Search Results

Trained medical professionals use Contour ProtégéAl as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0. In addition, Contour ProtégéAl supports the following indications:

· Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.

· Segmenting anatomical structures across a variety of CT anatomical locations.

· And segmenting the prostate, the seminal vesicles, and the urethra within T2-weighted MR images.

Appropriate image visualization software must be used to review and, if necessary, edit results automatically generated by Contour ProtégéAI.

Contour ProtégéAl+ is an accessory to MIM software that automatically creates contours on medical images through the use of machine-learning algorithms. It is designed for use in the processing of medical images and operates on Windows, Mac, and Linux computer systems. Contour ProtégéAl+ is deployed on a remote server using the MIMcloud service for data management and transfer; or locally on the workstation or server running MIM software.

Here's a breakdown of Contour ProtégéAI+'s acceptance criteria and study information, based on the provided text:

Acceptance Criteria and Device Performance

The acceptance criteria for each structure's inclusion in the final models were a combination of statistical tests and user evaluation:

Note: Cells highlighted in orange in the original document indicate non-demonstrated equivalence (not reproducible in markdown), and cells marked with '**' indicate that equivalence was not demonstrated because the minimum sample size was not met for that contour.

Study Details

1. Sample size used for the test set and the data provenance:

   • Test Set Sample Size: The Contour ProtégéAI+ subject device was evaluated on a pool of 770 images.
   • Data Provenance: The images were gathered from 32 institutions. The verification data used for testing is from a set of institutions that are totally disjoint from the datasets used to train each model. Patient demographics for the testing data are: 53.4% female, 31.3% male, 15.3% unknown; 0.3% ages 0-20, 4.7% ages 20-40, 20.9% ages 40-60, 50.0% ages 60+, 24.1% unknown; varying scanner manufacturers (GE, Siemens, Phillips, Toshiba, unknown). The data is retrospective, originating from clinical treatment plans according to the training set description.

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

   • The document implies that the ground truth for the test set was validated against "original ground-truth contours" when measuring Dice and MDA against MIM Maestro. However, the expert qualifications are explicitly stated for the training set ground truth, which often implies a similar standard for the test set.
   • Ground truth (for training/re-segmentation) was established by:
     • Consultants (physicians and dosimetrists) specifically for this purpose, outside of clinical practice.
     • Initial segmentations were reviewed and corrected by radiation oncologists.
     • Final review and correction by qualified staff at MIM Software (MD or licensed dosimetrists).
     • All segmenters and reviewers were instructed to ensure the highest quality training data according to relevant published contouring guidelines.

3. Adjudication method for the test set:

   • The document doesn't explicitly describe a specific adjudication method like "2+1" or "3+1" for the test set ground truth. However, it does state that "Detailed instructions derived from relevant published contouring guidelines were prepared for the dosimetrists. The initial segmentations were then reviewed and corrected by radiation oncologists against the same standards and guidelines. Qualified staff at MIM Software (MD or licensed dosimetrists) then performed a final review and correction." This process implies a multi-expert review and correction process to establish the ground truth used for both training and evaluation, ensuring a high standard of accuracy.

4. 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:

   • A direct MRMC comparative effectiveness study measuring human readers' improvement with AI versus without AI assistance (i.e., human-in-the-loop performance) is not explicitly described in terms of effect size.
   • Instead, the study evaluates the standalone performance of the AI device (Contour ProtégéAI+) against a reference device (MIM Maestro atlas segmentation) and user evaluation of time savings.
   • The "Average user evaluation of 2 or higher" on a three-point scale (1=negligible, 2=moderate, 3=significant time savings) provides qualitative evidence of perceived improvement in workflow rather than a quantitative measure of diagnostic accuracy improvement due to AI assistance. "Preliminary user evaluation conducted as part of testing demonstrated that Contour ProtégéAI+ yields comparable time-saving functionality when creating contours as other commercially available automatic segmentation products."

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

   • Yes, a standalone performance evaluation was conducted. The primary comparisons for Dice score, MDA, and cumulative APL are between the Contour ProtégéAI+ algorithm's output and the ground truth, benchmarked against the predicate device's (MIM Maestro atlas segmentation) standalone performance. The results in Table 2 and Table 3 directly show the algorithm's performance.

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

   • Expert Consensus Contour (and review): The ground truth was established by expert re-segmentation of images (by consultants, physicians, and dosimetrists) specifically for this purpose, reviewed and corrected by radiation oncologists, and then subjected to a final review and correction by qualified MIM Software staff (MD or licensed dosimetrists). This indicates a robust expert consensus process based on established clinical guidelines.

7. The sample size for the training set:

   • The document states that the CT images for the "training set were obtained from clinical treatment plans for patients prescribed external beam or molecular radiotherapy". However, it does not provide a specific numerical sample size for the training set, only for the test set (770 images). It only mentions being "re-segmented by consultants... specifically for this purpose".

8. How the ground truth for the training set was established:

   • The ground truth for the training set was established through a multi-step expert process:
     • CT images from clinical treatment plans were re-segmented by consultants (physicians and dosimetrists), explicitly for the purpose of creating training data, outside of clinical practice.
     • Detailed instructions from relevant published contouring guidelines were provided to the dosimetrists.
     • Initial segmentations were reviewed and corrected by radiation oncologists against the same standards and guidelines.
     • A final review and correction was performed by qualified staff at MIM Software (MD or licensed dosimetrists).
     • All experts were instructed to spend additional time to ensure the highest quality training data, contouring all specified OAR structures on all images according to referenced standards.

Ask a specific question about this device

MIM software is used by trained medical professionals as a tool to aid in evaluation and information management of digital medical image modalities include but are not limited to, CT, MR, CR, DX, MG, US, SPECT, PET and XA as supported by ACR/NEMA DICOM 3.0. MIM assists in the following indications:

• · Receive, transmit, store, retrieve, display, print, and process medical images and DICOM objects.
• · Create, display and print reports from medical images.

· Registration, fusion display, and review of medical images for diagnosis, treatment evaluation, and treatment planning.

· Evaluation of cardiac left ventricular function and perfusion, including left ventricular end-diastolic volume, end-systolic volume, and ejection fraction.

· Localization and definition of objects such as tumors and normal tissues in medical images.

· Creation, transformation, and modification of contours for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.

· Quantitative and statistical analysis of PET/SPECT brain scans by comparing to other registered PET/SPECT brain scans.

· Planning and evaluation of permanent implant brachytherapy procedures (not including radioactive microspheres).

· Calculating absorbed radiation dose as a result of administering a radionuclide.

• Assist with the planning and evaluation of ablation procedures by providing visualization and analysis, including energy zone visualization through the placement of virtual ablation devices validated for inclusion in MM-Ablation. The software is not intended to predict specific ablation zone volumes or predict ablation success.

