Search Results

This device is indicated to acquire and display cross sectional volumes of the whole body, to include the head, with the capability to image whole organs in a single rotation. Whole organs include but are not limited to brain, heart, pancreas, etc. The Aquilion ONE has the capability to provide volume sets of the entire organ. These volume sets can be used to perform specialized studies, using indicated software/hardware, of the whole organ by a trained and qualified physician.

FIRST is an iterative reconstruction algorithm intended to reduce exposure dose and improve high contrast spatial resolution for abdomen, pelvis, chest, cardiac, extremities and head applications.

AiCE is a noise reduction algorithm that improves image quality and reduces image noise by employing Deep Convolutional Network methods for abdomen, pelvis, lung, cardiac, extremities, head, and inner ear applications.

The spectral imaging system allows the system to acquire two nearly simultaneous CT images of an anatomical location using distinct tube voltages and/or tube currents by rapid KV switching. The Xray dose will be the sum of the dose at each respective tube voltage and current in a rotation. Information regarding the material composition of various organs, tissues, and contrast materials may be gained from the differences in X-ray attenuation between these distinct energies. When used by a qualified physician, a potential application is to determine the course of treatment.

PIQE* is a Deep Learning Reconstruction method designed to enhance spatial resolution. By incorporating noise reduction into the Deep Convolutional Network (DCNN), it is possible to achieve both spatial resolution improvement and noise reduction for cardiac, abdomen and pelvis, and lung applications, in comparison to FBP and hybrid iterative reconstruction.

CLEAR Motion is a Deep Learning Reconstruction (DLR) method designed to reduce motion artifacts. A Deep Convolutional Network (DCNN) is used to estimate the patient's motion. This information is used in the reconstruction process to obtain lung images with less motion artifacts.

Aquilion ONE (TSX-308A/3) V1.5 is a whole body multi-slice helical CT scanner, consisting of a gantry, couch and a console used for data processing and display. This device captures cross sectional volume data sets used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician. This system is based upon the technology and materials of previously marketed Canon CT systems.

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

1. Table of Acceptance Criteria and Reported Device Performance:

The document doesn't explicitly state quantitative acceptance criteria in a dedicated section. However, it implicitly defines performance through comparisons to a predicate device and statements about image quality.

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

The document mentions the use of "phantoms" for image quality evaluations and "clinical images" for performance testing.

• Phantom Studies:
  • Sample Size: Not explicitly stated, but multiple phantoms were used (e.g., water phantom, thoracic dynamic phantom). The exact number of scans or reconstructed images from these phantoms is not provided.
  • Data Provenance: Not explicitly stated, but phantom studies typically involve controlled, non-clinical data generation.
• Clinical Image Evaluations:
  • Sample Size: Not explicitly stated; "Representative body, cardiac, chest, head, and extremity diagnostic images" were used. The exact number of cases is not provided.
  • Data Provenance: Implied to be retrospective clinical data, as they are "obtained using the subject device" and "reviewed by American Board-Certified Radiologists." No country of origin is specified.

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

• Number of Experts: Not explicitly stated for each specific evaluation. For clinical image evaluation, it states "American Board-Certified Radiologists" (plural), indicating more than one.
• Qualifications of Experts: "American Board-Certified Radiologists." No specific years of experience are mentioned.

4. Adjudication Method for the Test Set:

• Clinical Images: For the clinical image quality evaluation, it states "reviewed by American Board-Certified Radiologists." It doesn't specify an adjudication method (e.g., 2+1, 3+1, none). It implies a consensus or individual assessment to confirm diagnostic quality.
• Phantom Studies: Phantoms have inherent, objective ground truth based on their design and known properties, so expert adjudication isn't typically applicable in the same way.

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

• No, an MRMC comparative effectiveness study was not explicitly described in the provided text. The document focuses on showing substantial equivalence through phantom studies and a general statement about diagnostic quality of clinical images, rather than a comparative study of human readers with and without AI assistance to quantify improvement.

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

• Yes, standalone performance was evaluated. The "Image Quality Evaluations" and "CLEAR Motion Evaluations" using phantoms are examples of standalone performance testing. These tests assess the device's algorithms (PIQE, CLEAR Motion) directly against objective metrics or by comparing reconstructed images for specific features (e.g., noise reduction, motion artifact reduction) without human intervention in the diagnostic process.

7. Type of Ground Truth Used:

• For Phantom Studies (PIQE, CLEAR Motion): Objective ground truth derived from the known physical properties and design of the phantoms (e.g., known image metrics, controlled motion patterns).
• For Clinical Image Quality: Expert consensus/review by "American Board-Certified Radiologists" to confirm "diagnostic quality."

8. Sample Size for the Training Set:

• Not provided. The document describes the device, its features (some of which use Deep Learning Reconstruction), and details of performance testing. It does not include information about the size or nature of the training data used for the AI algorithms (AiCE, PIQE, CLEAR Motion).

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

• Not provided. As the training set details are absent, the method for establishing its ground truth is also not mentioned.

Ask a specific question about this device

This device is indicated to acquire and display cross sectional volumes of the whole the head, with the capability to image whole organs in a single rotation. Whole organs include but are not limited to brain, heart, pancreas, etc. The Aquilion ONE has the capability to provide volume sets of the entire organ. These volume sets can be used to perform specialized studies, using indicated software, of the whole organ by a trained and qualified physician.

FIRST is an iterative reconstruction algorithm intended to reduce exposure dose and improve high contrast spatial resolution for abdomen, pelvis, chest, cardiac, extremities and head applications.

AiCE is a noise reduction algorithm that improves image quality and reduces image noise by employing Deep Convolutional Network methods for abdomen, pelvis, lung, cardiac, brain, extremities, head, and inner ear applications.

The spectral imaging system allows the system to acquire two nearly simultaneous CT images of an anatomical location using distinct tube voltages and/or tube currents by rapid KV switching. The X-ray dose will be the sum of the dose at each respective tube voltage and current in a rotation.

Information regarding the material composition of various organs, tissues, and contrast materials may be gained from the differences in X-ray attenuation between these distinct energies. When used by a qualified physician, a potential application is to determine the course of treatment.

PIQE is a Deep Learning Reconstruction method designed to enhance spatial resolution. By incorporating noise reduction into the Deep Convolutional Network (DCNN), it is possible to achieve both spatial resolution improvement and noise reduction for cardiac, abdomen, and pelvis applications, in comparison to FBP and hybrid iterative reconstruction.

Aquilion ONE (TSX-308A/3) V1.4 with PIQE Reconstruction System is a whole body multi-sice helical CT scanner, consisting of a gantry, couch and a console used for data processing and display. This device captures cross sectional volume data sets used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician. This system is based upon the technology and materials of previously marketed Canon CT systems.

Aquilion ONE (TSX-308A/3) V1.4 with PIQE Reconstruction System is equipped with PIQE, a deep learning reconstruction technology designed to fully utilize the maximum resolution of the detector, intended to improve spatial resolution. Original image data is available to end users when PIQE images are used for diagnosis.

The provided text describes the regulatory clearance of a medical device, the Aquilion ONE (TSX-308A/3) V1.4 with PIQE Reconstruction System, and its performance evaluation. However, it does not explicitly detail a separate "acceptance criteria" table with specific thresholds or the complete results of a study designed solely to prove the device meets these criteria in a quantitative, acceptance-testing style format.

