormalities, and the "time to notification" indicates the speed of the algorithm's output.

7. Type of Ground Truth Used:

• Expert Consensus: The ground truth for the test set was established through "expert reader truthing of the data." This implies a clinical expert (radiologist) determined the presence or absence of ICH.

8. Sample Size for the Training Set:

• The document states that the "minor change causing this filing, is the use of additional data for training and validation," implying the training set for this iteration of the device included more data than the predicate. However, the specific sample size of the training set is not provided in the summary.

9. How the Ground Truth for the Training Set was Established:

• Similar to the test set, the document indicates that the device was trained and validated using "retrospective case data based on expert reader truthing of the data." This suggests the ground truth for the training set was also established by expert review/diagnosis by clinical experts.

Ask a specific question about this device

RAPID ICH is a radiological computer aided triage and notification software indicated for use in the analysis of nonenhanced head CT images.

The device is intended to assist hospital networks and trained radiologists in workflow triage by flagging and communication of suspected positive findings of pathologies in head CT images, namely Intracranial Hemorrhage (ICH). RAPID ICH uses an artificial intelligence algorithm to analyze images and highlight cases with detected ICH 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 ICH 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 ICH 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.

RAPID ICH is a clinical module which operates within the integrated RAPID Platform to provide triage and notification of suspected intra-cranial hemorrhage (ICH). The RAPID ICH module consists of the core RAPID Platform software which provides the administration and services for the RAPID; and the RAPID ICH module which functions as one of many image processing modules hosted by the platform.

The RAPID platform is a software package that provides for the visualization and study of changes in tissue using digital images captured by diagnostic imaging systems including CT (Computed Tomography), CTA, XA3D and MRI (Magnetic Image Resonance), as an aid to physician diagnosis. RAPID can be installed on a customer's Server or it can be accessed online as virtual system. It provides viewing, quantification, analysis and reporting capabilities. The RAPID platform has multiple modules a clinician may elect to run and provide analysis for decision making.

RAPID ICH provides an automatic analysis of received NCCT scan data for the triage and notification of ischemic hemorrhage (ICH). The application is selected via DICOM encoding which is processed by the DICOM handler within the RAPID Platform. Once the DICOM server identifies the selected ICH processing modality, the Job Manager initiates image processing using the ICH Module.

Upon processing a case with suspicion of ICH, the clinical team is notified via messaging, the case has a suspicion of ICH. The notifications are sent to the PACS/Workstation, via email and to a mobile application. The notification provides the attending clinical team physician that suspicion of hemorrhage has been identified and immediate attention to the case should be given. The messaging provides notification and compressed image data of the case. The compressed preview is informational only and labeling identifies it as not to be used for diagnostic use and to review the data within the PACS/Workstation prior to making any diagnostic decisions. No additional information markings are added to the case.

The notifications are pop-up messages or email with the appropriate case information and suspicion or non-suspicion of ICH labelled. In all cases, the normal standard of care workflow is adhered to, this ensures the case is still reviewed when non-suspicion is determined.

Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided FDA 510(k) summary for RAPID ICH:

Acceptance Criteria and Device Performance

1. Table of Acceptance Criteria and the Reported Device Performance

2. Sample Size and Data Provenance

• Test Set Sample Size: 336 cases
• Data Provenance: Retrospective, blinded, multicenter, multinational study.
  • Countries of Origin: US (4 Sites and 1 Multisite Study) and OUS (1 site, 1 multisite).
  • Nature of Study: Retrospective.
• Case Distribution: Approximately an equal number of positive cases (images with ICH) and negative cases (images without ICH) were included in the analysis.

3. Number of Experts and Qualifications for Ground Truth

The document states "reader truthing of the data" was used, but it does not explicitly state the number of experts or their specific qualifications (e.g., "radiologist with 10 years of experience") for establishing the ground truth of the test set.

4. Adjudication Method for the Test Set

The document mentions "reader truthing," but it does not specify the adjudication method used (e.g., 2+1, 3+1, none) for the test set.

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

No MRMC comparative effectiveness study was done or reported. The study focused on the standalone performance of the AI algorithm. Therefore, there is no information on how human readers improve with AI vs without AI assistance.

6. Standalone (Algorithm Only) Performance Study

Yes, a standalone (algorithm only) performance study was conducted. The entire "Performance Data" section describes the RAPID ICH software's ability to identify ICH independently, reporting its sensitivity and specificity.

7. Type of Ground Truth Used

The ground truth for the test set was established through "reader truthing of the data." This implies expert consensus or expert interpretation as the basis for ground truth, rather than pathology or outcomes data.

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 the device uses a "deep learning algorithm trained on medical images."

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

The document does not explicitly detail how the ground truth for the training set was established. It generally states the algorithm was "trained on medical images," implying that these images were annotated or labeled with ground truth information (e.g., presence or absence of ICH) by experts, but the process or expert involvement is not described.

Ask a specific question about this device