When using the device clinically, within the United States, the user should only use FDA approved radiopharmaceuticals. If using with unapproved ones, this device should only be used for research purposes.

Lossy compressed mammographic images and digitized film screen images must not be reviewed for primary image interpretations. Images that are printed to film must be printed using a FDAapproved printer for the diagnosis of digital mammography images. Mammographic images must be viewed on a display system that has been cleared by the FDA for the diagnosis of digital mammography images. The software is not to be used for mammography CAD.

MIM - Symphony HDR Fusion extends the existing features and capabilities of MIM -Monte Carlo Dosimetry (K232862) by offering enhanced capabilities to better support the High Dose Rate (HDR) brachytherapy workflow. It is designed for use in medical imaging and operates on Windows, Mac, and Linux computer systems. The intended use and indications for use in MIM - Symphony HDR Fusion are unchanged from the predicate device MIM – Monte Carlo Dosimetry.

MIM – Symphony HDR Fusion is a standalone software application within the MIM software suite that uses the existing functionality of the predicate device, applied now in the context of a High Dose Rate (HDR) brachytherapy clinical workflow.

MIM – Symphony HDR Fusion leverages the foundational functionalities that were introduced in the predicate device to support Low Dose Rate (LDR) brachytherapy clinical workflows. These features are extended with necessary enhancements and optimizations to optimally support the HDR workflow. Specifically, the subject device MIM - Symphony HDR Fusion provides the following core processes:

• Reslicing and Predictive Fusion presents data to inform the user's placement of medical devices (in this case, brachytherapy applicators). MIM receives and displays 2D images from a Trans-rectal Ultrasound (TRUS) probe and overlays contours from the registered pre-op image volume. The user is able to modify the position of the TRUS probe in the patient in order to match the visible pre-op contours. The user may also manually adjust the registration using software tools.
• . Ultrasound Capture: MIM receives an image feed from an US machine and position information from a stepper that holds the TRUS probe. The 2D TRUS images are processed into 3D image volumes-enabling their registration, fusion display, and storage as DICOM objects.
• Catheter Digitization provides tools for the user to localize and define HDR . brachytherapy applicators (catheters) in medical images.
• Registration Chaining allows the user to transfer information (contours) from the pre-op image (typically MR) through to the final planning image (US or CT). This is achieved using existing rigid registration tools from the predicate device to sequentially register each new image to its immediate predecessor in the clinical workflow.
• Export Data: The end of the MIM Symphony HDR Fusion workflow is to export the final planning image and user-defined structures-including organs and brachytherapy applicator models—into DICOM files for use in third-party radiation therapy treatment planning systems. Structured reports may also be created.

Here's a breakdown of the acceptance criteria and study information for MIM - Symphony HDR Fusion, 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: The document does not specify an exact numerical sample size for the test set. It mentions "each of the five core features" underwent testing.
• Data Provenance: The document does not explicitly state the country of origin of the data or whether it was retrospective or prospective.

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

• Number of Experts: Unspecified, but referred to as "trained medical professionals with extensive experience in HDR brachytherapy."
• Qualifications of Experts: "trained medical professionals with extensive experience in HDR brachytherapy."

4. Adjudication Method for the Test Set

• The document does not describe a specific adjudication method (e.g., 2+1, 3+1). It states that "external validation by trained medical professionals" was conducted.

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

• No, a multi-reader multi-case (MRMC) comparative effectiveness study explicitly measuring the effect size of human readers improving with AI vs. without AI assistance was not reported. The validation involved external medical professionals, but it was to validate the software's functionality, not a comparative study on reader performance.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

• Yes, the performance of the device's features was evaluated. The validation included "internal verification by MIM's own qualified testers" and "external validation by trained medical professionals." While the "external validation" implies human interaction for the final assessment of the software's utility in a workflow, the "internal verification" likely involved standalone testing of the algorithms comprising each feature. The description focuses on the software as a "tool to aid in evaluation," implying a human-in-the-loop context for clinical use, but the individual feature testing suggests standalone performance evaluation.

7. The Type of Ground Truth Used

• The document implies an expert consensus/determination based on the involvement of "trained medical professionals with extensive experience in HDR brachytherapy" for external validation. For internal verification, "MIM's own qualified testers" would have established the ground truth based on predefined specifications and expected outputs for each feature.

8. The Sample Size for the Training Set

• The document does not specify a sample size for a training set. This suggests that MIM - Symphony HDR Fusion is an extension of existing features from a predicate device (MIM - Monte Carlo Dosimetry) and leverages "foundational functionalities." The description focuses on verification and validation of the new and extended functionalities rather than the development of entirely new machine learning algorithms requiring a distinct training set.

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

• As a training set is not explicitly mentioned, the method for establishing its ground truth is also not described.

Ask a specific question about this device

MIM software is used by trained medical professionals as a tool to aid in evaluation and information management of digital medical images. The medical image modalities include, but are not limited to, CT, MR, CR, DX, MG, US, SPECT, PET and XA as supported by ACR/NEMA DICOM 3.0. MIM assists in the following indications:

· Receive, transmit, store, retrieve, display, print, and process medical images and DICOM objects.

· Create, display, and print reports from medical images.

· Registration, fusion display, and review of medical images for diagnosis, treatment evaluation, and treatment planning.

· Evaluation of cardiac left ventricular function and perfusion, including left ventricular end-diastolic volume, end-systolic volume, and ejection fraction.

· Localization and definition of objects such as tumors and normal tissues in medical images.

· Creation, transformation, and modification of contours for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.

· Quantitative and statistical analysis of PET/SPECT brain scans by comparing to other registered PET/SPECT brain scans.

· Planning and evaluation of permanent implant brachytherapy procedures (not including radioactive microspheres).

· Calculating absorbed radiation dose as a result of administering a radionuclide.

When using the device clinically, within the United States, the user should only use FDA approved radiopharmaceuticals. If used with unapproved ones, this device should only be used for research purposes.

Lossy compressed mammoaraphic images and digitized film screen images must not be reviewed for primary image interpretations. Images that are printed to film must be printed using an FDA-approved printer for the diagnosis of digital mammography images. Mammographic images must be viewed on a display system that has been cleared by the FDA for the diagnosis of digital mammography images. The software is not to be used for mammography CAD.

MIM - Centiloid Scaling extends the features of MIM - Additional Tracers (K223800). It is designed for use in medical imaging and operates on Windows, Mac, and Linux computer systems. The intended use and indications for use in MIM - Centiloid Scaling are unchanged from the predicate device, MIM - Additional Tracers (K223800).