Instead, the document details "Performance Testing - Bench" and "Performance Testing - Clinical Images" that demonstrate the device's substantial equivalence to a predicate device and its overall image quality.

Here's an attempt to structure the information based on your request, inferring acceptance implied by demonstrating equivalence and diagnostic quality:

Acceptance Criteria and Performance Study for Aquilion ONE (TSX-308A/3) V1.4 with PIQE Reconstruction System

The acceptance criteria for the Aquilion ONE (TSX-308A/3) V1.4 with PIQE Reconstruction System are implicitly derived from its claim of substantial equivalence to the predicate device (Aquilion ONE (TSX-306A/3) V10.12 with Spectral Imaging System) and its ability to produce diagnostic quality images. The study supporting this involved a combination of phantom-based bench testing and clinical image reviews.

1. Table of Implied Acceptance Criteria and Reported Device Performance

• Contrast-to-Noise Ratios
• CT Number Accuracy
• Uniformity
• Slice Sensitivity Profile
• Modulation Transfer Function
• Standard Deviation of Noise Power Spectra
• Low Contrast Detectability. |
  | PIQE Performance | PIQE (Deep Learning Reconstruction method) is designed to enhance spatial resolution by incorporating noise reduction into the Deep Convolutional Network (DCNN). This achieves both spatial resolution improvement and noise reduction for cardiac, abdomen, and pelvis applications, in comparison to FBP (Filtered Back Projection) and hybrid iterative reconstruction. Its performance was evaluated as part of the overall "General Image Quality" assessment, contributing to the "equivalent or improved" statement. |
  | SilverBeam Dose Reduction | A phantom study confirmed that DR-mode (with SilverBeam Filter) resulted in dose reduction in Head/Body modes compared to normal scan mode. |
  | Low Contrast Detectability| A phantom study supported the following claims:
• 2 mm (0.3% contrast) detectability at 15.3 mGy CTDIvol using AIDR3D.
• 2 mm (0.3% contrast) detectability at 14.7 mGy CTDIvol using AiCE.
• 3 mm (0.3% contrast) detectability at 5.7 mGy CTDIvol using AiCE. |
  | Diagnostic Quality | Representative body, cardiac, chest, head, and extremity clinical images reviewed by American Board-Certified Radiologists confirmed that the reconstructed images using the subject device were of diagnostic quality. |

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

• Bench Testing (Phantom Studies): The document does not specify a numerical sample size (e.g., number of phantom scans) but indicates that various phantoms were used for comprehensive image quality assessments, low contrast detectability, and dose reduction studies. The provenance type is retrospective, as it involves controlled phantom experiments. The geographical provenance is not specified but implicitly linked to Canon Medical Systems' testing facilities.
• Clinical Images: The document mentions "Representative body, cardiac, chest, head, and extremity diagnostic images." The exact number of clinical images or cases used for the review is not specified. The provenance is retrospective, as these were "obtained" and then reviewed. The country of origin for the clinical data is not explicitly stated.

3. Number of Experts Used and Their Qualifications

For the clinical image review:

• Number of Experts: The document states "reviewed by American Board-Certified Radiologists." The exact number of radiologists involved is not specified.
• Qualifications: "American Board-Certified Radiologists." No further details on years of experience are provided.

4. Adjudication Method

The document does not specify an adjudication method (e.g., 2+1, 3+1, none) for the clinical image review to establish ground truth or determine diagnostic quality. It only states that images were "reviewed" and "confirmed" to be of diagnostic quality.

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

An MRMC study was not explicitly detailed in the provided text as a an evaluation comparing human readers with AI assistance versus human readers without AI assistance. The clinical image review confirmed diagnostic quality of images produced by the device, but not the effect of AI assistance on human readers.

6. Standalone (Algorithm Only) Performance

The bench testing on phantoms and the evaluation of reconstruction algorithms (PIQE, AiCE, AIDR 3D) are indicative of standalone (algorithm only) performance regarding technical image quality metrics (CNR, MTF, noise, resolution, low contrast detectability). These tests were performed without human interaction for interpretation within the evaluation itself.

7. Type of Ground Truth Used

• Bench Testing (Phantom Studies): The ground truth for metrics like CT Number Accuracy, Uniformity, and specific low contrast targets is inherent to the known physical properties of the phantoms used. This is a form of phantom-derived ground truth.
• Clinical Images: The ground truth for "diagnostic quality" of clinical images was established by the consensus/judgment of American Board-Certified Radiologists. This is a form of expert consensus ground truth.

8. Sample Size for the Training Set

The document describes PIQE and AiCE as Deep Learning Reconstruction methods. Therefore, they would have been trained on large datasets. However, the sample size for the training set is not provided in this document.

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

For the deep learning algorithms (PIQE and AiCE) used in the device, the method for establishing the ground truth for their training set is not described in the provided text. Typically, this would involve extensive curated datasets with expert annotations or high-quality reference images.

Ask a specific question about this device

This device is indicated to acquire and display cross-sectional volumes of the whole the head, with the capability to image whole organs in a single rotation. Whole organs include, but are not limited to brain, heart, pancreas, etc.

The Aquilion ONE has the capability to provide volume sets of the entire organ. These volume sets can be used to perform specialized studies, using indicated software, of the whole organ by a trained and qualified physician.

FIRST is an iterative reconstruction algorithm intended to reduce exposure dose and improve high contrast spatial resolution for abdomen, pelvis, chest, cardiac, extremities and head applications.

AiCE is a noise reduction algorithm that improves image quality and reduces image noise by employing Deep Convolutional Network methods for abdomen, pelvis, lung, cardiac, brain, inner ear and extremities applications.

The spectral imaging system utilizes two scan modes, spectral imaging scan and dual energy scan.

The spectral imaging scan allows the system to acquire two nearly simultaneous CT images of an anatomical location using distinct tube voltages and/or tube currents by rapid KV switching. The X-ray dose will be the sum of the dose at each respective tube voltage and current in a rotation.

Information regarding the material composition of various organs, tissues, and contrast materials may be gained from the differences in X-ray attenuation between these distinct energies.

The dual energy scan, utilized for brain imaging, allows the system to acquire two CT images of the same anatomical location using distinct tube voltages and/or tube cotations. The X-ray dose will be the sum of the dose of each tube rotation at its respective tube voltage and current. Information regarding the material composition of various organs, tissues, and contrast may be gained from the differences in X-ray attenuation between these distinct energies. This information may also be used to reconstruct images at multiple energies within the available spectrum, and to reconstruct basis images that allow the visualization and analysis of anatomical and pathological materials. Performance of dual energy scan may be affected by body size and motion artifacts.

When used by a qualified physician, a potential application is to determine the course of treatment

PIQE is a Deep Learning Reconstruction method designed to enhance spatial resolution. By incorporating noise reduction into the Deep Convolutional Network (DCNN), it is possible to achieve both spatial resolution improvement and noise reduction for cardiac applications.

Vitrea Software Package is an application package developed for use on Vitrea®, a medical diagnostic system that allows the processing, review, analysis, communication and media interchange of multi-dimensional digital images acquired from a variety of imaging devices. Vitrea has the following additional indications:

The Cerebral Aneurysm Analysis application is intended to facilitate the extraction of user identified aneurysms on the cerebral arteries. The software can be used as an adjunct to diagnosis for the purposes of measurement of size and aspect ratio.

