Search Results
Found 3 results
510(k) Data Aggregation
(245 days)
Deep Recon is a data driven image reconstruction method based on deep learning technology. It is intended to produce cross-sectional images by computer reconstruction of X-ray transmission data taken at different angles planes, including Axial, Helical, and Cardiac acquisition.
Deep Recon is designed to generate CT images with lower image noise, and improved low contrast detectability, and it can reduce the dose required for diagnostic CT imaging.
Deep Recon can be used for head, chest, abdomen, cardiac and vascular CT applications for adults. Deep Recon is intended to be used with uCT 760 and uCT 780 only.
The Deep Recon is a data driven image reconstruction method based on deep learning technology. Dedicated deep neural networks are designed and trained for different body parts. As a part of reconstruction chain, the Deep Recon generates CT images with an appearance similar to traditional FBP, but with a decreased image noise, and an improved low contrast detectability. The Deep Recon was specifically trained on uCT 760 and uCT 780 (K172135). The function is integrated on the mentioned CT systems as a part of reconstruction chain.
The initial document provides a 510(k) summary for the Deep Recon device, a data-driven image reconstruction method based on deep learning technology for CT systems. The device is intended to produce cross-sectional images with lower image noise, improved low contrast detectability, and the ability to reduce the required dose for diagnostic CT imaging.
Here's an analysis of the acceptance criteria and study information provided:
1. Table of Acceptance Criteria and Reported Device Performance
The document does not explicitly state numerical acceptance criteria in a dedicated table. Instead, it describes performance goals as "equivalent or better performance" compared to Filtered Back Projection (FBP) for various image quality metrics, and "equivalent or better" diagnostic quality in clinical evaluations.
| Feature/Metric | Acceptance Criteria (Implied) | Reported Device Performance (Deep Recon vs. FBP) |
|---|---|---|
| Low Contrast Detectability (LCD) | Equivalent or better than FBP | Improved compared to FBP |
| Image Noise | Equivalent or better than FBP | Decreased compared to FBP |
| Mean CT Number | Equivalent to FBP | Equivalent to FBP |
| Uniformity | Equivalent to FBP | Equivalent to FBP |
| Spatial Resolution | Equivalent to FBP | Equivalent to FBP |
| Reconstructed Section Thickness | Equivalent to FBP | Equivalent to FBP |
| Diagnostic Quality (Reader Study 1) | Equivalent or better than FBP | Equivalent or better than FBP in diagnostic quality |
| Diagnostic Quality (Reader Study 2) | Low-dose Deep Recon equivalent to standard-dose FBP | Low-dose images with Deep Recon are equivalent or better than standard-dose images with FBP in diagnostic quality |
2. Sample Size and Data Provenance for Test Set
- Clinical Image Evaluation (Reader Study 1):
- Sample Size: 80 retrospectively collected clinical cases.
- Data Provenance: Retrospective, country of origin not specified, but the device manufacturer is based in Shanghai, China.
- Clinical Image Evaluation (Reader Study 2):
- Sample Size: 40 retrospectively collected clinical cases (20 low dose, 20 standard dose).
- Data Provenance: Retrospective, country of origin not specified, but the device manufacturer is based in Shanghai, China.
3. Number of Experts and Qualifications for Ground Truth Establishment (Test Set)
- Clinical Image Evaluation (Reader Study 1):
- Number of Experts: 2 board-certified radiologists.
- Qualifications: "board-certified radiologists." No further details on years of experience are provided.
- Clinical Image Evaluation (Reader Study 2):
- Number of Experts: "a board-certified radiologist." This implies only one radiologist was used.
- Qualifications: "board-certified radiologist." No further details on years of experience are provided.
4. Adjudication Method for the Test Set
- The document describes that for Reader Study 1, "Each image was read by 2 board-certified radiologists." It does not specify an adjudication method like 2+1 or 3+1 for discrepancies.
