Search Results

AI-Rad Companion Organs RT is a post-processing software intended to automatically contour DICOM CT imaging data using deep-learning-based algorithms.

Contours that are generated by AI-Rad Companion Organs RT may be used as input for clinical workflows including external beam radiation therapy treatment planning. AI-Rad Companion Organs RT must be used in conjunction with appropriate software such as Treatment Planning Systems and Interactive Contouring applications, to review, edit, and accept contours generated by AI-Rad Companion Organs RT.

The output of AI-Rad Companion Organs RT in the format of RTSTRUCT objects are intended to be used by trained medical professionals.

The software is not intended to automatically detect or contour lesions. Only DICOM images of adult patients are considered to be valid input.

AI-Rad Companion Organs RT is a post-processing software used to automatically contour DICOM CT imaging data using deep-learning-based algorithms. AI-Rad Companion Organs RT contouring workflow supports CT input data and produces RTSTRUCT outputs. The configuration of the organ database and organ templates defining the organs and structures to be contoured based on the input DICOM data is managed via a configuration interface. Contours that are generated by AI-Rad Companion Organs RT may be used as input for clinical workflows including external beam radiation therapy treatment planning.

The output of AI-Rad Companion Organs RT, in the form of RTSTRUCT objects, are intended to be used by trained medical professionals. The output of AI-Rad Companion Organs RT must be used in conjunction with appropriate software such as Treatment Planning Systems and Interactive Contouring applications, to review, edit, and accept contours generated by AI-Rad Companion Organs RT application.

At a high-level, AI-Rad Companion Organs RT includes the following functionality:

1. Automated contouring of Organs at Risk (OAR) workflow
   a. Input -DICOM CT
   b. Output DICOM RTSTRUCT
2. Organ Templates configuration (incl. Organ Database)
3. Web-based preview of contouring results to accept or reject the generated contours

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

The acceptance criteria for the AI-Rad Companion Organs RT are implicitly tied to demonstrating performance comparable to the predicate device, AccuContour, specifically in terms of contouring accuracy.

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

• Test Set Sample Size: 113 cases
• Data Provenance: Retrospective CT data previously acquired for RT treatment planning from multiple clinical sites across North America and Europe. The subcohort analysis also included CT data from multiple vendors (GE, Siemens, Phillips).

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

The document states: "Ground truth annotations were established following RTOG and clinical guidelines using manual annotation." It does not specify the number of experts or their specific qualifications (e.g., "radiologist with 10 years of experience"). However, the phrase "following RTOG and clinical guidelines using manual annotation" implies establishment by qualified medical professionals experienced in radiation therapy contouring.

4. Adjudication Method for the Test Set

The document does not explicitly describe an adjudication method (e.g., 2+1, 3+1). It only states that ground truth annotations were established via "manual annotation" following guidelines. This suggests a single expert or a consensus process without a detailed breakdown of the adjudication procedure in the provided text.

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

No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not reported as being done in this document. The study focuses on the standalone performance of the AI algorithm against a manual ground truth and a comparison to a predicate device's reported performance, not on how human readers' performance improves with or without AI assistance.

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

Yes, a standalone performance study of the algorithm was done. The reported DICE coefficient and Hausdorff Distance values directly assess the algorithm's output against the ground truth without human intervention in the contouring process itself. The "output of AI-Rad Companion Organs RT... are intended to be used by trained medical professionals" who "review, edit, and accept contours generated by AI-Rad Companion Organs RT," but the performance metrics provided are for the initial automated contouring.

7. The Type of Ground Truth Used

The ground truth used was expert consensus / manual annotation following RTOG and clinical guidelines.

8. The Sample Size for the Training Set

The document does not specify the sample size for the training set. It only discusses the validation set (113 cases).

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

The document does not specify how the ground truth for the training set was established. It only describes the ground truth establishment for the test/validation set: "Ground truth annotations were established following RTOG and clinical guidelines using manual annotation."

Ask a specific question about this device

This computed tomography system is intended to generate and process cross-sectional images of patients by computer reconstruction of x-ray transmission data.

The images delivered by the system can be used by a trained physician as an aid in diagnosis. The images delivered by the system can be used by trained staff as an aid in diagnosis, treatment preparation and radiation therapy planning.

This CT system can be used for low dose lung cancer screening in high risk populations *

• As defined by professional medical societies. Please refer to clinical literature, including the results of the National Lung Screening Trial (N Engl J Med 2011: 365:395-409) and subsequent literature, for further information.

Scan&GO:
The in-room scan application is a planning and information system designed to perform the necessary functions required for planning and controlling scans of supported Siemens CT scanners. It allows users to work in close proximity to the scanner.