MIM - Centiloid Scaling is a standalone software application that extends the functionality of the predicate device by providing:

• · Conversion of SUVr calculations to a standardized Centiloid scale for PET-based amyloid burden measurement with Florbetapir (Amvvid®), Florbetaben (Neuraceq®), and Flutemetamol (Vizamyl™)

The MIM - Centiloid Scaling device is intended to convert SUVr (Standardized Uptake Value ratio) calculations to a standardized Centiloid scale for PET-based amyloid burden measurement using specific radiopharmaceuticals (Florbetapir (Amyvid®), Florbetaben (Neuraceq®), and Flutemetamol (Vizamyl™)).

Here's an analysis of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are implicitly derived from the validation methods and desired outcomes of the Centiloid Project and comparisons to expert visual reads.

Ask a specific question about this device

MIM software is used by trained medical professionals as a tool to aid in evaluation and information management of digital medical images. The medical image modalities include, but are not limited to, CT, MR, CR, DX, MG, US, SPECT, PET and XA as supported by ACR/NEMA DICOM 3.0. MIM assists in the following indications:

• Receive, transmit, store, retrieve, display, print, and process medical images and DICOM objects.
• Create, display, and print reports from medical images.
• Registration, fusion display, and review of medical images for diagnosis, treatment evaluation, and treatment planning.
• Evaluation of cardiac left ventricular function and perfusion, including left ventricular end-diastolic volume, end-systolic volume, and ejection fraction.
• Localization and definition of objects such as tumors and normal tissues in medical images.
• Creation, transformation, and modification of contours for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.
• Quantitative and statistical analysis of PET/SPECT brain scans by comparing to other registered PET/SPECT brain scans.
• Planning and evaluation of permanent implant brachytherapy procedures (not including radioactive microspheres).
• Calculating absorbed radiation dose as a result of administering a radionuclide.
• Assist with the planning and evaluation of ablation procedures by providing visualization and analysis, including energy zone visualization through the placement of virtual ablation devices validated for inclusion in MIM-Ablation. The software is not intended to predict specific ablation zone volumes or predict ablation success.

When using the device clinically, within the United States, the user should only use FDA approved radiopharmaceuticals. If used with unapproved ones, this device should only be used for research purposes.

Lossy compressed mammographic images and digitized film screen images must not be reviewed for primary image interpretations. Images that are printed to film must be printed using an FDA-approved printer for the diagnosis of digital mammography images. Mammographic images must be viewed on a display system that has been cleared by the FDA for the diagnosis of digital mammography images. The software is not to be used for mammography CAD.

MIM - Monte Carlo Dosimetry (K232862) extends the features of MIM - Ablation (K220256). It is designed for use in medical imaging and operates on Windows, Mac, and Linux computer systems. The intended use and indications for use in MIM - Monte Carlo Dosimetry are unchanged from the predicate device, MIM - Ablation (K220256).

MIM - Monte Carlo Dosimetry (K232862) is a standalone software application that extends the functionality of the predicate device by providing:

• Dose calculation of radionuclides performed using a Monte Carlo method

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

Device: MIM - Monte Carlo Dosimetry (K232862)

Acceptance Criteria and Reported Device Performance:

Study Details:

1. Sample size used for the test set and the data provenance:

   • The test set used "an existing CT scan of the patient that was of height (1.7m) and weight (77kg) similar to the default Adult Male model in OLINDA (1.7m, 70kg)."
   • The data provenance is not explicitly stated as retrospective or prospective, nor is the country of origin. However, the use of "an existing CT scan" suggests it was retrospective. The patient data was used for all three main comparison studies.

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

   • The document describes comparisons against established dosimetry methods (OLINDA/EXM, MIM-Ablation's VSV, and GATE), which serve as the reference for "ground truth" in this context. It does not mention human experts establishing ground truth for the test set, as the evaluation is based on quantitative comparison of calculated doses.

3. Adjudication method for the test set:

   • Not applicable as the ground truth wasn't based on expert adjudication of diagnostic interpretations, but rather on computational agreement with established dosimetry methods.

4. 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 was done. This device is a dose calculation software, not an AI diagnostic assistant for human readers.

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

   • Yes, the performance evaluation described (comparison to OLINDA/EXM, VSV, and GATE, along with particle history characterization) is a standalone algorithm-only performance assessment. The "MIM - Monte Carlo Dosimetry" is described as a "standalone software application."

6. The type of ground truth used:

   • Computational Ground Truth: The ground truth was established by comparing the device's calculations to:
     • Model-based dosimetry from the predicate device OLINDA/EXM (K033960).
     • Voxel S-value (VSV) dose calculation from the predicate device MIM – Ablation (K220256).
     • A "well-established Monte Carlo dose calculation algorithm, GATE."

7. The sample size for the training set:

   • The document does not mention a training set, as this is a physics-based dose calculation software, not a machine learning or AI model that requires a labeled training set in the typical sense. Its development would rely on physical models and algorithms rather than statistical learning from data.

8. How the ground truth for the training set was established:

   • Not applicable, as there is no specific "training set" mentioned or implied for this type of software.

Ask a specific question about this device

MIM software is used by trained medical professionals as a tool to aid in evaluation and information management of digital medical images. The medical image modalities include, but are not limited to, CT, MR, CR, DX, MG, US, SPECT, PET and XA as supported by ACR/NEMA DICOM 3.0. MIM assists in the following indications:

• Receive, transmit, store, retrieve, display, print, and process medical images and DICOM objects.
• Create, display, and print reports from medical images.
• Registration, fusion display, and review of medical images for diagnosis, treatment evaluation, and treatment planning.
• Evaluation of cardiac left ventricular function and perfusion, including left ventricular end-diastolic volume, end-systolic volume, and ejection fraction.
• Localization and definition of objects such as tumors and normal tissues in medical images.
• Creation, transformation, and modification of contours for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.
• Quantitative and statistical analysis of PET/SPECT brain scans by comparing to other registered PET/SPECT brain scans.
• Planning and evaluation of permanent implant brachytherapy procedures (not including radioactive microspheres).
• Calculating absorbed radiation dose as a result of administering a radionuclide.
• Assist with the planning and evaluation of ablation procedures by providing visualization and analysis, including energy zone visualization through the placement of virtual ablation devices validated for inclusion in MIM-Ablation. The software is not intended to predict specific ablation zone volumes or predict ablation success.

When using the device clinically, within the United States, the user should only use FDA approved radiopharmaceuticals. If used with unapproved ones, this device should only be used for research purposes.

Lossy compressed mammographic images and digitized film screen images must not be reviewed for primary image interpretations. Images that are printed to film must be printed using an FDA-approved printer for the diagnosis of digital mammography images. Mammographic images must be viewed on a display system that has been cleared by the FDA for the diagnosis of digital mammography images. The software is not to be used for mammography CAD.