The MR Wall Motion Tracking application is intended to assist physicians with performing cardiac functional analysis based upon magnetic resonance images. It provides measurements of global and regional myocardial function that is used for patients with suspected heart disease.

The MR Coronary Tracking application is intended to assist physicians with performing coronary artery analysis for MR heart images which are intended for the qualitative and quantitative analysis of coronary arteries.

The SUREVolume Synthesis application is intended to load volume images acquired by whole-body X-ray CT scanners, X-ray angiography systems, and MRI systems and displays fusion images.

The Angio Viewer application displays image data acquired using an X-ray angiography system. It supports cine display, subtraction, and distance measurement.

The US Cardiac Fusion application enables fusion display of the analysis results obtained using the US 3D Wall Motion Tracking application and the CT Coronary Artery Analysis application.

The Ultrasound Clinical Applications are indication of structures, and dynamic processes with the human body using saved ultrasound DICOM images to provide image information for diagnosis.

The Spectral Stone Analysis application is intended to serve as an adjunct visualization tool for the differentiation between uric acid and non-uric acid stones greater than 3 mm with Spectral CT studies acquired on the Canon Medical Systems scanner.

The Spectral Composition Analysis application is intended to assist a physician in visualizing the presence of monosodium urate in anatomical structures. The clinical syndrome of gout is characterized by the presence of monosodium urate crystals in joints or soft tissue.

The Embolization Plan application is a post processing software that is intended to assist physicians in the visualization of the liver arterial tree using 3D images of CT or 3D images of Cone Beam CT acquired by Toshiba or Canon Medical Systems. It provides tools to assist the user in analysis of these images. The output is intended to be an adjunct means that allows automatic and manual planning of the liver arterial vessels for guidance of the output is a 3D visualization of the hepatic arteries to high dense lesion in the liver.

The Spectral Analysis application is a CT, non-invasive image analysis software package, which may be used to aid in the visualization of anatomical and pathological materials. The software provides quantification of Hounsfield units of iodine attenuation differences and iodine concentration and display by color.

Effective Z and Electron Density maps may aid in the differentiation of different tissues in the human body.

Aquilion ONE (TSX-306A/3) V10.12 with Spectral Imaging System is a whole body multi-slice helical CT scanner, consisting of a gantry, couch and a console used for data processing and display. This device captures cross sectional volume data sets used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician. This system is based upon the technology and materials of previously marketed Canon CT systems.

Vitrea Software Package, VSTP-001A, is an application package developed for use on Vitrea, a medical image processing software, marketed by Vital Images, Inc. Vitrea Software Package, VSTP-001A, currently includes eleven post processing applications, MR Wall Motion Tracking, Cerebral Aneurysm Analysis, MR Coronary Tracking, SUREVolume Synthesis, Angio Viewer, US Cardiac Fusion, Ultrasound Applications Package, Spectral Stone Analysis, Spectral Composition Analysis, Embolization Planning Tool and Spectral Analysis which use brain, body or cardiac image data, obtained from CT/XA/MR/US systems, to assist physicians in performing specialized measurements and analysis.

Here's a summary of the acceptance criteria and study information for the Vitrea Software Package, VSTP-001A, based on the provided FDA 510(k) summary:

Acceptance Criteria and Device Performance for Vitrea Software Package, VSTP-001A

The provided document primarily focuses on demonstrating substantial equivalence to a predicate device rather than presenting specific quantitative acceptance criteria for improved performance. The "acceptance criteria" can be inferred from the studies conducted to show that the new/modified features (specifically related to Spectral Analysis and additional display options) meet specifications and perform as intended, or offer equivalent/improved performance compared to the predicate.

1. Table of Acceptance Criteria and Reported Device Performance

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

The document does not explicitly state a sample size for a "test set" in the context of clinical images for the Vitrea Software Package. The performance testing mentioned primarily involved:

• Bench testing: This would involve laboratory-based tests on the software without human data.
• Phantom studies: These use artificial objects (phantoms) instead of human subjects.
• Clinical Images (for Aquilion ONE): For the associated CT scanner (Aquilion ONE), "Representative body, cardiac, low dose chest and head diagnostic images, reviewed by an American Board Certified Radiologist" were obtained. However, this refers to the acquisition capabilities of the CT scanner, not directly to the evaluation of the Vitrea Software Package's specific new features.

Therefore, for the Vitrea Software Package, there is no stated sample size for a clinical test set directly evaluating the software's new features. The primary evaluation focused on technical performance using phantoms.

Data Provenance: Not explicitly stated for specific test sets related to the Vitrea software. The studies mentioned are primarily technical (bench and phantom studies).

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

• For the Vitrea Software Package: The document does not specify the use of experts to establish ground truth for a test set directly related to the new software features. Evaluation appears to be based on technical specifications and phantom studies.
• For the Aquilion ONE (CT Scanner) clinical images: "An American Board Certified Radiologist" reviewed images for diagnostic quality. This indicates at least one expert was used, but the purpose was general diagnostic quality rather than establishing ground truth for specific algorithm performance metrics.

4. Adjudication Method for the Test Set

Not applicable, as no external expert-adjudicated test set is described for the Vitrea Software Package's new features.

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

A Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not described for the Vitrea Software Package. The document focuses on demonstrating substantial equivalence based on technical performance and phantom studies, and the new features are primarily tools for visualization and analysis, not AI assistance for human readers in a diagnostic setting that would typically require an MRMC study.

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

The performance testing primarily involved standalone algorithm components evaluated through bench testing and phantom studies. For example, the Spectral Analysis application's ability to generate monochromatic images, iodine maps, electron density images, etc., and the functionality of new display modes, would have been evaluated primarily in a standalone manner against technical specifications.

7. The Type of Ground Truth Used (expert consensus, pathology, outcomes data, etc.)

For the Vitrea Software Package:

• Technical specifications/expected outputs: The "ground truth" for the new features was primarily based on predefined technical specifications and expected outputs from phantom studies (e.g., accurate generation of monochromatic images, proper functioning of display modes).
• Phantom studies: Phantoms are designed with known properties, serving as a form of "ground truth" for evaluating image quality metrics.

No mention of expert consensus, pathology, or outcomes data being used as ground truth for the Vitrea Software Package's new features.

8. The Sample Size for the Training Set

Not applicable. The document describes a "Software Package" with modifications to existing features, not a deep learning algorithm that requires a "training set." While the Aquilion ONE CT scanner mentions AI-based reconstruction (AiCE, PIQE), the FDA submission for the Vitrea Software Package itself does not describe a training set for the software's functional updates.

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

Not applicable, as no training set is mentioned for the Vitrea Software Package in this submission.

Ask a specific question about this device

This device is indicated to acquire and display cross-sectional volumes of the whole the head, with the capability to image whole organs in a single rotation. Whole organs include, but are not limited to brain, heart, pancreas, etc.

The Aquilion ONE has the capability to provide volume sets of the entire organ. These volume sets can be used to perform specialized studies, using indicated software, of the whole organ by a trained and qualified physician.

