Search Results

This device is indicated to acquire and display cross-sectional volumes of the whole body, to include the head.

The Aquilion Lightning has the capability to provide volume sets. These volume sets can be used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician.

The Aquilion Lightning, TSX-036A/1, v8.4 is an 80-row CT System that is intended to acquire and display cross-sectional volumes of the whole body, including the head. This system is based upon the technology and materials of previously marketed Toshiba CT systems.

The provided text describes a 510(k) submission for the Aquilion Lightning, TSX-036A/1, V8.4 CT system. As such, it focuses on demonstrating substantial equivalence to a predicate device rather than presenting a detailed study with specific acceptance criteria and performance metrics for a novel AI-powered diagnostic device.

Therefore, much of the requested information, particularly regarding acceptance criteria for diagnostic performance, sample sizes for test sets in an AI context, expert ground truth establishment, MRMC studies, and standalone AI performance, is not present in this document because it describes a computed tomography x-ray system, not an AI software.

However, I can extract information related to the device's technical specifications and how its performance was assessed for regulatory clearance.

Here's a breakdown of the available information based on your request:

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

The document doesn't define specific quantitative "acceptance criteria" in the typical sense of a diagnostic performance study (e.g., sensitivity, specificity thresholds). Instead, it states that the device was evaluated against performance metrics relevant to CT image quality and found to be "substantially equivalent" to the predicate.

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

• Test Set Sample Size: Not specified in terms of a patient cohort. The testing involved "representative diagnostic images" and phantom studies.
• Data Provenance: Not explicitly stated but implies images were generated by the device itself and likely from standard clinical scenarios (retrospective or prospective is not specified).

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

• Number of Experts: One.
• Qualifications of Experts: An "American Board Certified Radiologist."

4. Adjudication method for the test set:

Not applicable/specified. The document states a single American Board Certified Radiologist reviewed representative diagnostic images. There is no mention of a multi-reader adjudication process for establishing ground truth for a test set in the context of diagnostic performance.

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. This document does not mention an MRMC comparative effectiveness study, nor does it discuss AI assistance for human readers. This device is an imaging system, not an AI diagnostic tool.

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

No. This is a CT imaging system. The performance assessment relates to the image acquisition and display capabilities, not a standalone algorithm.

7. The type of ground truth used:

• For CT Image Quality metrics (phantom studies): The ground truth is the physical properties of the phantoms and established CT physics principles for measuring image quality.
• For diagnostic image quality: Expert opinion of an American Board Certified Radiologist ("produces images of diagnostic quality").

8. The sample size for the training set:

Not applicable. This document describes a CT scanner, not an AI algorithm that requires a training set in the typical sense. The "training" of the system involves its design, manufacturing under quality systems, and adherence to engineering specifications.

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

Not applicable. (As above, not an AI algorithm with a training set).

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 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.

Aquilion ONE (TSX-305A/3) V7.3 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.

Here's an analysis of the provided text regarding the acceptance criteria and study for the Aquilion ONE (TSX-305A/3) V7.3:

1. Table of Acceptance Criteria and Reported Device Performance:

The document is a 510(k) summary for a premarket notification for a Computed Tomography X-ray System. It is not a clinical study report with specific acceptance criteria directly tied to a diagnostic performance metric (like sensitivity or specificity) of a disease-detecting AI algorithm. Instead, it demonstrates substantial equivalence to a predicate device, focusing on technical specifications and image quality for general diagnostic use.

Therefore, the "acceptance criteria" here relate to demonstrating that the new device performs acceptably for its intended use and is equivalent to the predicate. The "performance" is primarily a comparison of technical specifications and image quality metrics against the predicate.

• View rate: Maximum 2910 views/s (same)
• Detector: 896 channels x 320 rows (same)
• Pitch factor: Range 0.555 to 1.575 / 0.555 to 1.5 (very similar)
• FOV: 240/320/500mm / 180/240/320/400/500mm (subject has slightly reduced range, but still within typical diagnostic needs)
• Metal artifact reduction: SEMAR (Volume, Helical, ECG gated) / SEMAR (Volume, Helical) (subject has added ECG gated capability)
• Gantry opening size: 780 mm (same)
• All previously cleared software options are listed as "no change" in functionality, with some having "workflow improvements" (e.g., Lung Volume Analysis, surESubtraction Lung, MyoPerfusion, Dual Energy System Package, 4D Airways Analysis) which are enhancements rather than regressions.