MIM – Dose Analysis extends features of MIM – Ablation (K220256). It is designed for use in medical imaging and operates on both Windows and Mac computer systems. MIM - Dose Analysis is a standalone software application that provides:

• Calculation of Biologically Effective Dose (BED) and Equieffective Dose in specified fractions (e.g., EQD2)
• Evaluation of radiation dose on different image sets through the use of rigid and deformable registration objects
• Accumulation of doses

The provided text describes the regulatory clearance of "MIM - Dose Analysis" software. While it outlines the device's functionality and states that testing was performed, it does not provide explicit acceptance criteria in a quantitative table or the detailed study results needed to prove those criteria were met. Instead, it offers broad statements about the testing methodology and validation.

However, based on the provided text, we can infer the acceptance criteria and reconstruct what the study aimed to prove, along with the limited information available regarding the study itself.

Inferred Acceptance Criteria and Reported Device Performance

Given the nature of the device (calculating radiation dose and accumulating doses) and the statements in the "Testing and Performance Data" section, the acceptance criteria would revolve around the accuracy and consistency of these calculations.

Table 1: Inferred Acceptance Criteria and Reported Device Performance

Study Details from the Provided Text:

Unfortunately, the provided document is a 510(k) summary for regulatory clearance and does not contain the extensive detail of a full clinical study report. Therefore, many of the requested details are not explicitly stated.

1. Sample sized used for the test set and the data provenance:
* Sample Size: Not specified. The text mentions "Several different applications of the standard Linear-Quadratic (LQ) BED Model were tested" and "radiation doses delivered at separate timepoints for each patient dataset." This implies a case-based testing approach, but the number of cases is not provided.
* Data Provenance: Not specified regarding country of origin or whether it was retrospective or prospective data. The use of "patient datasets" could imply retrospective patient data, but this is not explicitly stated.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
* Not specified. The ground truth for BED, EQD, and TCP calculations was established by "manual calculations using the same equations referenced in literature." For dose accumulation, it was compared against "manual dose accumulation calculations" and involved "physician-contoured anatomical structures." This suggests expert involvement in contouring, but the number or qualifications of these physicians are not detailed.

3. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
* Not applicable/Not specified. The testing described primarily compares algorithmic output against manual calculations or established formulas, not against a human reader's interpretation that would require adjudication.

4. 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 or comparative effectiveness study involving human readers is mentioned or described. The study focuses on the software's calculation accuracy in a standalone manner.

5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
* Yes, implicitly. The testing described ("BED, EQD, and TCP calculations implemented in MIM were compared against manual calculations" and "Accumulated dose was calculated in MIM... and these dose results were compared against manual dose accumulation calculations") is a standalone performance assessment of the algorithm against a defined ground truth (manual calculations/literature equations).

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
* Calculated Ground Truth / Standardized Equations: For BED, EQD, and TCP, the ground truth was derived from "manual calculations using the same equations referenced in literature."
* Manual Calculation/Physician Contours: For dose accumulation, the ground truth was derived from "manual dose accumulation calculations" applied to "physician-contoured anatomical structures of interest."

7. The sample size for the training set:
* Not specified. This document only describes verification and validation testing, not the training of a potential machine learning model. If this software does not contain machine learning components, then "training set" is not applicable. The text does not indicate the presence of AI/ML components beyond standard imaging processing and calculation functionalities.

8. How the ground truth for the training set was established:
* Not applicable, as a training set or its ground truth establishment are not mentioned for this device in the provided text.

Ask a specific question about this device

Trained medical professionals use Contour ProtégéAI as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0. In addition, Contour ProtégéAI supports the following indications:

· Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.

· Segmenting anatomical structures across a variety of CT anatomic locations.

· And segmenting the prostate, the seminal vesicles, and the urethra within T2-weighted MR images.

Appropriate image visualization software must be used to review and, if necessary, edit results automatically generated by Contour ProtégéAI.

Contour ProtégéAI is an accessory to MIM software that automatically creates contours on medical images through the use of machine-learning algorithms. It is designed for use in the processing of medical images and operates on Windows, Mac, and Linux computer systems. Contour ProtégéAl is deployed on a remote server using the MIMcloud service for data management and transfer; or locally on the workstation or server running MIM software.

Here's a detailed breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided FDA 510(k) summary:

Acceptance Criteria and Reported Device Performance for Contour ProtégéAI

1. Table of Acceptance Criteria and Reported Device Performance

2. Statistical non-inferiority of the MDA score compared with the reference predicate (MIM Maestro atlas segmentation).
3. Average user evaluation score of 2 or higher (on a 3-point scale).
   A structure is deemed acceptable if it passes two or more of these three tests. | Dice Score: For all reported structures in the Head and Neck, Thorax, and Whole Body - Physiological Uptake Organs CT models, Contour ProtégéAI generally showed higher mean Dice scores (indicating better overlap with ground truth) and often superior lower 95th percentile confidence bounds compared to MIM Atlas. Equivalence (defined as lower 95th percentile confidence bound of ProtégéAI Dice > 0.1 Dice lower than MIM Atlas mean) was demonstrated for most structures, often with direct improvement.

MDA Score: For most reported structures, Contour ProtégéAI showed lower mean MDA scores (indicating better boundary accuracy/distance to ground truth) and often superior upper 95th percentile confidence bounds compared to MIM Atlas. Equivalence was demonstrated for most structures, again often with direct improvement.

External Evaluation Score: All reported structures achieved an average user evaluation score of 2 or higher (ranging from 2.0 to 3.0), indicating moderate to significant time savings.

Overall: The summary states: "Contour ProtégéAl results were equivalent or had better performance than the MIM Maestro atlas segmentation reference device." And "only structures that pass two or more of the following three tests could be included in the final models". This indicates successful performance against the criteria for all included structures. |
| Model-as-a-Whole Performance | Statistically non-inferior cumulative Added Path Length (APL) compared to the reference predicate. | Cumulative APL (mm):

• Head and Neck CT: MIM Atlas: 38.69 ± 33.36; Contour ProtégéAI: 28.61 ± 29.59. Equivalence demonstrated.
• Thorax CT: MIM Atlas: 89.24 ± 82.73; Contour ProtégéAI: 65.44 ± 68.85. Equivalence demonstrated.
• Whole Body - Physiological Uptake Organs CT: MIM Atlas: 138.06 ± 142.42; Contour ProtégéAI: 98.20 ± 127.11. Equivalence demonstrated.

This indicates that Contour ProtégéAI performs with lower or equivalent APL, suggesting less editing time for the entire model. |
| Localization Accuracy (Informational) | No passing criterion, but results included for user understanding. | Percentage of images successfully localized by Contour ProtégéAI is provided for each structure and model. Most structures show 100% localization accuracy within their relevant FOV for Head and Neck and Thorax models. Some structures (e.g., Cochlea_L/R, OpticChiasm, Pancreas) show slightly lower percentages, indicating instances where the structure was not localized. For Whole Body CT, many structures also show 100%, with a few exceptions (e.g., Bladder: 95%, LN_Iliac: 64%). |

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

• Test Set Sample Size: 754 independent images.
• Data Provenance: Gathered from 27 institutions. The document does not explicitly state the countries of origin for the test set, but for the training set, it mentions "across multiple continents" and lists "USA" and "Hong Kong" and "Australia." It is reasonable to infer the test set would also be from diverse institutions/countries. The data is retrospective as it was gathered from existing clinical treatment plans.