FIRST is an iterative reconstruction algorithm intended to reduce exposure dose and improve high contrast spatial resolution for abdomen, pelvis, chest, cardiac, extremities and head applications.

AiCE is a noise reduction algorithm that improves image quality and reduces image noise by employing Deep Convolutional Network methods for abdomen, pelvis, lung, cardiac, brain, inner ear and extremities applications.

The Spectral Imaging System allows the system to acquire two nearly simultaneous CT images of an anatomical location using distinct tube voltages and/or tube currents by rapid KV switching. The X-ray dose will be the sum of the dose at each respective tube voltage and current in a rotation.

Information regarding the material composition of various organs, tissues, and contrast materials may be gained from the differences in X-ray attenuation between these distinct energies.

When used by a qualified physician, a potential application is to determine the course of treatment.

Aquilion ONE (TSX-306A/3) V10.4 with Spectral Imaging System is a whole body multi-slice helical CT scanner, consisting of a gantry, couch and a console used for data processing and display. This device captures cross sectional volume data sets used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician. This system is based upon the technology and materials of previously marketed Canon CT systems.

Here's a breakdown of the acceptance criteria and the study details for the Aquilion ONE (TSX-306A/3) V10.4 with Spectral Imaging System, 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

• Bench Testing (Phantom Studies): The text indicates "phantoms" were used for various image quality and dose-related tests. Specific numbers of phantoms are not provided, nor is their provenance (obviously not country of origin for phantoms).
• Clinical Images: "Representative abdomen/pelvis, lung, extremity and cardiac Spectral Images" were obtained. The exact sample size (number of images or patients) is not specified. The provenance (e.g., country of origin, retrospective/prospective) for these clinical images is also not explicitly stated, beyond being "representative."

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

• Clinical Images: "An American Board Certified Radiologist" reviewed the clinical images. This indicates one expert.
• Qualifications: American Board Certified Radiologist. The number of years of experience is not specified.

4. Adjudication Method for the Test Set

• For the clinical images, the review was performed by a single American Board Certified Radiologist. This implies no multi-reader adjudication method (like 2+1 or 3+1). The radiologist's assessment directly served as the ground truth for diagnostic quality.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, If So, What Was the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance

• No MRMC comparative effectiveness study is mentioned. The primary focus of the clinical image review was to confirm the diagnostic quality of the spectral images themselves, not to assess human reader performance with or without AI assistance. The device includes "AiCE is a noise reduction algorithm," but its impact on human reader performance is not evaluated in this submission documentation.

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

• Yes, the majority of the image quality metrics (CNR, CT Number Accuracy, Uniformity, SSP, MTF-Wire, SD, NPS, LCD, Dose Neutral, Noise Reduction, CNR Improvement, Iodine Reduction) were evaluated in a standalone fashion using phantoms. This represents an algorithm-only performance assessment against predefined physical criteria.

7. The Type of Ground Truth Used

• Bench Testing: The ground truth for the bench tests was derived from known phantom properties and measurements against those properties (e.g., known iodine concentrations in phantoms, measurements of noise, etc.). This is essentially physical/objective measurements against known standards.
• Clinical Images: The ground truth for the clinical images was expert opinion/consensus from a single American Board Certified Radiologist regarding diagnostic quality.

8. The Sample Size for the Training Set

• The document does not specify a sample size for the training set for any algorithms (like FIRST or AiCE). The submission focuses on verification and validation of the device changes, not the development or training of its internal algorithms.

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

• The document does not provide information on how the ground truth for the training set (if applicable to the algorithms like AiCE, which uses Deep Convolutional Network methods) was established. This information would typically be detailed during the development and initial validation of such algorithms, not necessarily in a submission for an updated device version that incorporates these existing algorithms.

Ask a specific question about this device

This device is indicated to acquire and display cross-sectional volumes of the whole the head, with the capability to image whole organs in a single rotation. Whole organs include, but are not limited to brain, heart, pancreas, etc.

The Aquilion ONE has the capability to provide volume sets of the entire organ. These volume sets can be used to perform specialized studies, using indicated software, of the whole organ by a trained and qualified physician.

FIRST is an iterative reconstruction algorithm intended to reduce exposure dose and improve high contrast spatial resolution for abdomen, pelvis, chest, cardiac, extremities and head applications.

AiCE is a noise reduction algorithm that improves image quality and reduces image noise by employing Deep Convolutional Neural Network methods for abdomen, pelvis, inner ear and extremities applications.

The Spectral Imaging System allows the system to acquire two nearly simultaneous CT images of an anatomical location using distinct tube voltages and/or tube currents by rapid KV switching. The X-ray dose will be the sum of the dose at each respective tube voltage and current in a rotation.

Information regarding the material composition of various organs, tissues, and contrast materials may be gained from the differences in X-ray attenuation between these distinct energies.

When used by a qualified physician, a potential application is to determine the course of treatment.

Aquilion ONE (TSX-306A/3) V10.0 with Spectral Imaging System is a whole body multi-slice helical CT scanner, consisting of a gantry, couch and a console used for data processing and display. This device captures cross sectional volume data sets used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician. This system is based upon the technology and materials of previously marketed Canon CT systems.

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

1. Table of Acceptance Criteria and Reported Device Performance

The FDA 510(k) summary does not explicitly list numerical "acceptance criteria" in the format of a table with pass/fail thresholds. Instead, it describes various tests and claims that demonstrate substantial equivalence to previously cleared devices. The performance is reported in terms of qualitative assessments (e.g., "diagnostic quality," "substantially equivalent") and quantitative improvements (e.g., "dose reduction," "improvement in low contrast detectability," "noise reduction").

Here's an interpretation of the implied acceptance criteria and reported performance:

• More similar to high dose FBP (compared to FIRST);
• More similar to FBP (compared to FIRST);
• Improved (compared to FIRST);
• More natural (compared to FIRST). | AiCE noise appearance/texture:
• More similar to high dose filtered backprojection (compared to FIRST);
• More similar to filtered backprojection (compared to FIRST);
• Improved (compared to FIRST);
• More natural (compared to FIRST). |
  | Clinical Image Quality | AiCE images for abdomen/pelvis, brain, inner ear, and extremity applications are of diagnostic quality. | Representative abdomen/pelvis, brain, inner ear and extremity AiCE images were confirmed to be of diagnostic quality by an American Board Certified Radiologist. |

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

• Spectral Imaging Performance Testing - Bench: The sample size for phantoms used in bench testing is not specified beyond "various phantoms" and "Catphan and Body Phantom." Data provenance is laboratory bench testing.
• Spectral Imaging Performance Testing - Clinical Images: The sample size is not specified beyond "Representative abdomen/pelvis, lung, and extremity Spectral Images." The data provenance of these clinical images (country of origin, retrospective/prospective) is not specified in the provided text.
• Non-Spectral Imaging and AiCE Performance Testing - Bench: The sample size for phantoms used in bench testing is not specified beyond "various phantoms." Data provenance is laboratory bench testing.
• AiCE Imaging Performance Testing - Clinical Images: The sample size is not specified beyond "Representative abdomen/pelvis, brain, inner ear and extremity AiCE images." The data provenance of these clinical images (country of origin, retrospective/prospective) is not specified in the provided text.