- For Reader Study 2, "Each image was read by a board-certified radiologist," indicating no inter-reader adjudication was performed.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- No explicit MRMC comparative effectiveness study is mentioned that quantifies the "effect size of how much human readers improve with AI vs without AI assistance." The studies performed are comparative evaluations of the image quality and diagnostic usefulness of Deep Recon images versus FBP images, as assessed by human readers. They do not describe a scenario where AI assists human readers and measures the improvement.
6. Standalone Performance
- No explicit standalone performance study (algorithm only without human-in-the-loop performance) is described in terms of diagnostic accuracy. The document focuses on the output of the algorithm (the reconstructed images) and how those images are perceived by human readers. The phantom studies (bench testing) could be considered standalone in the sense that they evaluate the algorithm's direct image quality metrics, but not diagnostic performance.
7. Type of Ground Truth Used
- Clinical Image Evaluation Studies: The "ground truth" for the clinical evaluations (Reader Studies 1 & 2) was the expert consensus of the radiologists regarding image noise, structure fidelity, image quality, and clinical features based on a 4-point or 5-point scale. It does not refer to pathology, patient outcomes data, or an independent gold standard for diagnosis.
- Bench Testing: The ground truth for bench testing (LCD, image noise, CT number, uniformity, spatial resolution, section thickness) involved standard phantom measurements and model observers, which represent established physical metrics rather than clinical ground truth derived from patients.
8. Sample Size for the Training Set
- The document states that the Deep Recon's "Dedicated deep neural networks are designed and trained for different body parts." It also mentions that the "Deep Recon was specifically trained on uCT 760 and uCT 780 (K172135)."
- However, the specific sample size (number of images or cases) used for the training set is not provided in this document.
9. How Ground Truth for the Training Set Was Established
- The document mentions that the DNN is "trained on low dose FBP images to get normal dose (high quality) FBP images." This implies that the training likely used pairs of low-dose FBP images (as input to the network) and corresponding "normal dose (high quality) FBP images" (as the target or ground truth for the network to learn from).
- The methodology for establishing the "normal dose (high quality) FBP images" as ground truth for training is not explicitly detailed. It's implicitly assumed that these are standard-of-care FBP reconstructions from standard-dose acquisitions.
Ask a specific question about this device
(53 days)
The uCT Computed Tomography X-ray System uCT530/550 is a computed tomography X-ray system intended to produce cross-sectional images of the body by computer reconstruction of X-ray transmission data taken at different angles and planes and indicated for the whole body, including head, neck, cardiac (calcium scoring) and vascular.
The uCT 530/uCT 550 is a multi-slice X-ray computed tomography scanner which features a continuously rotating tube-detector system and functions according to the fan beam principle. The system provides the filter back-projection (FBP) algorithm to reconstruct images in DICOM format, which can be used by post-processing applications. The system consists of the Gantry, X-ray System, Data Management System, Patient Table, Console, Power Supply Cabinet, Image Processing Computer, and Software. The system software is a program used for patient management, data management, X-ray scan control, image reconstruction, and image archive. A motorized patient table moves the patient through a circular opening in the Gantry. As the patient passes through the Gantry, a source of x rays rotates around the inside of the circular opening. Detectors on the exit side of the patient record the X rays exiting the section of the patient's body being irradiated as an X-ray "snapshot". Many different "snapshots" (angles) are collected during one complete rotation. The data are sent to a computer to reconstruct all of the individual "snapshots" into a cross-sectional image (slice) of the internal organs and tissues for each complete rotation of the source of x rays. There are two features for denoising and reduce metal artifact, which are KARL iterative denoising reconstruction algorithm and MAC Metal artifact correction algorithm.
The provided text does not contain the detailed acceptance criteria or a study proving the device meets specific performance criteria, especially for the newly added "cardiac (calcium scoring)" indication. The document is a 510(k) summary for a Computed Tomography X-ray System (uCT 530, uCT 550) seeking clearance for a modification that adds cardiac calcium scoring functionality.