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

The ground truth for the test set was established by a multi-stage process involving:

• Initial Segmentation: Consultants (physicians and dosimetrists).
• Review and Correction: A radiation oncologist.
• Final Review and Correction: Qualified staff at MIM Software (M.D. or licensed dosimetrists).

While the exact number of experts is not specified, it involved multiple individuals with specialized qualifications (physicians, dosimetrists, radiation oncologists, M.D.s, licensed dosimetrists).

4. Adjudication Method for the Test Set

The ground truth generation involved a multi-stage review and correction process:

1. Initial segmentations by consultants (physicians and dosimetrists).
2. Review and correction by a radiation oncologist against established standards and guidelines.
3. Final review and correction by qualified staff at MIM Software (M.D. or licensed dosimetrists).

This indicates a sequential refinement process, potentially similar to a "cascading consensus" or "expert review and correction" rather than a specific numeric adjudication method like 2+1 or 3+1 for resolving disagreements among multiple initial segmenters. The explicit mentioning of "correction" at multiple stages suggests an iterative process where initial segmentations were refined based on expert review.

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

No, a traditional MRMC comparative effectiveness study was not explicitly stated in the provided text in the context of comparing human readers with and without AI assistance to measure an effect size on human performance.

Instead, the study primarily focused on the standalone performance of the AI model (Contour ProtégéAI) compared to an existing atlas-based segmentation system (MIM Maestro) using quantitative metrics (Dice, MDA, APL) and a user evaluation for "time savings functionality." The user evaluation (average score of 2 or higher on a three-point scale for time savings) provides an indirect measure of the AI's utility, but not a direct MRMC study on human reader improvement with AI.

6. If a Standalone Study Was Done

Yes, a standalone study was done.

• Contour ProtégéAI (the algorithm under review) was evaluated in comparison to a reference predicate device, MIM Maestro (K071964), which uses an atlas-based segmentation approach.
• The comparison involved quantitative metrics like Dice score, MDA, and cumulative APL, as well as a qualitative user evaluation. The goal was to show that Contour ProtégéAI was equivalent or superior in performance to the reference predicate in a standalone capacity.

7. The Type of Ground Truth Used

The ground truth used for the test set was expert consensus / expert-derived segmentation.

• It was derived from clinical treatment plans, but the original segmentations were not used.
• The images were re-segmented by consultants (physicians and dosimetrists) specifically for this purpose, following detailed clinical contouring guidelines.
• These initial segmentations were then reviewed and corrected by a radiation oncologist.
• A final review and correction was performed by qualified staff at MIM Software (M.D. or licensed dosimetrists).
• All segmenters were instructed to ensure the "highest quality training data" and contour according to referenced standards.

8. The Sample Size for the Training Set

• CT Models: A total of 550 CT images from 41 clinical sites.
• The document implies that these 550 images are specifically for the training of the final 4.1.0 neural network models for CT. It does not explicitly state the training set size for MR models if they were separate.

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

The ground truth for the training set was established through a rigorous, multi-stage expert-driven process, identical to the description for the test set ground truth:

1. Initial Segmentation: Performed by consultants (physicians and dosimetrists) following detailed instructions derived from published clinical contouring guidelines.
2. Review and Correction: By a radiation oncologist against the same standards and guidelines.
3. Final Review and Correction: By qualified staff at MIM Software (M.D. or licensed dosimetrists).
   • The goal was "to ensure the highest quality training data."
   • Segmenters were asked to contour all specified OAR structures on all images according to referenced standards, regardless of proximity to the treatment field.

Ask a specific question about this device

Trained medical professionals use Contour ProtégéAI as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0. In addition, Contour ProtégéAI supports the following indications:

· Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transfering contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.

· Segmenting anatomical structures across a variety of CT anatomic locations.

· And segmenting the prostate, the seminal vesicles, and the urethra within T2-weighted MR images.

Appropriate image visualization software must be used to review and, if necessary, edit results automatically generated by Contour ProtégéAI.

Contour ProtégéAI is an accessory to MIM software that automatically creates contours on medical images through the use of machine-learning algorithms. It is designed for use in the processing of medical images and operates on Windows, Mac, and Linux computer systems. Contour ProtégéAl is deployed on a remote server using the MIMcloud service for data management and transfer; or locally on the workstation or server running MIM software.

Here's a breakdown of the acceptance criteria and study details for Contour ProtégéAI, based on the provided document:

Acceptance Criteria and Device Performance Study for Contour ProtégéAI

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for Contour ProtégéAI were based on a non-inferiority study comparing its segmentation performance (measured by Dice coefficient) to a predicate device, MIM Maestro (K071964), specifically using atlases built from the same training data. The key acceptance criterion was:

Equivalence is defined such that the lower 95th percentile confidence bound of the Contour ProtégéAI segmentation is greater than 0.1 Dice lower than the mean MIM atlas segmentation reference device performance.

This translates to being either equivalent to or having better performance than the MIM Maestro atlas segmentation reference device. The acceptance was demonstrated at a p=0.05 significance level.

The table below summarizes the reported mean ± standard deviation Dice coefficients for both the MIM Atlas (predicate) and Contour ProtégéAI, along with the lower 95th percentile confidence bound for Contour ProtégéAI, for various anatomical structures across different CT models (4.0.0 CT Model). The asterisk (*) next to Contour ProtégéAI performance indicates that equivalence was demonstrated at p=0.05.

Note: The document presents a single large table for all structures and models. For clarity, a few representative examples from each CT Model are extracted below to illustrate the reported performance against the acceptance criteria. The full table from the document should be consulted for comprehensive results.

2. Sample Size and Data Provenance for the Test Set

• Sample Size for Test Set: 819 independent images.
• Data Provenance: The images were gathered from 10 institutions. The document explicitly states that the test set institutions are "totally disjoint from the training datasets used to train each model." The countries of origin for the test set are not explicitly detailed, but since the training data included multiple countries (USA, Hong Kong, Australia), it's implied the test set could also be diverse. The data was retrospective clinical data, re-segmented for this specific purpose.

3. Number of Experts and Qualifications for Ground Truth

• Number of Experts: The ground truth for the test set was established by "consultants (physicians and dosimetrists)." The exact number is not specified, but it implies a team. These initial segmentations were then "reviewed and corrected by a radiation oncologist." Finally, "Qualified staff at MIM Software (M.D. or licensed dosimetrists) then performed a final review and correction."
• Qualifications of Experts:
  • Consultants: Physicians and dosimetrists.
  • Review and Correction: Radiation oncologist.
  • Final Review and Correction: Qualified staff at MIM Software (M.D. or licensed dosimetrists).
  • All segmenters and reviewers were given "detailed instructions derived from relevant published clinical contouring guidelines" and instructed to ensure the "highest quality training data."