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

• For both Spectral Imaging Clinical Images and AiCE Clinical Images: "an American Board Certified Radiologist" was used. This indicates one expert. The specific years of experience are not mentioned, but "American Board Certified" signifies a high level of qualification.

4. Adjudication Method for the Test Set

• For both Spectral Imaging Clinical Images and AiCE Clinical Images, only one expert (an American Board Certified Radiologist) reviewed the images. Therefore, there was no adjudication method between multiple experts employed for these clinical image quality assessments. The phrase "it was confirmed" implies a singular decision.

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

• No, a MRMC comparative effectiveness study was not explicitly mentioned in the provided text for evaluating human reader improvement with AI assistance. The clinical image reviews were done by a single radiologist to confirm "diagnostic quality" of the AI-processed images, not to compare human reader performance with and without AI. The quantitative performance (dose reduction, LCD improvement, noise reduction) was assessed via model observer evaluation or phantom studies.

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

• Yes, extensive standalone performance testing was done for both Spectral Imaging and AiCE.
  • Spectral Imaging: Bench testing utilizing phantoms assessed various image quality metrics (CNR, CT Number Accuracy, Uniformity, SSP, MTF-Wire, SD, NPS, LCD) to demonstrate substantial equivalence to the predicate. Other phantom studies supported claims of beam hardening artifact reduction and iodine concentration correlation.
  • AiCE: Bench testing utilizing phantoms assessed similar image quality metrics and pediatric phantom/protocol performance, demonstrating substantial equivalence. A model observer evaluation was specifically mentioned for quantitative assessments of dose reduction, LCD improvement, and noise reduction compared to FBP and AIDR 3D, which is a standalone algorithm-only performance assessment. Other phantom studies supported claims regarding artifact appearance, spatial resolution, and noise appearance/texture.

7. The Type of Ground Truth Used

• For Bench Testing (Spectral and AiCE): The ground truth was based on physical phantom measurements and known properties/compositions of the phantoms. For example, known concentrations of materials for iodine correlation, or predefined structures for MTF and LCD assessments.
• For Clinical Image Reviews (Spectral and AiCE): The ground truth was expert consensus (single expert), specifically the judgment of an American Board Certified Radiologist that the images were of "diagnostic quality." This is a form of expert opinion or interpretation.

8. The Sample Size for the Training Set

• The document does not specify the sample size for the training set for either AiCE (Deep Convolutional Neural Network) or FIRST (Iterative Reconstruction Algorithm). It only mentions that AiCE employs "Deep Convolutional Neural Network methods."

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

• The document does not provide details on how the ground truth for the training set was established. It mentions the use of "Deep Convolutional Neural Network methods" for AiCE, which implies a supervised learning approach requiring labeled training data, but the specifics of that labeling process or who performed it are omitted.

Ask a specific question about this device

This device is indicated to acquire and display cross-sectional volumes of the whole the body, to include the head, with the capability to image whole organs in a single rotation. Whole organs include, but are not limited to brain, heart, pancreas, etc.

The Aquilion ONE has the capability to provide volume sets of the entire organ. These volume sets can be used to perform specialized studies, using indicated software, of the whole organ by a trained and qualified physician.

Aquilion ONE (TSX-306A/3) V10.0 is a whole body multi-slice helical CT scanner, consisting of a gantry, couch and a console used for data processing and display. This device captures cross sectional volume data sets used to perform specialized studies, using indicated software, c by a trained and qualified physician. This system is based upon the technology and materials of previously marketed Canon CT systems.

Here's a breakdown of the acceptance criteria and the study information based on the provided text, focusing on what is explicitly stated:

Acceptance Criteria and Reported Device Performance

The provided document describes modifications to an existing Computed Tomography (CT) system (Aquilion ONE) and seeks clearance based on substantial equivalence to a predicate device. The core "acceptance criteria" here relate to demonstrating that the modified device performs similarly to or better than the predicate device, particularly concerning image quality, and that the modifications do not introduce new safety concerns.

The study presented focuses not on the clinical performance of an AI algorithm, but on demonstrating the image quality equivalence of the modified CT system (Aquilion ONE (TSX-306A/3) V10.0) compared to its predicate (Aquilion ONE (TSX-305A/6) V8.9 with AiCE).

Study Details

It's important to note that this submission is for a CT system hardware and software update, not a separate AI-driven diagnostic device or AI assistance tool for human readers. Therefore, many of the typical questions for AI performance studies (like human reader improvement with AI, ground truth for AI training, etc.) are not directly applicable to this specific submission. The "study" here is a technical verification rather than a clinical trial.

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

   • Test Set Sample Size: Not applicable in the sense of patient data. The testing was phantom-based. The document does not specify the number of phantoms used, but describes the type of tests performed (e.g., assessing different image quality parameters).
   • Data Provenance: Not applicable as it was phantom-based testing, not clinical data from patients.

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

   • Not applicable. Ground truth for phantom-based image quality metrics is inherently defined by the physical characteristics of the phantom and the measurement techniques, not by expert interpretation.

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

   • Not applicable. This was phantom-based objective measurement, not subjective expert adjudication of clinical cases.

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, an MRMC comparative effectiveness study was not done. This submission is for a CT system update, not an AI assistance device that impacts human reader performance.
   • Effect size of human readers improving with AI vs. without AI assistance: Not applicable.

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

   • Not applicable as this is a CT scanner, not a standalone AI algorithm. While the predicate device included "AiCE" (Canon's Advanced Intelligent Clear-IQ Engine, an AI-based reconstruction technology), the primary focus of this submission's testing is on the general image quality equivalence of the updated system with its modified hardware and software, rather than a specific performance study of AiCE itself or any new AI feature. The subject device removes AiCE as an option from this specific version, indicating the focus is elsewhere.

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

   • Objective physical measurements from phantom-based tests. The "ground truth" for image quality metrics like noise, spatial resolution, and density resolution is established through controlled measurements of known phantom properties, not clinical outcomes or expert consensus.

7. The sample size for the training set:

   • Not applicable. This document describes the verification of a modified CT system, not the training of a new AI model.

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

   • Not applicable. (See #7).

Ask a specific question about this device

This device is indicated to acquire and display cross-sectional volumes of the whole the head, with the capability to image whole organs in a single rotation. Whole organs include, but are not limited to brain, heart, pancreas, etc.

The Aquilion ONE has the capability to provide volume sets of the entire organ. These volume sets can be used to perform specialized studies, using indicated software, of the whole organ by a trained and qualified physician.

FIRST is an iterative reconstruction algorithm intended to reduce exposure dose and improve high contrast spatial resolution for abdomen, pelvis, chest, cardiac, extremities and head applications.

AiCE is a noise reduction algorithm that improves image quality and reduces image noise by employing Deep Convolutional Network methods for abdomen, pelvis, lung and cardiac applications.

Aguilion ONE (TSX-305A/6) V8.9 with AiCE is a whole body multi-slice helical CT scanner, consisting of a gantry, couch and a console used for data processing and display. This device captures cross sectional volume data sets used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician. This system is based upon the technology and materials of previously marketed Canon CT systems. In addition, the subject device incorporates the latest reconstruction technology, AiCE (Advanced intelligent Clear-IQ Engine), intended to reduce image noise and improve image quality by utilizing Deep Convolutional Neural Network methods. These methods can more fully explore the statistical properties of the signal and noise. By learning to differentiate structure from noise, the algorithm produces fast, high quality CT reconstruction.