Here's a breakdown of what is stated regarding performance and what is missing for the specific request:
What is provided:
- Device Name: uCT 530, uCT 550
- New Indication for Use: Cardiac (calcium scoring)
- Justification for new indication: ECG gating tests and sample clinical images evaluation of calcium scoring scan showed the proposed device can obtain clinically acceptable calcium scoring images.
- Performance Verification Tests conducted (general):
- Clinical Evaluation for sample clinical images evaluation
- AEC Test Report for AEC performance study
- MAC Performance Evaluation Report (Metal Artifact Correction)
- ECG R-Wave Detection Algorithm Performance Evaluation Report
- Standards Conformance: A wide range of electrical safety, EMC, product, software, and biocompatibility standards were conformed to.
- General assertion of equivalence: "The features described in this premarket submission are supported with the results of the testing mentioned above, the uCT 530/uCT 550 was found to have a safety and effectiveness profile that is similar to the predicate device."
- No Clinical Study: The document explicitly states "No Clinical Study is included in this submission."
What is missing from the document to answer your specific questions:
The document does not provide any specific quantitative acceptance criteria or detailed results of the studies for the new cardiac (calcium scoring) feature. It merely states that "ECG gating tests and sample clinical images evaluation of calcium scoring scan showed the proposed device can obtain clinically acceptable calcium scoring images." This is a general statement of positive outcome, not a detailed report of criteria or performance.
Therefore, for almost all your specific questions (1-7, and details for 8 and 9), the information is not present in the provided text.
Based on the provided text, I can only provide the following:
Acceptance Criteria and Study for uCT 530/550 (Cardiac Calcium Scoring)
1. Table of Acceptance Criteria and Reported Device Performance:
| Acceptance Criteria (Not explicitly stated in text for calcium scoring) | Reported Device Performance (Summary from text for calcium scoring) |
|---|---|
| No specific quantitative criteria are provided in the document for the calcium scoring feature. For example, it does not specify a target accuracy for calcium score measurement, a minimum detectability of calcifications, or a maximum acceptable motion artifact level. | "the proposed device can obtain clinically acceptable |
| calcium scoring images." |
2. Sample size used for the test set and the data provenance:
- Sample Size: Not specified. The document mentions "sample clinical images evaluation" but does not give a number of images or patients.
- Data Provenance: Not specified (e.g., country of origin, retrospective/prospective).
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Number of Experts: Not specified.
- Qualifications of Experts: Not specified.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- Adjudication Method: Not specified. The phrase "clinical images evaluation" implies assessment by experts, but the method (e.g., consensus, independent reads) is not detailed.
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. The document explicitly states "No Clinical Study is included in this submission." The evaluation appears to be a review of image quality rather than a human-in-the-loop study with human readers.
- Effect Size: Not applicable, as no such study was conducted or reported.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- The document implies the system's ability to produce "clinically acceptable calcium scoring images" based on "ECG gating tests and sample clinical images evaluation." This suggests an assessment of the system's output, which can be considered a form of standalone performance evaluation for the imaging capability.
- The ECG R-Wave Detection Algorithm Performance Evaluation Report would be a standalone evaluation of a component algorithm. However, the details of its performance or acceptance criteria are not provided.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Type of Ground Truth: The ground truth for "clinically acceptable" images would implicitly be based on expert subjective evaluation of the images. No objective ground truth (e.g., quantitative calcium scores from a reference standard, pathology confirmed calcifications, or patient outcomes) is mentioned.
8. The sample size for the training set:
- The document does not describe the training set for any algorithms, as it focuses on the device's substantial equivalence to predicates and its modified features. The mention of "KARL iterative denoising reconstruction algorithm" and "MAC Metal artifact correction algorithm" implies underlying algorithms, but their training data specifics are not provided.
9. How the ground truth for the training set was established:
- Not applicable, as details on training sets for any algorithms are not provided.