4. Adjudication Method for the Test Set

The adjudication method involved a multi-stage process:

1. Initial Segmentation: Done by consultants (physicians and dosimetrists).
2. First Review & Correction: By a radiation oncologist.
3. Final Review & Correction: By qualified staff (M.D. or licensed dosimetrists) at MIM Software.

This indicates a sequential review process, rather than a specific (e.g., 2+1, 3+1) consensus model among peers at the same stage.

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

No MRMC comparative effectiveness study was explicitly described comparing human readers with AI assistance versus without AI assistance. The study focused on the algorithm's standalone performance compared to an atlas-based predicate device, and a preliminary user evaluation for time-saving was mentioned, but not in the context of an MRMC study.

6. Standalone (Algorithm Only) Performance

Yes, a standalone (algorithm only) performance study was conducted. The Dice coefficient results presented in the table demonstrate the performance of the Contour ProtégéAI algorithm compared to the MIM Maestro atlas-based segmentation, without human intervention in the segmentation process being evaluated. The document explicitly states the "performance of both segmentation devices was measured by calculating the Dice score of the novel segmentations with the original ground-truth contours."

7. Type of Ground Truth Used

The ground truth used was expert consensus. It was established by a multi-stage review and correction process involving physicians, dosimetrists, a radiation oncologist, and qualified MIM Software staff who re-segmented images "specifically for this purpose, outside of clinical practice" and were instructed to adhere to "relevant published clinical contouring guidelines."

8. Sample Size for the Training Set

The training set consisted of 326 CT images gathered from 37 clinical sites across multiple countries (USA, Hong Kong, Australia).

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

The ground truth for the training set was established through a rigorous, multi-step expert review process:

1. CT images (from clinical treatment plans) were re-segmented by consultants (physicians and dosimetrists).
2. These initial segmentations were then reviewed and corrected by a radiation oncologist against the same standards and guidelines.
3. A final review and correction was performed by qualified staff at MIM Software (M.D. or licensed dosimetrists).
   All involved in ground truth establishment were given "detailed instructions derived from relevant published clinical contouring guidelines" and were explicitly asked "to spend additional time to ensure the highest quality training data" and to contour all specified structures "according to referenced standards."

Ask a specific question about this device

MIM software is used by trained medical professionals as a tool to aid in evaluation and information management of digital medical images. The medical image modalities include, but are not limited to, CT, MRI, CR, DX, MG, US, SPECT, PET and XA as supported by ACR/NEMA DICOM 3.0. MIM assists in the following indications:

· Receive, transmit, store, retrieve, display, print, and process medical images and DICOM objects.

• · Create, display and print reports from medical images.
  · Registration, fusion display, and review of medical images for diagnosis, treatment evaluation, and treatment planning.

· Evaluation of cardiac left ventricular function and perfusion, including left ventricular end-systolic volume, and ejection fraction.

· Localization and definition of objects such as tumors and normal tissues in medical images.

· Creation, transformation, and modification of contours for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.

· Quantitative and statistical analysis of PET/SPECT brain scans by comparing to other registered PET/SPECT brain scans.

· Planning and evaluation of permanent implant brachytherapy procedures (not including radioactive microspheres).

• · Calculating absorbed radiation dose as a result of administering a radionuclide.
  When using this device clinically within the United States, the user should only use FDA-approved radiopharmaceuticals. If used with unapproved ones, this device should only be used for research purposes.

Lossy compressed mammographic images and digitized film screen images must not be reviewed for primary image interpretations. Images that are printed to film must be printed using a FDA-approved printer for the diagnosis of digital mammography images. Mammographic images must be viewed on a display system that has been cleared by the FDA for the diagnosis of digital mammography images. The software is not to be used for mammography CAD.

MIM - Additional Tracers is an expansion of the standalone software application MIM -On Linux (K190379). MIM - Additional Tracers Indications for Use have not been modified, and the Intended Use is the same as in MIM - On Linux. With the addition of additional tracers to the standalone MIM software, engineering drawings, schematics, etc. are not applicable to the device.

These Indications for Use include quantitative and statistical analysis of PET/SPECT brain scans by comparing to other registered PET/SPECT brain scans. MIM - Additional Tracers includes the above features and capabilities and adds support for new PET/SPECT new tracers. While MIM - On Linux supported FDG and HMPAO tracers, this special 510k submission includes new tracers support for: Amyvid™ (Florbetapir), Vizamyl™ (Flutemetamol), Neuraceq™ (Florbetaben), and DaTscan™ (Joflupane).

MIM - Additional Tracers operates on Windows, Mac, and Linux computer systems.

The provided text describes the acceptance criteria and a study proving the device meets those criteria, specifically for the "MIM - Additional Tracers" software (K223800). The key information is extracted below.

1. Table of Acceptance Criteria and Reported Device Performance

The document doesn't explicitly list numerical acceptance criteria. Instead, it describes a validation process focused on the accuracy of template registration and quantitative analysis for the newly supported PET/SPECT tracers. The "reported device performance" is a qualitative statement of approval by experts.

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

• Sample Size: "Normal patient scans were curated to span demographics appropriate for each tracer." The exact number of scans is not specified.
• Data Provenance: Not explicitly stated (e.g., country of origin, retrospective/prospective). It is implied that these are existing patient scans curated for the purpose of creating normal databases and performing quantitative analysis.

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

• Number of Experts: A "radiologist" (singular) and "MIM technical experts" (plural) were involved. The exact number of MIM technical experts is not specified.
• Qualifications: "Radiologist" and "MIM technical experts" are provided as qualifications. No specific experience levels (e.g., "10 years of experience") are given for the radiologist.

4. Adjudication Method for the Test Set

The adjudication method appears to be a consensus-based approach with expert review and approval. For template registration accuracy, a radiologist and MIM technical experts assessed and approved. For normal patient database creation, the same group approved. For quantitative analysis, results were compared to expert reads and reviewed by physicians and MIM technical experts. Discrepancies were noted and deemed acceptable only under specific conditions (misregistered, borderline, or poor-quality scans). This suggests a process where expert input directly defines or validates the acceptable outcomes.

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

No MRMC comparative effectiveness study is mentioned. The study described focuses on the standalone performance of the software with expert review, not on human readers' improvement with AI assistance.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study

Yes, a standalone performance study was done. The description indicates that the software's template registration accuracy and quantitative analysis results were assessed by experts. The "risk mitigation for automatic registration error" built into the software, which includes "registration adjustment and verification steps" by the user, implies a workflow involving a human in the loop. However, the initial assessment and comparison of quantitative results to expert reads suggest an evaluation of the algorithm's output independently, even if human verification is part of the clinical workflow.