Acceptance Criteria and Study Proving Device Performance: Canon Medical Systems Corporation Aquilion ONE (TSX-305A/6) V8.9 with AiCE

This document outlines the acceptance criteria and the study conducted to prove that the Aquilion ONE (TSX-305A/6) V8.9 with AiCE (Advanced intelligent Clear-IQ Engine) CT system meets its performance claims, as detailed in the provided FDA 510(k) summary (K183046).

The AiCE algorithm is a noise reduction algorithm that utilizes Deep Convolutional Neural Network methods to improve image quality and reduce image noise for abdomen, pelvis, lung, and cardiac applications. The primary goal of the study was to demonstrate that the AiCE system provides improved image quality, reduced noise, and better low-contrast detectability compared to the predicate device and other reconstruction methods, while maintaining diagnostic quality.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for the Aquilion ONE (TSX-305A/6) V8.9 with AiCE are based on the comparison of its image quality metrics against established benchmarks, primarily the predicate device (Aquilion ONE (TSX-305A/3) V8.3 with FIRST 2.1) and other reconstruction methods (AIDR 3D, Filtered Back Projection).

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

• Sample Size for Test Set: The documentation does not explicitly state a specific number of cases or scans used for the quantitative phantom studies or the qualitative diagnostic image review. The qualitative evaluation mentions "Representative abdomen/pelvis, lung, and cardiac diagnostic images," implying a selection of cases rather than a large cohort. The quantitative evaluations used various phantoms.
• Data Provenance: The studies were phantom-based for quantitative metrics and included "Representative ... diagnostic images" for qualitative assessment. There is no information provided regarding the country of origin of the diagnostic images, nor whether they were retrospective or prospective patient data. Given the context of a 510(k) submission primarily relying on technical performance and equivalence, large-scale patient outcome studies (prospective or retrospective) are often not required if technical equivalence and safety can be demonstrated through phantom studies and limited clinical image review.

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

• Number of Experts: For the qualitative assessment of diagnostic images, it is stated that images were "reviewed by an American Board Certified Radiologist." This indicates one expert was used for this part of the evaluation.
• Qualifications of Experts: The expert was an "American Board Certified Radiologist." No specific years of experience are mentioned. For the quantitative phantom studies, experts establishing ground truth would typically be physicists or engineers, but this is not explicitly detailed.

4. Adjudication Method for the Test Set

For the qualitative assessment of diagnostic images, the documentation states that images were "reviewed by an American Board Certified Radiologist, and it was confirmed that the AiCE reconstructed images were of diagnostic quality." This implies a single-reader assessment for confirming diagnostic quality, rather than a multi-reader or consensus-based adjudication method (e.g., 2+1 or 3+1). For the phantom studies, adjudication methods for ground truth are generally not applicable as the "ground truth" is derived from the known properties of the phantoms or precise measurements.

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

No, an MRMC comparative effectiveness study was not explicitly reported in this 510(k) summary. The evaluation of the AiCE algorithm appears to have focused on:

• Quantitative measurements using phantoms (spatial resolution, LCD, noise, dose reduction).
• Qualitative review by a single radiologist to confirm diagnostic quality of "representative" patient images.
• A phantom-based "noise texture reader study" to compare noise appearance, but this is not described as a full MRMC study for diagnostic accuracy or reader performance.

Therefore, no effect size or improvement in human reader performance with AI assistance vs. without AI assistance is detailed in this submission. The AiCE algorithm acts as an inherent image reconstruction/processing component of the CT system, improving the fundamental image quality before interpretation, rather than an AI-assisted diagnostic tool for specific pathologies that would typically undergo MRMC studies.

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

Yes, a standalone (algorithm only) performance assessment was conducted for the key imaging metrics. The phantom studies directly evaluate the performance of the AiCE algorithm in terms of:

• Quantitative Spatial Resolution (7.4 lp/cm MTF)
• Quantitative Body LCD Improvement (12.2% better)
• Quantitative Noise Reduction (29.2% less noise)
• Quantitative Dose Reduction (79.6-82.4% reduction vs. FBP)
• Noise Texture/Appearance (more similar to FBP than FIRST)
• General CT Image Quality metrics (CNR, CT Number Accuracy, Uniformity, SSP, MTF, SD, NPS, Pediatric water phantom).

These measurements assess the intrinsic performance of the AiCE algorithm's output (the reconstructed image) independent of human interpretation.

7. Type of Ground Truth Used

• Quantitative Studies (Spatial Resolution, LCD, Noise, Dose): The ground truth was established through phantom studies. Phantoms have precisely known physical properties and are designed to provide a measurable "truth" for imaging performance metrics (e.g., specific object sizes for spatial resolution, known contrast differences for LCD).
• Qualitative Assessment (Diagnostic Image Quality): The ground truth was expert consensus / opinion from a "single American Board Certified Radiologist" who confirmed the "diagnostic quality" of the AiCE reconstructed images.
• Noise Texture Reader Study: The ground truth for noise texture comparison was based on the collective assessment/preference of readers in that dedicated phantom study, establishing what "more similar to standard filtered backprojection" means.

8. Sample Size for the Training Set

The 510(k) summary does not provide details on the sample size used for training the AiCE Deep Convolutional Neural Network. This information is typically proprietary and not always required in detail for a 510(k) submission, especially when the device is an image reconstruction algorithm rather than a diagnostic AI that provides a specific clinical output (e.g., detection of a disease). The focus of this submission is on the output performance of the trained algorithm in terms of image quality metrics.

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

The 510(k) summary does not provide details on how the ground truth was established for the training set of the AiCE Deep Convolutional Neural Network. For deep learning image reconstruction algorithms, the training process often involves:

• Pairs of "noisy" and "clean" image data: The network learns to transform low-dose/noisy inputs into high-quality outputs. The "clean" ground truth might be derived from high-dose scans, or synthetic data generated with known characteristics, or even "ideal" images produced by traditional, more time-consuming reconstruction methods.
• Simulations: Using physics-based models to simulate noise and artifacts, then generating ideal "ground truth" images.
• Existing clinical data: Using a large dataset of patient images, processed or curated to serve as ground truth for noise reduction and image enhancement.

Without explicit information, the specific method used to establish ground truth for AiCE's training set remains undetailed in this FDA submission.

Ask a specific question about this device

Optional movable gantry base unit for use with an Aquilion ONE (TSX-305A) system to support longitudinal movement and allow acquisition of images in the z-direction (z-axis).

Note: When installed with the movable gantry base unit, Aquilion ONE can be used with the INFX-8000C system in the same room.

