Search Results

TribusConnect is a medical image management and processing device which facilitates remote viewing on different locations of CT images stored in DICOM format, allowing healthcare professionals to access, review, manipulate, measure and visualize simulated patient data from different sources.

TribusConnect is suitable for, but not limited to, pre-procedural planning of cardiac interventions and post procedural analysis of cardiac interventions.

To facilitate the above, TribusConnect uses CT DICOM data to output volume renderings, ML-based heart segmentation, a simulated fluoroscopy view, endo views, MIP and MPR views. Measurements can only be performed on MPR views.

TribusConnect is not intended for diagnostic use on mobile displays.

TribusConnect is not intended to serve as the primary archive for medical imaging data.

TribusConnect is an adjunct tool and is not intended to replace a physician's own review on the images.

The intended patient population is comprised of adult patients (22 years of age and older)

TribusConnect is a cloud-based software application that facilitates remote viewing on different locations of CT images stored in DICOM format, allowing healthcare professionals to access, review, manipulate, measure and visualize simulated patient data from different sources through a secure internet connection. TribusConnect offers a secure and efficient solution for accessing, reviewing, and visualizing CT images in DICOM format. Measurements can be performed on MPR views. TribusConnect is provided to its users as a Saas solution, TribusMed will be in control of installation, configuration, updates etc.

TribusConnect can be used in healthcare facilities, hospitals, and remote locations, including home environments, when accessed by authorized users through a secure internet connection.

CT scans can be uploaded to the server and will be de-identified before the upload. On the server the heart structures will be segmented using a ML based algorithm. Also a ML based semi-automated localization of the Left Atrial Appendage (LAA) will be performed.

All rendering of image data is done on the server te results can be viewed on standard computing devices (e.g., desktop, laptop, tablet, smartphone) with an internet connection.

Images can be rendered as Volume rendering, segmentations, simulated fluoroscopy, endo view, MIP or MPR. The user can interact with the data like zooming, rotating, changing the window. On the MPR views measurements can be performed.

The device's cloud infrastructure and encryption protocols ensure data protection and compliance with security standards.

Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

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

• Test Set Sample Size: 100 CTA volumes.
• Data Provenance:
  • Country of Origin: Collected from 14 sites across the United States.
  • Retrospective/Prospective: Retrospectively gathered.
  • Anonymization: Data was anonymized before being shared with TribusMed.
  • Uniqueness: Each dataset is acquired from a unique patient.
  • Exclusion from Training: Sites providing data for the training dataset are different from the validation dataset (no sites used for training are used for validation).

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

• Number of Experts: 2.
• Qualifications: US board certified radiologists.

4. Adjudication Method for the Test Set

The text states that two US board certified radiologists qualitatively evaluated the performance of heart segmentation. It does not explicitly mention an adjudication method (e.g., 2+1, 3+1). It describes direct qualitative evaluation by these two experts. For LAA localization, the validation was done by comparing results with the predicate device, TruPlan, rather than human expert consensus on the test set.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done

No MRMC comparative effectiveness study involving human readers assisting with or without AI was explicitly described for the performance of TribusConnect's AI features (heart segmentation, LAA localization). The study validated the algorithm's performance against expert qualitative review (for segmentation) and comparison to a predicate device (for LAA localization).

6. If a Standalone (algorithm only without human-in-the-loop performance) was done

Yes, the validation of the machine learning-based algorithms for heart segmentation and LAA localization focused on the standalone performance of the algorithm without a human in the loop, by comparing its output to established ground truth (expert qualitative review or predicate device computed results).

7. The Type of Ground Truth Used

• Heart Segmentation: Expert consensus/qualitative review by 2 US board certified radiologists.
• LAA Localization: Comparison to the predicate device TruPlan (K222593)'s computed landing zone contour center.
• Manual Quantitative Imaging Functions: Phantom studies compared against "true value."

8. The Sample Size for the Training Set

The document does not explicitly state the sample size for the training set. It only mentions that the sites used for the training dataset were different from those used for the validation dataset.

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

The document does not provide details on how the ground truth for the training set was established. It only implies the use of a training dataset distinct from the validation dataset.

Ask a specific question about this device