Ask a specific question about this device
(76 days)
The uCT Computed Tomography X-ray System uCT 530/550 is a computed tomography x-ray system intended to produce cross-sectional images of the body by computer reconstruction of x-ray transmission data taken at different angles and planes and indicated for the whole body (including head, neck, vascular).
The uCT 530/uCT 550 is a multi-slice X-ray computed tomography scanner which features a continuously rotating tube-detector system and functions according to the fan beam principle. The system provides the filter back-projection (FBP) algorithm to reconstruct images in DICOM format. which can be used by post-processing applications.
The system consists of the Gantry, X-ray System, Data Management System, Patient Table, Console, Power Supply Cabinet, Image Processing Computer, and Software. The system software is a program used for patient management, data management, Xray scan control, image reconstruction, and image archive.
A motorized patient table moves the patient through a circular opening in the Gantry. As the patient passes through the Gantry, a source of x rays rotates around the inside of the circular opening. Detectors on the exit side of the patient record the X rays exiting the section of the patient's body being irradiated as an X-ray "snapshot". Many different "snapshots" (angles) are collected during one complete rotation. The data are sent to a computer to reconstruct all of the individual "snapshots" into a crosssectional image (slice) of the internal organs and tissues for each complete rotation of the source of x rays.
There are two features for denoising and reduce metal artifact, which are KARL iterative denoising reconstruction algorithm and MAC Metal artifact correction algorithm.
The provided text does not contain acceptance criteria or a detailed study that proves the device meets specific acceptance criteria in the format requested.
Instead, the document is a 510(k) summary for a Computed Tomography X-ray System (uCT 530, uCT 550) seeking FDA clearance based on substantial equivalence to a predicate device (uCT 760, uCT 780).
Here's what can be extracted and inferred from the text regarding performance, though it doesn't meet all your requested points:
1. Table of Acceptance Criteria and Reported Device Performance
The document provides a comparison table of technological characteristics between the proposed device (uCT 530, uCT 550) and a predicate device (uCT 760, uCT 780). While not explicitly "acceptance criteria" against numerical targets, it implies that comparable performance to the predicate is the acceptance criterion for substantial equivalence.
| ITEM | Acceptance Criteria (Implicit: Similar to Predicate) | Reported Device Performance (uCT 530, uCT 550) | Notes/Differences from Predicate |
|---|---|---|---|
| Intended Use | The uCT Computed Tomography X-ray System is a computed tomography x-ray system intended to produce cross-sectional images of the body by computer reconstruction of x-ray transmission data taken at different angles and planes and indicated for the whole body (including head, neck, cardiac and vascular). | The uCT Computed Tomography X-ray System uCT530/uCT550 is a computed tomography x-ray system intended to produce cross-sectional images of the body by computer reconstruction of X-ray transmission data taken at different angles and planes and indicated for the whole body (including head, neck, vascular). | Note 1: Does not include cardiac, but states this does not affect safety and effectiveness. |
| Product Code | JAK | JAK | Same |
| Regulation No. | 21 CFR 892.1750 | 21 CFR 892.1750 | Same |
| Class | II | II | Same |
| Scan Regime | Continuous Rotation | Continuous Rotation | Same |
