Search Results

3D inVIVO™ is a PACS System used to acquire, transmit, view, and store image or patient data. This data can be transmitted, stored, and viewed over a computer network or off-site using an Internet connection. The typical users of this system are trained professionals including radiologist, physicians, technicians, and nurses.

The 3D inVIVO™ is indicated for capture and storage of 2D images from ultrasound systems and reconstructing them into 3D ultrasound images. These images provide an approximate representation of the 3D volume and are not intended for use in diagnosis or quantitative measurements.

The CSIST 3D inVIVO™ interactive 3D diagnostic ultrasound system (3DUS) acquires a set of successive 2D ultrasound images from Ultrasound system and coverts them into a 3D ultrasound image, stores the 3DUS images in the Image Archive of DICOM Image Server, views patient data and medical images via the Intranet or Internet, supplies image availability information to the Department System Scheduler/Order Filler with DICOM 3.x standard, and accept schedule information and updated procedure information from the Department System Scheduler/Order Filler with HL7 version 2.4 standard.

CSIST Interactive 3DUS system is a PC-based Modality with operating system of MS Windows NT/2000/XP. 3D inVIVO™ provides the advanced process functions, such as :

• Supports the freehand scanning, 3D image acquisition, and 3D reconstruction.
• Acquires images from ultrasound system in real time.
• Supports the ultrasound image storage and ultrasound multi-frame image storage.
• Converts non-DICOM to DICOM format.
• Supports the GUI functions: zoom, pan, pseudo-color setting, threshold setting, Gamma correction, and thumbnail.
• Provides the DICOM query and retrieve services (images and presentation states).
• Displays the cut planes of 3D image along the long axis or the short axis.
• Provides 3D rendering modes: MIP, X-ray, translucent, surface, slicing, and 3D cine.
• Supports distance/area/angle measurements.
• Supports Tele-medicine (audio, video, and text) between regional medicine center and remote clients.

The provided text does not contain detailed information about specific acceptance criteria or a study that proves the device meets those criteria in the form of a performance study. The 510(k) summary focuses on establishing substantial equivalence to a predicate device rather than presenting a performance study with quantitative acceptance criteria.

However, based on the provided text, we can infer some general "acceptance criteria" related to functionality and safety, and describe what the document does say about meeting these.

Here's an attempt to structure the answer based on the information available, while acknowledging what is not present:

1. A table of acceptance criteria and the reported device performance

The document does not explicitly present a table of acceptance criteria with corresponding performance metrics like sensitivity, specificity, accuracy, etc., which would typically be found in a performance study for AI-powered diagnostic devices. The acceptance criteria here are more related to functional equivalence and safety.

2. Sample size used for the test set and the data provenance

The document does not mention any specific test set, sample size, or data provenance (e.g., country of origin, retrospective/prospective) for evaluating the performance of the device against quantitative metrics. This 510(k) submission primarily relies on demonstrating functional equivalence to a predicate device and software verification/validation, rather than a clinical performance study with patient data.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

This information is not provided as the submission does not detail a study involving expert-established ground truth for a test set of medical images.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

This information is not provided as no test set with expert-established ground truth is described.

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

No MRMC comparative effectiveness study is mentioned in the provided text. The device's indications for use explicitly state that the images "are not intended for use in diagnosis or quantitative measurements," which further suggests that a study comparing human reader performance with and without device assistance for diagnostic tasks would not be relevant to this specific 510(k) submission.

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

No standalone performance study is described for the algorithm. Given the device's purpose as an image archiving, transmission, and visualization system for approximate 3D reconstruction, and its explicit statement that it's "not intended for use in diagnosis or quantitative measurements," a standalone diagnostic performance study would not be applicable or expected for this submission. The device is a tool for visualizing and managing ultrasound data, with human interpretation always in the loop.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

No specific ground truth type is mentioned or implied as there is no detailed performance study described that would require ground truth data for diagnostic or quantitative purposes. The "ground truth" for this device's acceptance is primarily its functional consistency with its description and substantial equivalence to the predicate device, as well as adherence to standards (DICOM, HL7).

8. The sample size for the training set

The document does not mention any training set size as it pertains to a traditional machine learning or AI model development that would typically require a training set. The "3D inVIVO™" is described as an "interactive 3D diagnostic ultrasound system" that converts 2D images into 3D, and provides various processing and rendering functions. While 3D reconstruction can involve algorithms, the submission doesn't present it as a deep learning or AI model in the modern sense requiring a large labeled training set for diagnostic purposes.

9. How the ground truth for the training set was established

As no training set is described (see point 8), no information is provided on how ground truth for a training set was established.

Ask a specific question about this device