The in-room scan application runs on standard information technology hardware and software, utilizing the standard information technology operating systems and user interface. Communication and data exchange are done using special protocols.

Siemens intends to update software version, SOMARIS/10 syngo CT VA30 (Update) for Siemens SOMATOM Computed Tomography (CT) Scanner Systems with mobile workflow and 3D Camera options.

This update includes support of additional hardware for the go. platform and includes reuse of optional postprocessing applications for Recon&GO for all scanners subject of this submission.

The SOMATOM CT Scanner Systems that support the same software platform update include:

• . SOMATOM go.Up
• . SOMATOM go.Now
• SOMATOM go.Top .
• SOMATOM go.All .
• . SOMATOM ao.Sim
• . SOMATOM go.Open Pro
• SOMATOM X.cite
• Scan&GO Mobile Medical Application (optional mobile workflow component) .

The subject device SOMATOM go. platform and SOMATOM X.cite with SOMARIS/10 syngo CT VA30 (update) are Computed Tomography X-ray Systems which feature one continuously rotating tubedetector system and function according to the fan beam principle. The SOMATOM go. platform and SOMATOM X.cite with software SOMARIS/10 syngo CT VA30 (update) produces CT images in DICOM format. These images can be used by trained staff for post-processing applications commercially distributed by Siemens Healthcare and other vendors. These images aid in diagnosis, treatment preparation and therapy planning support (including, but not limited to, Brachytherapy, Particle including Proton Therapy, External Beam Radiation Therapy, Surgery), The computer system delivered with the CT scanner is able to run optional post processing applications.

The Scan&GO mobile workflow is an optional planning and information software designed to perform the necessary functions required for planning and controlling of the SOMATOM X.cite and SOMATOM go. platform CT scanners. Scan&GO can be operated on a Siemens provided tablet or personal computer that meets certain minimum technical requirements. It allows users to work in close proximity to the scanner and the patient.

The software version for the SQMATOM go, platform and SOMATOM X.cite, syngo CT VA30 (update) (SOMARIS/10 syngo CT VA30 (update)), is a command-based program used for patient management, data management, X-ray scan control, image reconstruction, and image archive/evaluation. The software platform SOMARIS/10 syngo CT VA30 (update) is designed to support a software plugin interface to reuse a subset of stand-alone, cleared processing software applications.

The Siemens Medical Solutions USA, Inc. K200524 submission describes an update to the SOMATOM X.cite and SOMATOM Go Platform CT Scanners (software version SOMARIS/10 syngo CT VA30). The submission focuses on demonstrating substantial equivalence to previously cleared devices through non-clinical testing.

Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided text:

1. Table of Acceptance Criteria and the Reported Device Performance:

The document primarily discusses verification and validation testing, with acceptance criteria tied to the fulfillment of requirements and comparable performance to predicate devices. Specific quantitative acceptance criteria are not explicitly detailed in a separate table format within the provided text, but rather described in the "Testing Performed" column of Table S5-06.

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

• FAST Integrated Workflow (3D Camera): "Clinical data based software validation" was conducted. The specific number of cases or patients is not quantified in the provided text. Data provenance is not explicitly stated beyond "clinical data based software validation," implying it's likely from a clinical setting, but country of origin or retrospective/prospective nature is not specified.
• Contrast Media Protocol: The evaluation was based on factory protocols and comparison to approved drug labeling. This does not involve a "test set" in the traditional sense of patient data.
• Scan&GO Supported Hardware: "Bench test." The sample size for this is not specified. Data provenance is a bench test, presumably conducted by the manufacturer.
• Wireless Coexistence Testing: No specific sample size (number of wireless devices or test scenarios) is mentioned.
• Customer Use Testing:
  • Internal Clinical Use Test: "The CT scanner customer environment is simulated in Siemens Test Cabins. For such a test, customers with clinical expertise are typically invited to perform tests." The number of "customers with clinical expertise" or individual test cases is not quantified.
  • External Clinical Use Test: "The CT scanner is tested in the environment of the clinic/hospital. Typically we perform these tests with selected customer before rollout of the CT scanner." The number of "selected customer" sites or test cases is not quantified. Data provenance is clinical environments.

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

• For the FAST Integrated Workflow, the ground truth for "isocenter deviation" and "landmark boundaries" was established presumably by comparison to a reference or manual measurements, but the document does not specify the number or qualifications of experts involved in establishing this ground truth.
• For the Contrast Media Protocol, the ground truth is established by the "approved labeling of Ultravist® or Visipaque®." No external experts were involved in establishing new ground truth for this test.
• For Customer Use Testing, "customers with clinical expertise" were invited for internal tests, and "selected customer" (presumably clinical staff) performed external tests. The exact number and specific qualifications (e.g., "radiologist with 10 years of experience") are not provided.