Differences (addressed through testing or not raising new concerns):

• Gantry Rotation Speed: 0.35s (Optional max 0.275s) for subject vs. 0.275s (Standard or optional) for predicate. This indicates a minor hardware difference, likely addressed by showing image quality is maintained.
• Wedge filter types: Two types for subject vs. Three for predicate. This is a minor design change.
• X-ray tube voltage/current: Max 72kW (Optional Max 90kW) for subject vs. Max 90kW (for one model) or Max 72kW (for others) for predicate. Comparable.
• Image reconstruction time: Up to 80 images/s for subject vs. Up to 50 images/s for predicate. Improvement in subject device.
• Helical reconstruction method: 20 rows or more: TCOT+ for subject vs. 16 rows or more: TCOT+ for predicate. Improvement in subject device (more rows).
• Patient Couch Type and related dimensions/weights: Various configurations/differences between subject and predicate models, indicating design variations but within expected functional range.
• Gantry tilt angle: ±30° for subject vs. ±22° for predicate. Improvement in subject device.
• Minimum area for installation: Smaller for subject (27m² vs 37.2m²). Improvement in subject device.
• Area finder: Optional for subject vs. NA for predicate. New feature on subject device. |
  | Image Quality | Image quality metrics (spatial resolution, CT number magnitude/uniformity, noise properties, low contrast detectability/CNR performance) should meet established specifications and be comparable to the predicate device. Images obtained should be of diagnostic quality. | CT image quality metrics performed using phantoms demonstrated that the subject device is substantially equivalent to the predicate device with regard to: spatial resolution, CT number magnitude/uniformity, noise properties, and low contrast detectability/CNR performance. Representative diagnostic images (head, chest, abdomen/pelvis, extremity, cardiac) were also reviewed and demonstrated diagnostic quality. |
  | Safety and Standards Compliance | The device must be designed and manufactured under Quality System Regulations (21 CFR 820, ISO 13485) and conform to applicable performance standards for ionizing radiation-emitting products (21 CFR, Subchapter J, Part 1020). It must also comply with various IEC, NEMA, and other relevant standards. | The device is designed and manufactured under QSR and ISO 13485. It conforms to applicable performance standards for Ionizing Radiation Emitting Products [21 CFR, Subchapter J, Part 1020] and numerous international standards including IEC60601-1 series, IEC60601-2 series, IEC60825-1, IEC62304, IEC62366, NEMA PS 3.1-3.18, NEMA XR-25 and NEMA XR-26. |
  | Software Validation | Software documentation must comply with FDA guidance for a Moderate Level of Concern, and validation testing should be successfully completed. | Software Documentation for a Moderate Level of Concern was included. Successful completion of software validation is cited in the conclusion. |
  | Risk Management | Risk analysis should be conducted. | Risk analysis was conducted. |

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

• Test Set Description: The "test set" for this submission primarily consists of:
  • Phantoms: Used for evaluating CT image quality metrics (spatial resolution, CT number, noise, low contrast detectability). The number and specific types of phantoms are not explicitly stated but are typically standard phantoms used in CT performance testing.
  • Representative Diagnostic Images: Clinical images covering various body regions (head, chest, abdomen/pelvis, extremity, cardiac). The number of cases/patients is not specified.
• Data Provenance: The document does not explicitly state the country of origin for the diagnostic images. Given Toshiba Medical Systems Corporation is based in Japan and Toshiba America Medical Systems, Inc. is in the US, the data could originate from either region or a combination. The document also does not specify if the data was retrospective or prospective. However, for a 510(k) clearance based on substantial equivalence, particularly for a hardware/software update to a CT scanner, diagnostic images are often retrospectively collected or acquired as part of internal validation.

3. Number of Experts Used to Establish Ground Truth and Qualifications:

• Number of Experts: One (1) expert is explicitly mentioned.
• Qualifications of Experts: An "American Board Certified Radiologist." No specific years of experience are stated. This expert reviewed the representative diagnostic images to confirm diagnostic quality.

4. Adjudication Method for the Test Set:

• The document describes a single American Board Certified Radiologist reviewing images to confirm diagnostic quality. This indicates no formal adjudication method involving multiple readers (like 2+1 or 3+1) was used for this specific part of the evaluation. The assessment of image quality from phantoms would not typically involve expert adjudication.

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

• No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done. This document is for a general-purpose CT scanner system, not an AI-specific diagnostic tool that assists human readers. Therefore, there is no mention of an effect size for human reader improvement with or without AI assistance.

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

• No, a standalone performance study in the context of an AI algorithm was not done. The Aquilion ONE (TSX-305A/3) V7.3 is a complete CT system where the "algorithm" refers to the image reconstruction and processing capabilities, which are inherent to the device's function. The study validates the overall system's ability to produce diagnostic images, not a separate AI algorithm's diagnostic accuracy. The performance is assessed on the system output.

7. The Type of Ground Truth Used:

• For the phantom studies, the "ground truth" is typically known physical properties of the phantoms (e.g., known dimensions, densities, contrast levels).
• For the representative diagnostic images, the "ground truth" for confirming "diagnostic quality" is based on the expert opinion/consensus of an American Board Certified Radiologist. This is a form of expert consensus, albeit from a single expert in this stated context. There is no mention of pathology or outcomes data being used as ground truth for this submission.

