Search Results

Tempus Pixel software is a medical diagnostic application that displays, processes, stores, and transfers DICOM and non-DICOM medical data. It provides the capability to store images and patient information, and perform filtering, digital manipulation, and quantitative measurements. The client software is designed to run on standard personal and business computers and on monitors/screens that meet appropriate technical specification for image diagnosis.

Tempus Pixel includes an optional Pixel Cardio module which is used to analyze the heart and its major vessels using multi-slice, multi-phase, and velocity-encoded cardiovascular magnetic resonance (MR) images. It provides clinically relevant and reproducible, quantitative data, and has been tested and validated on MR images acquired from both 1.5T and 3.0T MR Scanners.

Tempus Pixel includes an optional Pixel Therapy Response Evaluation module which provides analytical tools to help the user assess and document changes in morphological activity at diagnostic and therapy follow-up examinations. It is a tool used to support the oncological workflow by helping the user confirm the absence or presence of lesions, including evaluation, quantification, follow-up, and documentation of any such lesions.

Tempus Pixel software is intended to be used as a support tool by trained healthcare professionals to aid in diagnosis. It is intended to provide image and related information that is interpreted by a trained professional to render findings and/or diagnosis, but it does not directly generate any diagnosis or potential findings.

Tempus Pixel is a dedicated software application for evaluating various DICOM modalities, including multi-slice and multi-phase computed tomography (CT), magnetic resonance (MR) images, ultrasound, x-ray, and mammography. Pre-existing images are uploaded into Tempus Pixel in a DICOM format from a PACS or a scanner. Tempus Pixel is completely hosted in the cloud and is accessed using a Google Chrome™ web browser by navigating to the following https://app.arterys.com. Cloud servers are provided by the Google Cloud Platform and service is accessible globally.

The Pixel application is designed around a modular architecture of separate components that make up a basic image viewer. These include the studylist, from which studies are selected and opened, the image display (2D, 3D, MIP, etc), view synchronization, metadata information, and various navigational, measurement, and other action tools.

Functionality provided by the Pixel application is extended by additional Cardio and Therapy Response Evaluation application modules which add support for specific clinical workflows. The Cardio module includes support for 4D flow studies and provides additional functionality specific to cardiac studies. The Therapy Response Evaluation module includes functionality specific to evaluating and measuring lesions.

The device also facilitates several FDA-cleared third party inference models. Each third party inference integration implements a set of filters that determine whether a given study or workflow is eligible to be submitted to the third-party inference model. If eligible, a request is queued and processed by the third party model. Results are returned to Pixel from the third party model and are stored in the study collection.

The clinician retains the ultimate responsibility for making the pertinent diagnosis based on their standard practices and visual comparison of images. Tempus Pixel software is a complement to these standard procedures.

Acceptance Criteria and Study for Tempus Pixel Device

This report details the acceptance criteria and the study conducted to demonstrate that the Tempus Pixel device meets these criteria, based on the provided FDA 510(k) Clearance Letter.

1. Table of Acceptance Criteria and Reported Device Performance

The provided document focuses on the performance validation of the T1/T2 Inline Map functionality of the Tempus Pixel device.

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

For T1/T2 Inline Map Validation:

• Sample Size (Test Set):
  • T1: 30 series, with 4 ROIs assessed per series, resulting in a total of 120 ROIs.
  • T2: 30 series, with 4 ROIs assessed per series, resulting in a total of 120 ROIs.
• Data Provenance: The document states that the validation was performed by comparing Tempus Pixel generated inline maps to "maps generated from FDA-cleared MR scanners using the same input series." This indicates the use of retrospective clinical reference data acquired from various commercial MR scanners. The document lists the specific GE, Philips, and Siemens MR scanner models (both 1.5T and 3T) used for comparison, implying a diverse set of data origins within the general scope of clinical practice operating these systems. The country of origin is not explicitly stated but can be inferred to be from medical institutions where these FDA-cleared scanners are used (likely US and potentially international).

For Human Factors/Design Validation:

• Sample Size (Test Set): Not explicitly stated how many "studies" were used, but the validation involved three physicians. The focus was on the user interface and functionality with the new inline maps.

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

For T1/T2 Inline Map Validation:

• Number of Experts: Not applicable in the traditional sense, as the ground truth was established by FDA-cleared MR scanners themselves. The "ground truth" for this test was the T1/T2 inline maps generated directly by these commercial MR scanners using the same input series.
• Qualifications of Experts: Not applicable, as expert consensus was not the primary method for ground truth establishment for this specific test.

For Human Factors/Design Validation:

• Number of Experts: Three physicians.
• Qualifications of Experts: Described as "physicians experienced in assessing and quantifying parametric mapping data within Pixel Cardio software."

4. Adjudication Method for the Test Set

For T1/T2 Inline Map Validation:

• Adjudication Method: Not applicable. The comparison was a direct quantitative assessment (absolute error) between the device's output and the reference output from the FDA-cleared MR scanners. There was no mention of an adjudication process involving human reviewers for numerical discrepancy resolution.

For Human Factors/Design Validation:

• The document implies a qualitative assessment by three physicians. While "no deviations or unexpected results" were observed, no specific adjudication method (e.g., 2+1, 3+1) is described for resolving potential discrepancies among the three physicians' observations. It seems their individual assessments collectively confirmed the acceptance criteria were met.

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

• No, an MRMC comparative effectiveness study was not done. The document explicitly states: "No clinical studies were performed for the modified Tempus Pixel device." The validation focused on the standalone performance of the T1/T2 inline map generation against scanner-generated maps and human factors of the user interface. There is no mention of comparing human reader performance with and without AI assistance from Tempus Pixel.

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

• Yes, a standalone performance study was done for the T1/T2 Inline Map validation. The "Summary of Non-Clinical Studies" section describes comparing the "Tempus Pixel generated inline maps" (algorithm output) directly to "maps generated from FDA-cleared MR scanners" (reference standard). This is a direct measure of the algorithm's accuracy without human intervention influencing the map generation process itself.

7. The Type of Ground Truth Used

For T1/T2 Inline Map Validation:

• The ground truth used was "clinical reference data" in the form of T1/T2 maps generated by commercially available, FDA-cleared MR scanners using the same input series. This serves as a highly reliable, established standard for quantitative T1/T2 mapping.

For Human Factors/Design Validation:

• The ground truth was established based on pre-defined user needs/specifications and the subjective assessment of functionality by experienced physicians, ensuring the device met its intended usability goals.

8. The Sample Size for the Training Set

• The document does not provide information regarding the sample size used for the training set. The validation described is for the T1/T2 inline map generation, which is a new functionality. While the device incorporates "deep learning model" for LV contouring in the predicate (mentioned in the "T1 and T2" characteristic comparison), the specifics of the training data for any underlying AI/ML components for T1/T2 inline map generation are not detailed in this 510(k) summary. The T1/T2 inline map validation focuses on the output comparison, not the training of the model itself.

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

• As the document does not provide information on the training set, it also does not specify how the ground truth for the training set was established.

Ask a specific question about this device