4. Adjudication Method for the Test Set:

The document does not describe any explicit adjudication method (like 2+1 or 3+1) for any of the described tests. Performance for FAST Integrated Workflow appears to be based on direct measurement comparison. For customer use tests, it's implied that feedback from "customers with clinical expertise" determined meeting acceptance criteria, but no formal adjudication process is detailed.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done:

No, the document does not describe a Multi-Reader Multi-Case (MRMC) comparative effectiveness study to measure the effect size of how much human readers improve with AI vs. without AI assistance. The study focuses on the technical performance of the device's features.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done:

Yes, the testing described appears to be primarily focused on standalone algorithm/device performance for the modifications. For example:

• FAST Isocentering and FAST Range: The measurement of deviation is a direct assessment of the algorithm's accuracy in proposing an isocenter or landmark, independent of a human reader's interpretation improvement.
• Recon&GO features are noted as "post-processing algorithms" or "inline results methods" that appear to be integrated for improved workflow, implying their standalone function in generating these results.
• Software Specifications and Wireless Coexistence testing inherently evaluate the algorithm and system performance without explicit human-in-the-loop assessment as the primary outcome.

7. The Type of Ground Truth Used:

• FAST Integrated Workflow: The ground truth for isocenter deviation and landmark boundaries seems to be based on a reference standard or manual measurements to which the device's output is compared. This is an implicit form of expert consensus or highly accurate measurement.
• Contrast Media Protocol: The ground truth is the approved labeling of Ultravist® or Visipaque®, which serves as a regulatory and clinical standard.
• Scan&GO Supported Hardware: The ground truth is the information shown on tablets, serving as a reference for comparison of the in-room monitor's display.
• Software Specifications: Ground truth is defined by the software requirements/specifications themselves.
• Wireless Coexistence Testing: Ground truth is adherence to technical standards and successful communication parameters.
• Customer Use Testing: Ground truth seems to be based on expert opinion/feedback from "customers with clinical expertise" or "selected customer" in clinical environments, confirming the safety and effectiveness of the intended use.
• National Lung Screening Trial (NLST): This is referenced as supportive data for lung cancer screening indications. The ground truth for this external study (NLST) would have been clinical outcomes data (e.g., biopsy-confirmed cancer, mortality). However, it's important to note this is not the ground truth created for the current device's primary testing but rather cited clinical evidence supporting an indication for use.

8. The Sample Size for the Training Set:

The document does not specify any sample sizes for training sets. The submission describes updates to existing CT scanner systems and software, and the testing focuses on the verification and validation of these updates against predicate devices and defined requirements. This implies the core algorithms were likely developed and trained prior to this specific update.

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

Since no training set is mentioned, the method for establishing its ground truth is not described in this document.

Ask a specific question about this device

This computed tomography system is intended to generate and process cross-sectional images of patients by computer reconstruction of x-ray transmission data.

The images delivered by the system can be used by a trained physician as an aid in diagnosis. The images delivered by the system can be used by trained staff as an aid in diagnosis, treatment preparation and radiation therapy planning.

This CT system can be used for low dose lung cancer screening in high risk populations .*

• As defined by professional medical societies. Please refer to clinical literature, including the results of the National Lung Screening Trial (N Engl J Med 2011; 365:395-409) and subsequent literature, for further information.

Scan&GO:

The in-room scan application is a planning and information system designed to perform the necessary functions required for planning and controlling scans of supported SIEMENS CT scanners. It allows users to work in close proximity to the scanner.

The in-room scan application runs on standard information technology hardware and software, utilizing the standard information technology operating systems and user interface. Communication and data exchange are done using special protocols.

The subject device SOMATOM go.Platform with SOMARIS/10 syngo CT VA30 are Computed Tomography Xray Systems which feature one continuously rotating tube-detector system and function according to the fan beam principle. The SOMATOM go.Platform with Software SOMARIS/10 syngo CT VA30 produces CT images in DICOM format, which can be used by trained staff for post-processing applications commercially distributed by Siemens Healthcare and other vendors as an aid in diagnosis, treatment preparation and therapy planning support (including, but not limited to, Brachytherapy, Particle including Proton Therapy, External Beam Radiation Therapy, Surgery). The computer system delivered with the CT scanner is able to run optional post processing applications.