The Infinix 4DCT is composed of the INFX-8000C interventional angiography system and the dynamic volume CT system, Aquilion ONE, TSX-305A/3. This combination enables patient access and efficient workflow for interventional procedures. Aquilion ONE Self-Propelled Scan Base Kit for IVR-CT, CGBA-034A, is an optional kit intended to be used in conjunction with an Aquilion ONE / INFX-8000C based IVR-CT system. This device is attached to the Aquilion ONE CT gantry to support longitudinal movement and allow image acquisition in the z-direction (Z-axis), both axial and helical. When this option is installed, the standard CT patient couch is replaced with the patient handling system utilized by the interventional x-ray system, INFX-8000C. The Aquilion ONE Self-Propelled Scan Base Kit for IVR-CT, CGBA-034A, will be used as part of an Aquilion ONE / INFX-8000C based IVR-CT system. Please note, the intended uses and technological characteristics of the Aquilion ONE CT System and the INFX-8000C Interventional X-Ray System remain the same. There have been no modifications made to the imaging chains in these FDA cleared devices and the base system software remains the same. Since both systems will be installed in the same room and to prevent interference during use, system interlocks have been incorporated into the systems.

The provided text does not contain detailed information on the acceptance criteria and a study proving the device meets those criteria in a format suitable for the requested table. The document is a 510(k) summary for a medical device, which focuses on demonstrating substantial equivalence to a predicate device rather than presenting a detailed clinical study with specific performance metrics and acceptance thresholds.

However, based on the text, I can infer some aspects and extract relevant information to address your request as much as possible.

Here's an attempt to answer your questions by extracting information from the provided document:

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

The document describes verification/validation testing conducted through bench testing to demonstrate that requirements for modifications have been met. It lists several functional aspects that were evaluated. While it doesn't provide specific quantitative acceptance criteria or numerical performance results in the provided text, it states that these functions "performed according to specifications."

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

The document mentions "bench testing" but does not specify a sample size for the test set or the data provenance (e.g., country of origin, retrospective or prospective). Bench testing typically involves testing a device or its components in a laboratory setting.

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

No information is provided regarding experts used to establish ground truth for a test set. Bench testing for this type of device (a movable gantry base unit) would likely involve technical verification against engineering specifications rather than expert interpretation of images or clinical outcomes.

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

No information is provided regarding an adjudication method. This is typically relevant for studies involving human interpretation or clinical endpoints, which are not detailed here.

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 study was mentioned or performed. This device is a mechanical accessory (movable gantry base unit for a CT scanner), not an AI-powered diagnostic tool, so an MRMC study comparing human reader performance with and without AI assistance is not applicable.

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

No standalone algorithm performance was done/mentioned. As noted, this is a hardware accessory, not an algorithm.

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

Given the nature of the device and the "bench testing" described, the "ground truth" would likely be defined by engineering specifications and functional requirements for mechanical movement, positioning accuracy, safety interlocks, and imaging capabilities (e.g., ability to acquire images in the z-direction). It is not based on clinical "truth" such as pathology or expert consensus on disease.

8. The sample size for the training set:

No training set is mentioned or applicable. This device is a mechanical component, not a machine learning model that requires a training set.

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

Not applicable, as there is no training set for this type of device.

In summary, the provided document (a 510(k) summary) focuses on regulatory submission for a device modification and substantial equivalence. It highlights that "bench testing" was conducted to confirm the device's functional performance against specifications and to ensure safety and effectiveness. However, it does not detail specific quantitative acceptance criteria or results in a table, nor does it conduct the types of studies (like MRMC or AI performance studies) that would involve human readers, ground truth established by experts, or training sets.

Ask a specific question about this device

This device is indicated to acquire and display cross sectional volumes of the whole body, to include the head, with the capability to imaqe whole organs in a single rotation. Whole organs include but are not limited to brain, heart, pancreas, etc.

The Aquilion ONE has the capability to provide volume sets of the entire These volume sets can be used to perform specialized studies, orqan. usinq indicated software/hardware, of the whole orqan by a trained and qualified physician.

FIRST 2.1 is an iterative reconstruction alqorithm intended to reduce exposure dose and improve high contrast spatial resolution for abdomen, pelvis, chest, cardiac, extremities and head applications.

Aquilion ONE (TSX-305A/3) V8.3 with FIRST 2.1 is a whole body multi-slice helical CT scanner, consisting of a gantry, couch and a console used for data processing and display. This device captures cross sectional volume data sets used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician. This system is based upon the technology and materials of previously marketed Toshiba CT systems. In addition, the subject device incorporates the latest iterative reconstruction technology, FIRST 2.1, intended to reduce exposure dose while maintaining and/or improving image quality.

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

1. Table of Acceptance Criteria and Reported Device Performance

The provided document primarily focuses on demonstrating substantial equivalence to a predicate device and discussing improvements. Explicit, quantitative acceptance criteria are not clearly defined as pass/fail thresholds in the same way they might be for a novel device with a specific performance claim. Instead, the "acceptance criteria" are implied by showing the new device performs "substantially equivalent to or demonstrates an improvement" over the predicate device, particularly in terms of image quality and dose reduction.

• Diagnostic quality images for head, chest, abdomen/pelvis, extremity, and cardiac exams.
• Superior LCD performance with FIRST. |
  | Dose Reduction | Demonstrate dose reduction while maintaining/improving image quality. | - Dose reduction claim with a range of 59.2% to 82.4% supported.
• 49.2% noise reduction demonstrated with FIRST.
• Low Contrast Detectability (LCD) comparability at full dose (FBP) and reduced dose (FIRST 2.1) using a non-inferiority analysis with a model observer approach. |
  | suRE Subtraction Angio | Improved visibility of contrast enhancement over predicate device. | - Visual assessment by doctors and clinical case examples determined "the visualization of contrast enhancement, especially blood vessels, are improved when compared to the predicate device." |
  | suRE Subtraction Iodine Mapping | Functions as intended to assess contrast enhancement and provide additional information for abdominal organs. | - Visual assessment of clinical case examples determined "provides additional information for the assessment of abdominal organs, differential enhancement mapping images are useful visualization tools and CE Boost images provide improved CNR." |

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

• Test Set Sample Size:
  • Image Quality Metrics (Phantoms): Not explicitly stated, as it uses phantoms for objective measurements.
  • Quantitative Dose Reduction/Spatial Resolution Evaluations: Not explicitly stated, as it uses phantoms and a model observer approach to simulate human perception.
  • suRE Subtraction Angio & Iodine Mapping (Clinical Evaluation): Refers to "clinical case examples" but does not specify the number of cases.
  • Diagnostic Image Quality: "Representative diagnostic images" were obtained, but the number is not specified.
• Data Provenance: Not specified in terms of country of origin. The studies are described as "bench testing" (phantoms), "model observer studies," and "clinical evaluation/case examples," implying a mix of objective measurements and human review of images. It does not state whether the clinical evaluations were retrospective or prospective, though "clinical case examples" often implies retrospective.

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

• Quantitative Dose Reduction/Spatial Resolution Evaluations (LCD): "A model observer approach which incorporates some aspects of human vision was used." This is not human expert ground truth but an algorithm mimicking human perception.
• suRE Subtraction Angio & Iodine Mapping: "Visual assessment scoring method by doctors" (Angio) and "visual assessment scoring method of clinical case examples" (Iodine Mapping). The number of doctors/experts is not specified. Their qualifications are broadly stated as "doctors" for Angio and not specified beyond "visual assessment" for Iodine Mapping.
• Diagnostic Image Quality: "Representative diagnostic images, reviewed by an American Board Certified Radiologist." The number of radiologists is not specified (could be one or more). The qualification is "American Board Certified Radiologist."