8. The Sample Size for the Training Set:

• The document does not specify a separate "training set" sample size. This submission is for a medical imaging device (CT scanner) rather than a deep learning AI algorithm that undergoes distinct training. The underlying algorithms for image reconstruction and processing (e.g., TCOT+, SEMAR) are developed and refined through engineering and iterative testing, but not typically in the same "training set" paradigm as AI for diagnostic interpretation. The software validation is mentioned, which refers to standard software development lifecycle testing.

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

• As a "training set" in the context of AI development is not explicitly mentioned as relevant to this submission, the establishment of ground truth for a training set is not applicable/described. The "ground truth" during the development of a CT scanner's image reconstruction algorithms would typically involve engineering specifications, physical models, and potentially early clinical data used for empirical tuning and validation, but not a formally labeled training set in the AI sense.

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 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.

The Aquilion ONE Vision, TSX-301C/3 and 301C/4, v6.00 are 320-row CT Systems and the TSX-301C/5, v6.00 is a 160-row CT system consisting of the same gantry, couch and console used for data processing and display. These devices capture cross sectional volume data sets used to perform specialized studies, using indicated software, by a trained and qualified physician. These systems are based upon the technology and materials of previously marketed Toshiba CT systems.

The provided 510(k) summary focuses on demonstrating substantial equivalence of a modified CT system (Aquilion ONE Vision, TSX-301C/3, 301C/4 and 301C/5, v6.00) to its predicate device. This submission is for a hardware and associated software modification to an existing CT scanner, not a new AI/CADe device. Therefore, many of the typical acceptance criteria and study elements pertinent to AI systems that you requested are not directly applicable or explicitly detailed in this document.

However, I can extract the information that is present and explain why some of your requested details might not be found in this type of submission.

1. Table of Acceptance Criteria and Reported Device Performance

For this type of device modification, the "acceptance criteria" are generally related to demonstrating that the modified device performs at least as well as the predicate device in terms of image quality and safety, and continues to meet relevant standards. The performance is assessed through various tests, primarily utilizing phantoms and review by an expert.

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

• Sample Size for Test Set: The document mentions "representative diagnostic images" but does not specify a numerical sample size for the clinical image review. This type of submission often relies on a qualitative assessment of a small, representative set of images rather than a large statistical study.
• Data Provenance: Not explicitly stated (e.g., country of origin). Since it's a Toshiba America Medical Systems submission, the testing would likely have occurred in the US or Japan. The assessment of diagnostic images by an "American Board Certified Radiologist" suggests at least some of the data review, if not acquisition, was US-based.
• Retrospective or Prospective: Not specified. Given the nature of a device modification test, it could involve prospective acquisition of new images for evaluation, or retrospective review of images acquired with the modified device.

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" (singular).
• Qualifications: "American Board Certified Radiologist." The duration of experience is not specified.

4. Adjudication method for the test set

• Adjudication Method: Not applicable or specified. With only one radiologist reviewing, there is no inter-reader discrepancy to adjudicate.

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

• MRMC Study: No, an MRMC comparative effectiveness study was not done. This device is a CT scanner itself, not an AI/CADe accessory intended to assist human readers in image interpretation. Therefore, assessing human reader improvement with/without "AI assistance" is not relevant to this submission.

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

• Standalone Performance: Not applicable. This is a conventional CT imaging system, not an algorithm being evaluated for standalone performance. The "softwar" mentioned refers to control software and image processing pathways within the CT system, not an independent AI diagnostic algorithm.

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

• Type of Ground Truth: For the image quality assessment, the "ground truth" was the qualitative judgment of an "American Board Certified Radiologist" that the images were of "diagnostic quality" and that the device "performs as intended." For the phantom studies, the ground truth is against known physical properties and measurements within the phantoms, assessed by quantitative metrics (spatial resolution, CT number, CNR, noise, uniformity).

8. The sample size for the training set

• Sample Size for Training Set: Not applicable/not provided. This submission is for a hardware and control software modification to a CT scanner. The concept of a "training set" in the context of machine learning (AI) does not apply here. The system's performance is based on engineered design and physical principles, not on being trained on a dataset of images with ground truth labels.

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

• How Ground Truth for Training Set was Established: Not applicable, as there is no "training set" in the AI sense for this device. The system's operational parameters and calibration are established through design specifications, factory calibration, and quality control processes.

In summary: This 510(k) submission is for a modification to a general-purpose CT imaging system. The performance evaluation focuses on demonstrating that the modified hardware and software maintain or improve the fundamental imaging capabilities and safety profiles compared to the predicate device, primarily through phantom testing and qualitative clinical image review by a radiologist. It does not involve the a-typical AI/CADe specific study designs and ground truth methodologies you would expect for an AI-powered diagnostic tool.

Ask a specific question about this device