7. Type of Ground Truth Used

The ground truth was established through expert consensus/reads. For template registration and contour accuracy, a radiologist and MIM technical experts approved the alignment and contours. For quantitative analysis, results were compared to "expert reads" and reviewed by physicians and MIM technical experts.

8. Sample Size for the Training Set

The document does not explicitly mention a separate "training set" or its sample size. The description focuses on creating "normal patient scans...curated to span demographics appropriate for each tracer" which were then aligned to template space to create "databases for each tracer." These databases could be considered analogous to a reference/training set for the software's internal operations or for comparison, but it's not explicitly labeled as such for a machine learning model.

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

If the "normal patient scans" and "databases for each tracer" are considered part of a training or reference set, then their ground truth was established by:

• Expert Approval: "Alignment to template space and accuracy of analysis contours for each tracer were approved by a radiologist and MIM technical experts before each series could be included in the normals database."

Ask a specific question about this device

MIM software is used by trained medical professionals as a tool to aid in evaluation and information management of digital medical images. The medical image modalities include, but are not limited to, CT, MRI, CR, DX, MG, US, SPECT, PET and XA as supported by ACR/NEMA DICOM 3.0. MIM assists in the following indications:

· Receive, transmit, store, retrieve, display, print, and process medical images and DICOM objects.

· Create, display and print reports from medical images.

· Registration, fusion display, and review of medical images for diagnosis, treatment evaluation, and treatment planning.

· Evaluation of cardiac left ventricular function, including left ventricular end-diastolic volume, end-systolic volume, and ejection fraction.

· Localization and definition of objects such as tumors and normal tissues in medical images.

· Creation, transformation, and modification of contours for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.

· Quantitative and statistical analysis of PE7/SPECT brain scans by comparing to other registered PET/SPECT brain scans.

· Planning and evaluation of permanent implant brachytherapy procedures (not including radioactive microspheres).

· Calculating absorbed radiation dose as a result of administering a radionuclide.

· Assist with the planning and evaluation procedures by providing visualization and analysis, including energy zone visualization through the placement of virtual ablation devices validated for inclusion in MIM-Ablation. The software is not intended to predict specific ablation zone volumes or predict ablation success.

When using device clinically, within the United States, the user should only use FDA approved radionly, If using with unapproved ones, this device should only be used for research purposes.

Lossy compressed mammographic images and digitized film screen images must not be reviewed for primary image interpretations. Images that are printed to film must be printed using a FDA-approved printer for the diagnosis of digital mammography images. Mammographic images must be viewed on a display system that has been cleared by the FDA for the diagnosis of digital mammography images. The software is not to be used for mammography CAD.

MIM - Ablation is a standalone software application that allows for the planning and evaluation of ablation procedures. This is achieved by utilizing the following functionality:

• . Manual and automatic tools for normal structure, target region, and ablation zone segmentation
• lmage re-slicing and reorientation orthogonally to a user-defined angle to give a . "probe's-eye view" image for planning
• Manual and constraint-driven placement of virtual ablation devices on medical . imaging in order to visualize the ablation energy zones.
• . The calculation of the percentage of designated structures that are covered by each energy zone during planning, as well as a calculation of the final ablation zone coverage after the ablation has been performed
• Multimodality image registration, including rigid and deformable fusion, for the . comparison of images taken at different times during the ablation planning and treatment administration
  MIM - Ablation is run on a dedicated workstation in the hospital healthcare environment and can be used with an 3D DICOM image. The software can be used on image data for any patient demographic that is undergoing ablation treatment with devices validated for inclusion in MIM - Ablation.

The acceptance criteria and study proving MIM-Ablation meets these criteria are detailed below, based on the provided FDA 510(k) summary.

MIM-Ablation Acceptance Criteria and Performance Study

The MIM-Ablation software, as described in the 510(k) summary, demonstrates its efficacy and safety through specific performance testing. The core functionality validated is the accurate representation and calculation of "energy zones" (simulated ablation zones) based on manufacturer specifications of validated ablation devices within the software.

1. Table of Acceptance Criteria and Reported Device Performance

• Percent difference ranged from -3.75% to 1.53%, falling within the manufacturer tolerance. |
  | Image Resolution Independence | Volume of imported 3D energy zone objects in MIM is independent of image resolution (0.5mm, 1.0mm, 1.5mm). | Percent Difference in Volume:
• At 0.5 mm: -1.65% to 0.29%
• At 1.0 mm: -0.87% to -0.6%
• At 1.5 mm: -0.70% to 0.00%.
  This demonstrates consistency across various resolutions. |
  | Contour Resolution Independence | Dimensions of imported 3D energy zone objects in MIM match manufacturer specifications, independent of contour resolution. | Percent Difference in Dimensions:
• At 0.25 mm: -1.43% to -0.43%
• At 0.5 mm: -1.16% to 0.00%
• At 1.0 mm: -2.15% to -0.38%.
  This indicates independence from contour resolution. |
  | Image Modality Independence | Dimensions of 3D energy zone objects in MIM match manufacturer specifications across four image modalities. | Percent Difference in Dimensions:
• Ranged from 0.64% to 0.89%.
  This highlights the minimal effect of image modality on contour dimensions. |
  | Percent Coverage Calculation | Accuracy of "Percent Coverage" statistic (volume of structure covered by energy zone) with one or multiple ablation probes. | Percent Difference in Calculation:
• With one ablation probe: 0.00% to 1.57%
• With two ablation probes: 0.00% to 0.19%.
  This validates the accuracy of the coverage calculation feature. |
  | HIFU Energy Zone Dimensions | Dimensions of the unique HIFU energy zone (3cm and 4cm transducer treatment heights, overlap) match manufacturer specifications. | Percent Error in Measurements (MIM vs. SonoBlate):
• Ranged from 0.00% to 6.00%.
  These measurements fell within the manufacturer tolerance, verifying consistency in HIFU dimensions. |
  | Constraint-Driven Planning | Constraint-driven planning functionality places energy zones that adhere to user-set constraints, and indicates when planning is not possible. | Verified Functionality:
• The work verified that energy zones adhere to constraints and that an "unattainable plan" is indicated when a scenario is not possible, ensuring appropriate targeting and preventing impossible planning. |

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

The provided document does not specify the exact sample size for the test set in terms of an overall number of cases or images. Instead, the testing appears to be highly controlled and synthetic or semi-synthetic, focusing on the accuracy of internal calculations and representations based on input parameters (manufacturer specifications, different resolutions, different contour resolutions, multiple modalities, and various probe configurations).