4. Adjudication Method for the Test Set

• Not explicitly stated for the qualitative assessments. For the "visual assessment scoring method by doctors," it's not clear if there was a consensus process (e.g., 2+1, 3+1), or if individual assessments were aggregated.
• For the dose reduction and image quality metrics using phantoms, adjudication is not applicable as it involves objective measurements or a model observer.

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

• No, an MRMC comparative effectiveness study involving human readers comparing AI-assisted vs. non-AI-assisted performance was not explicitly described.
• The document describes comparisons between images reconstructed with FIRST 2.1 (iterative reconstruction, the "AI" component) and Filtered Back Projection (FBP, the older method) in terms of objective metrics and model observer studies for dose reduction/spatial resolution.
• For the suRE Subtraction features, it mentions "visual assessment scoring method by doctors" comparing the new features to the predicate device, but this is not framed as a rigorous MRMC study comparing human performance with and without the AI (FIRST 2.1) or the suRE features' assistance. It focuses on the features providing improved image visibility or additional information.

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

• Yes, a standalone performance assessment was conducted.
• The "Image Quality Evaluation" and "Quantitative Dose Reduction/Spatial Resolution Evaluations" sections primarily describe the standalone performance of the FIRST 2.1 algorithm. This includes:
  • CT image quality metrics (spatial resolution, noise, CNR, etc.) performed using phantoms.
  • The dose reduction and noise reduction claims were established by comparing FIRST 2.1 to FBP using phantom data and a model observer approach, which is an algorithmic assessment of image quality.
  • PUREViSION Optics (LCD and Noise Improvement) was evaluated via model observer studies using phantoms.

7. Type of Ground Truth Used

• Phantom Data: For image quality metrics, dose reduction, and noise reduction evaluations (e.g., spatial resolution, CT number accuracy, low contrast detectability). This is an objective, physical standard.
• Model Observer Approach: Used for low contrast detectability in the dose reduction studies. This is an algorithmic simulation of human perception, not human expert consensus or clinical outcomes.
• Expert Visual Assessment/Clinical Case Examples: For suRE Subtraction Angio and Iodine Mapping, and for overall diagnostic image quality. This relies on medical expertise but is qualitative and subject to inter-observer variability if not rigorously adjudicated.

8. Sample Size for the Training Set

• The document does not provide any information regarding the training set sample size for the FIRST 2.1 iterative reconstruction algorithm. This information is typically proprietary to the manufacturer and not usually included in 510(k) summaries unless specifically relevant to a new and unprecedented AI claim.

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

• Since the training set sample size is not provided, how its ground truth was established is also not described in this document. Iterative reconstruction algorithms like FIRST 2.1 are developed by optimizing parameters based on phantom data, simulated noise models, and sometimes a large dataset of patient images with associated image quality metrics, but the specifics are not detailed here.

Ask a specific question about this device

This device is indicated to acquire and display cross sectional volumes of the whole body, to include the head, with the capability to imaqe whole orqans in a sinqle rotation. Whole organs include but are not limited to brain, heart, pancreas, etc.

The Aquilion ONE has the capability to provide volume sets of the entire orqan. These volume sets can be used to perform specialized studies, usinq indicated software/hardware, of the whole organ by a trained and qualified physician.

FIRST 2.0 is an iterative reconstruction alqorithm intended to reduce exposure dose and improve high contrast spatial resolution for abdomen, pelvis, chest, cardiac and extremities applications.

Aquilion ONE Vision with FIRST 2.0 (CCRS-001B) V7.4 is a whole body multi-slice helical CT scanner, consisting of a gantry, couch and a console used for data processing and display. This device captures cross sectional volume data sets used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician. This system is based upon the technology and materials of previously marketed Toshiba CT systems. In addition, the subject device incorporates the latest iterative reconstruction technology, FIRST 2.0, intended to reduce exposure dose while maintaining and/or improving image quality.

The provided text describes a 510(k) premarket notification for a Computed Tomography (CT) system named Aquilion ONE Vision with FIRST 2.0 (CCRS-001B) V7.4. The documentation focuses on demonstrating substantial equivalence to a predicate device (Aquilion ONE Vision with FIRST 1.0) by updating an iterative reconstruction algorithm (FIRST 2.0).

Based on the provided document, here's a breakdown of the requested information:

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

The document does not explicitly state "acceptance criteria" in a numerical or pass/fail table format for clinical performance. Instead, it describes performance in terms of improvements or equivalency compared to the predicate device and filtered back projection (FBP).

• Contrast-to-noise ratio
• CT number accuracy
• Uniformity
• Slice sensitivity profile
• Modulation transfer function
• Line pair gauge
• Low contrast detectability
• Standard deviation of noise
• Noise power spectra. |
  | Diagnostic Quality | Representative diagnostic images reviewed by an American Board Certified Radiologist demonstrated that the device produces images of diagnostic quality and performs as intended. |

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

• Test Set Sample Size: The document mentions "Representative diagnostic images... including chest, abdomen and pelvis, extremity and cardiac exams." However, it does not specify the number of images or cases used for this review.
• Data Provenance: The document implies the data was collected from the device itself ("obtained using the subject device"). There is no information about the country of origin of the data or whether it was retrospective or prospective. Given it's a premarket submission for a new version of an existing device, it's likely part of internal validation testing.

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

• Number of Experts: "an American Board Certified Radiologist" – indicates one expert was used.
• Qualifications: "American Board Certified Radiologist" – this implies a board certification, which generally requires specific training, residency, and passing board examinations, demonstrating a certain level of expertise in radiology. The document does not specify the years of experience of this radiologist.

4. Adjudication method for the test set

• Since only one radiologist was used for the diagnostic image review, an adjudication method (like 2+1 or 3+1) was not applicable/performed. The single expert's review served as the assessment.

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 study was mentioned or performed as part of this submission. The evaluation was primarily focused on technical image quality metrics and a single radiologist's review of diagnostic image quality, not the impact on human reader performance with or without AI assistance. The "AI" here (FIRST 2.0) is an iterative reconstruction algorithm, which enhances image quality/dose reduction, rather than an AI-driven diagnostic aid that would typically warrant a comparative reader study.

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

• Yes, an algorithm-only (standalone) performance evaluation was done in terms of quantitative image quality metrics (e.g., contrast-to-noise ratio, spatial resolution, dose reduction claims) using phantoms.
• "A model observer evaluation" directly assessed the algorithm's performance in achieving "equivalent low contrast detectability" with reduced dose, which is a standalone assessment of the algorithm's output.

7. The type of ground truth used

• For quantitative image quality metrics and dose reduction claims: Phantoms (e.g., MITA-FDA phantom) were used to provide a known, controlled ground truth.
• For diagnostic image quality review: The "ground truth" was established by the expert opinion of an American Board Certified Radiologist who reviewed representative images for "diagnostic quality." This is essentially expert consensus (albeit from a single expert).

8. The sample size for the training set

• The document does not provide any information about the sample size used for training the FIRST 2.0 iterative reconstruction algorithm. This information is typically proprietary to the manufacturer and not required in this level of 510(k) summary.

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

• The document does not provide any information on how the ground truth for the training set (if any, as iterative reconstruction algorithms might use phantoms or specific patient data for training their models without needing human-labeled "ground truth" for diagnosis) was established. This detail is also not typically shared in a 510(k) summary.

Ask a specific question about this device