The Scan&GO mobile workflow is an optional planning and information software designed to perform the necessary functions required for planning and controlling of the SOMATOM go.Platform CT scanners. Scan&GO can be operated on a Siemens provided tablet or a commercially available tablet that meets certain minimum technical requirements. It allows users to work in the scanner and the patient.

The provided text describes acceptance criteria and testing for the Siemens SOMATOM go.Platform CT Scanners with software version SOMARIS/10 syngo CT VA30, and the Scan&GO mobile medical application.

Here's the breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

The document details various non-clinical tests conducted, with statements of the test results meeting the acceptance criteria. However, it does not present a single consolidated table of specific, quantifiable acceptance criteria alongside reported performance values for those criteria. Instead, it offers narrative summaries of the testing and its outcomes, indicating successful verification and validation.

Below is a table constructed from the provided text, outlining the features tested and the reported performance (which is generally stated as "met acceptance criteria" or "similar/improved performance").

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

• Check&GO Testing:

  • Sample size: 500 CT-series from 100 patients.
  • Data provenance: Not explicitly stated, but clinical datasets were used ("clinical datasets from 100 patients"). It's specified as a "bench test," which implies it was likely retrospective from an existing data archive. Country of origin is not mentioned.

• Other Non-Clinical Testing (Phantom, Integration, Functional): The document frequently refers to "phantom images," "test levels," "development activities," and "bench tests." No specific sample sizes for these tests (e.g., number of phantom scans) or data provenance are provided beyond the general descriptions of the tests themselves, which are stated as having been conducted "during product development."

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

• Check&GO Testing:

  • Ground Truth Establishment: The datasets were "manually annotated with a detailed GT contrast-state (None, Low, InhomogeneousLow, Standard, InhomogeneousHigh, High)."
  • Number & Qualifications of Experts: Not specified.

• Other Tests: For other tests, such as those involving image quality or physical measurements (e.g., Flex 4D Spiral, DirectDensity), the ground truth is typically derived from physical measurements, reference standards (e.g., known phantom properties), or established technical specifications, rather than expert consensus on clinical interpretation. The document does not mention the use of experts to establish ground truth for these tests. The indication for the new "Kidney Stones" feature notes: "Only a well-trained radiologist can make the final diagnosis under consideration of all available information," suggesting the involvement of radiologists in the clinical context, but not for ground truth establishment specifically for the device's technical validation.

4. Adjudication method for the test set

• The document does not explicitly describe an adjudication method (like 2+1, 3+1, etc.) for any of the test sets. For the Check&GO test, ground truth was "manually annotated," implying a single process for ground truth establishment rather than a consensus/adjudication method among multiple experts.

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 involving human readers and AI assistance is reported for this device in the provided text. The device itself is a CT scanner system and its associated software, not explicitly an AI-assisted diagnostic tool for interpretation in collaboration with human readers. The Check&GO feature is described as "aiding the user," but no study on human performance improvement is included.

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

• The non-clinical performance testing, particularly phantom studies and specific feature evaluations like "kV and Filter independent CaScore," "Recon&GO - Spectral Recon," "TwinSpiral Dual Energy," "Flex 4D Spiral," "DirectDensity," "HD FoV," and "InjectorCoupling," can be considered standalone algorithm/device performance evaluations. These tests assess the technical output and accuracy of the device and its software features independent of human interpretation or interaction during the measurement process. The Check&GO feature's "Bench Test" also evaluates the algorithm's performance against annotated ground truth.

7. The type of ground truth used

• Check&GO: Expert annotation of "detailed GT contrast-state" (None, Low, InhomogeneousLow, Standard, InhomogeneousHigh, High) for 500 CT series.
• Other Feature Tests (e.g., CaScore, Spectral Recon, Flex 4D Spiral, DirectDensity, HD FoV, TwinSpiral DE): Primarily derived from physical phantom measurements, comparison to established technical specifications, or reference images/algorithms (e.g., comparing to initial release versions or conventional 120kV images).
• National Lung Screening Trial (NLST): Outcomes data from a large clinical trial (N Engl J Med 2011; 365:395-409) is cited to support the "low dose lung cancer screening" indication for use, not for direct ground truth establishment during this device's specific validation, but rather as supportive clinical literature for the screening concept itself.

8. The sample size for the training set

• The document does not specify any training set sizes. The studies described are primarily for verification and validation, not for training machine learning models. The Check&GO feature describes "500 CT-series from 100 patients were used for the testing of the algorithm," but this is explicitly called "testing," not training.

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

• Since no training set details (size or establishment method) are provided, this information is not available in the document.

Ask a specific question about this device