• Data Provenance: The data provenance is primarily "internal" verification data, based on manufacturer specifications (user guides, marketing material) of the Varian V-Probes and SonaBlate HIFU system. This suggests a controlled environment, likely using simulated or canonical data derived from these specifications. The document does not explicitly state the country of origin of the data or whether it was retrospective or prospective clinical data. Given the "verification and validation testing" language and the precise percentage differences, it points towards rigorous technical and performance testing rather than a clinical study on patient data.

3. Number of Experts and Qualifications for Ground Truth

The document does not describe the establishment of ground truth through expert consensus for the performance testing. The ground truth for the device's calculations and representations of ablation zones is explicitly stated as the manufacturer specifications of the validated ablation devices (Varian V-Probes and SonaBlate HIFU system). This is a technical ground truth rather than a clinical one from human experts evaluating medical images.

4. Adjudication Method for the Test Set

No adjudication method is mentioned, as the ground truth is derived from manufacturer specifications rather than subjective expert interpretations requiring adjudication.

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

No MRMC comparative effectiveness study was done. The study described focuses on the device's technical performance in accurately representing and calculating energy zones, not on its impact on human reader performance or the improvement of human readers with AI assistance.

6. Standalone Performance (Algorithm Only)

The testing described is primarily focused on the standalone performance of the algorithm in generating and calculating energy zones based on defined inputs (manufacturer specifications, different image/contour resolutions, etc.). It verifies the internal consistency and accuracy of the software's representations independent of a human operator, beyond the initial input parameters.

7. Type of Ground Truth Used

The ground truth used for this study is manufacturer specifications (from user guides and marketing material) of the validated ablation devices. This serves as the engineering/technical ground truth against which the software's generated energy zone dimensions and volumes are compared.

8. Sample Size for the Training Set

The document does not describe a "training set" in the context of a machine learning model that learns from data. MIM-Ablation's functionality, as described, appears to be based on predefined models and algorithms derived from the physical specifications of ablation devices, rather than a data-driven machine learning approach requiring a training set. The descriptions point to a rule-based or model-based system.

9. How Ground Truth for the Training Set Was Established

As no training set (for machine learning) is implied, there is no mention of how ground truth for a training set was established. The "ground truth" for the device's functionality is the engineering specifications of the ablation devices, which were used to "generate models and import into MIM-Ablation."

Ask a specific question about this device

Trained medical professionals use Contour ProtégéAI as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0. In addition, Contour ProtégéAI supports the following indications:

· Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transfering contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.

· Segmenting normal structures across a variety of CT anatomical locations.

· And segmenting normal structures of the prostate, seminal vesicles, and urethra within T2-weighted MR images.

Appropriate image visualization software must be used to review and, if necessary, edit results automatically generated by Contour ProtégéAI.

Contour ProtégéAl is an accessory to MIM software that automatically creates contours on medical images through the use of machine-learning algorithms. It is designed for use in the processing of medical images and operates on Windows, Mac, and Linux computer systems. Contour ProtégéAl is deployed on a remote server using the MIMcloud service for data management and transfer; or locally on the workstation or server running MIM software.

The user is asking for information about the acceptance criteria and study proving the device's performance from the provided FDA 510(k) summary. I will extract the relevant information to answer each point.

Here's a breakdown of the requested information and where to find it in the document:

1. A table of acceptance criteria and the reported device performance: This will primarily come from the "Testing and Performance Data" section, specifically the table comparing MIM Atlas and Contour ProtégéAI Dice coefficients and the equivalence definition.
2. Sample sizes used for the test set and the data provenance: Found in the "Testing and Performance Data" section.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Found in the "Testing and Performance Data" section.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set: Found in the "Testing and Performance Data" section regarding ground truth generation.
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: The document describes a comparison between the AI (Contour ProtégéAI) and an atlas-based segmentation (MIM Maestro reference device), not a human-in-the-loop study with human readers comparing performance with and without AI assistance.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: The provided data compares the algorithm's performance against a ground truth and an atlas-based reference algorithm. The use of "appropriate image visualization software must be used to review and, if necessary, edit results automatically generated by Contour ProtégéAI" implies it's an AI-assisted tool, but the testing itself appears to be an algorithmic comparison.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.): Found in the "Testing and Performance Data" section.
8. The sample size for the training set: Found in the "Device Description" and "Testing and Performance Data" sections.
9. How the ground truth for the training set was established: Found in the "Testing and Performance Data" section.

Here's the detailed response based on the provided document:

Acceptance Criteria and Study Proving Device Performance

The study evaluated the performance of Contour ProtégéAI, specifically its new 3.0.0 CT neural network models, by comparing its segmentation accuracy (Dice coefficient) against a reference atlas-based segmentation device, MIM Maestro (K071964).

1. Table of Acceptance Criteria and Reported Device Performance:

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

• Test Set Size: 739 independent images.
• Data Provenance: Gathered from 12 institutions. The specific countries for the test set are not explicitly stated, but the training data (from which test subjects were explicitly excluded) was from Australia, France, Hong Kong, and the USA. The data collection was prospective in the sense that the training data explicitly excluded patients from the institutions contributing to the test set, ensuring independence.

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

• Number of Experts: Not explicitly stated as a fixed number.
• Qualifications of Experts: Ground truth segmentations were generated by a "trained user (typically, a dosimetrist or radiologist)" and then reviewed and approved by a "supervising physician (typically, a radiation oncologist or a radiologist)."

4. Adjudication method for the test set:

• The ground truth generation process involved: initial segmentation by a trained user, followed by review and approval by a supervising physician. If necessary, the data was sent back for re-segmentation and re-review. This constitutes an iterative consensus-building method rather than a strict 2+1 or 3+1 type of adjudication.

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, an MRMC comparative effectiveness study involving human readers' improvement with AI vs. without AI assistance was not conducted or reported in this summary. The study focused on the standalone algorithmic performance of the AI tool (Contour ProtégéAI) compared to an existing atlas-based automatic segmentation method (MIM Maestro). The device is intended as a "tool to assist" and mandates review/editing by users, but the performance study itself was not a human-in-the-loop clinical trial.

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

• Yes, the primary study reported is a standalone algorithmic performance comparison. The Dice coefficients were calculated for the algorithm's output directly against the established ground truth, and then compared to the performance of the MIM Maestro atlas segmentation reference device.

7. The type of ground truth used:

• The ground truth used was expert consensus segmentation, established by trained users (dosimetrists or radiologists) and approved by supervising physicians (radiation oncologists or radiologists).

8. The sample size for the training set:

• Training Set Size: 4061 CT images.

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

• The ground-truth segmentations used for both training and validation (test set) were established using the same method: generated by a "trained user (typically, a dosimetrist or radiologist)" that were then "reviewed and approved by a supervising physician (typically, a radiation oncologist or a radiologist) and sent back for re-segmentation and re-review as necessary."

Ask a specific question about this device