Search Results

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

The Aquilion RXL 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 TSX-101A/R is a whole body CT scanner. This device captures helical volumetric data sets. The device consists of a gantry, patient couch (table) and peripheral cabinets used for data processing and display.

The provided text describes a 510(k) premarket notification for a CT scanner (Aquilion RXL; TSX-101A) by Toshiba America Medical Systems, Inc. The submission is for a modification to a cleared device (TSX-301A/2 - K093891), specifically to reduce radiation exposure to the patient using an iterative reconstruction algorithm (AIDR algorithm).

However, the document does not provide specific acceptance criteria, detailed study results, or the other requested granular information (sample sizes, ground truth establishment, expert qualifications, adjudication methods, MRMC studies, standalone performance, or training set details) that would typically be found in a comprehensive clinical study report. The submission focuses on demonstrating substantial equivalence based on the device's design, manufacturing controls, and general performance testing using phantoms, rather than detailed clinical validation with specific endpoints.

Here's a breakdown of what can be extracted from the document, and what is missing:

1. Table of Acceptance Criteria and Reported Device Performance

Study Proving Acceptance Criteria:

The document states: "Testing was conducted utilizing phantoms and accepted image quality metrics. The results of this testing is contained in the user information for the device. Additional testing was conducted at numerous beta sites to provide clinical data for the validation of the software change."

This implies that the "study" involved:

• Phantom Testing: To verify image quality metrics with the new iterative reconstruction algorithm.
• Beta Site Clinical Data: To validate the software change in a clinical setting, presumably to confirm that the lower-dose scans with AIDR produce clinically acceptable images.

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

• Sample Size for Test Set: Not specified. The document mentions "numerous beta sites" for clinical data, but no specific number of patients or scans.
• Data Provenance: The "clinical data" was gathered at "numerous beta sites," which suggests a prospective collection for validation, but the country of origin is not specified. The submitting company, Toshiba America Medical Systems, Inc., is based in Tustin, CA, USA, which might suggest US-based beta sites, but this is not explicitly stated.

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

• Not specified. The document mentions that the volume sets "can be used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician." This refers to the general use of the device, not specifically the establishment of ground truth for the validation study of the AIDR algorithm. There is no information on how many experts, or their qualifications, were involved in assessing the clinical data from the beta sites.

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

• Not specified. The document does not provide any details on adjudication methods for the clinical data collected.

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

• Not specified and unlikely. The AIDR algorithm is an iterative reconstruction technique, which is a core image processing function of the CT scanner, not typically an AI-assisted diagnostic tool that would directly lead to "human readers improve with AI vs without AI assistance" in the way an MRMC study evaluates. The focus here is on achieving image quality comparable to higher-dose scans. The document does not mention any MRMC study or effect sizes of human reader improvement.

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

• Implicitly, yes for image quality metrics. The "testing was conducted utilizing phantoms and accepted image quality metrics" would represent a standalone performance evaluation of the algorithm's effect on image characteristics (e.g., noise, spatial resolution, contrast-to-noise ratio) without human interpretation initially. However, the exact metrics and results are not detailed. The beta site data would then involve human observers (physicians) assessing clinical utility.

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

• Not explicitly stated. For the clinical validation at beta sites, the "ground truth" would likely be the clinical assessment by trained and qualified physicians of the diagnostic utility of the lower-dose images generated with AIDR compared to standard-dose images (or the previous version of the device). This implicitly relies on clinical judgment and potentially comparison with the previous standard, but no specific external ground truth (like pathology or outcomes) is mentioned.

8. The sample size for the training set

• Not applicable/Not specified. AIDR is described as an "iterative reconstruction algorithm." While such algorithms can sometimes be machine learning-based, the document doesn't explicitly state it's a "deep learning" or AI model that requires a distinct "training set" in the modern sense. If it did involve such a training set, the sample size is not specified. It's more likely to be a sophisticated mathematical reconstruction technique rather than a data-trained AI model in this context (given the 2012 submission date).

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

• Not applicable/Not specified. As above, a "training set" and its "ground truth" establishment are not discussed in the context of this iterative reconstruction algorithm.

Ask a specific question about this device