| Scan Modes | Scout, Axial Scan, Helical Scan | Scout, Axial Scan, Helical Scan | Same |
| Detector Material | Solid-state GOS | Solid-state GOS | Same |
| Z-plane coverage | 40mm | 22mm | Note 2: Shorter coverage induces longer scanning time, but does not affect safety and effectiveness. |
| Size of detector element in Z-plane | 0.5mm | 0.55mm | Note 3: Bigger minimum element size induces lower spatial resolution, but does not affect safety and effectiveness. |
| Number of element per row | 936 | 864 | Note 4: Smaller number of elements induces less data sampling but provides sufficient data for reconstruction, not affecting safety and effectiveness. |
| Number of detector row | 80 | 40 | Note 5: Determined by Z-plane coverage and element size, not affecting safety and effectiveness. |
| Maximum slices generated per rotation | 128 for uCT 760, 160 for uCT 780 | 40 for uCT 530, 80 for uCT 550 | Note 6: Smaller slice number induces longer scanning time, but does not affect safety and effectiveness. |
| Minimum slice thickness | 0.625mm for uCT 760, 0.5mm for uCT 780 | 0.55mm | Note 7: Determined by detector element size, not affecting safety and effectiveness. |
| Maximum sampling rate | Up to 4800 views per 360° | Up to 4800 views per 360° | Same |
| Tube anode storage capacity | 7.5MHU | 5.3MHU | Note 8: Lower capacity affects continuous scans' throughput, but not independent scans' effect. |
| Maximum cooling rate | 1386 kHU/min | 815 kHU/min | Note 9: Lower rate affects continuous scans' throughput, but not independent scans' effect. |
| Focal spot size | 0.7x0.7mm, 1.0x1.0mm | 0.5x1.0mm, 1.0x1.0mm | Note 10: Smaller size helps resolution, but image spatial resolution is "equivalent substantially." |
| Power | 80kW for uCT 760, 100 kW for uCT 780 | 50kW | Note 11: Smaller power output induces lower ability of x-ray penetration for high BMI, but safety evaluated. |
| mA Range | 6-667mA for uCT 760, 6-833mA for uCT 780 | 10-420mA | Note 12: Smaller mA output induces lower ability of x-ray penetration for high BMI, but safety evaluated. |
| kV Settings | 70, 80, 100, 120, 140 | 70, 80, 100, 120, 140 | Same |
| Aperture | 700mm | 700mm | Same |
| Rotation speed | Up to 0.35 sec per 360° rotation for uCT 760, Up to 0.3 sec per 360° rotation for uCT 780 | Up to 0.5 sec per 360° rotation | Note 13: Slower rotation speed induces longer scan time, but does not affect safety and effectiveness. |
| Gantry Tilt | ± 30° with 0.5 increment | ± 30° with 0.5 increment | Same |
| Scannable range | 1700 mm | 1700 mm | Same |
| Horizontal motion range | 2180 mm | 2180 mm | Same |
| Table Horizontal Speed | Up to 200mm/sec | Up to 200mm/sec | Same |
| Vertical motion range | 480 mm-950 mm from the floor | 480 mm-950 mm from the floor | Same |
| Vertical speed | Up to 40 mm/sec | Up to 40 mm/sec | Same |
| Table Horizontal Position accuracy | ± 0.25mm | ± 0.25mm | Same |
| Table Maximum table load | 205kg | 205kg | Same |
| Image Spatial Resolution | High mode: >20 lp/cm @ MTF 0%, 16.5 ± 1.7 lp/cm @ MTF10%, 11.5 ± 1.2 lp/cm @ MTF50% | High mode: >20 lp/cm @ MTF 0%, 16.5 ± 1.7 lp/cm @ MTF10%, 11.5 ± 1.2 lp/cm @ MTF50% | Same |
| Image Noise | 3.0 ± 0.5 HU at 120 kV, 5 mm slice thickness, CTDIvol 29.1mGy | 3.0 ± 0.5 HU at 120 kV, 5 mm slice thickness, CTDIvol 28.9 mGy | Note 14: Noise level is "equivalent substantially" considering slightly different CTDIvol. |
| CT Number Display Range | -1024 ~+8191 HU | -1024 ~+8191 HU | Same |
| Scan Field of View | Up to 500 mm, 600mm with extend FOV | Up to 500 mm, 600mm with extend FOV | Same |
| Reconstruction Field of View | 40mm-500mm, 40mm-600mm with extend FOV | 40mm-500mm, 40mm-600mm with extend FOV | Same |
| Maximum scannable length | 1700mm | 1700mm | Same |
| Image Matrix | Up to 1024 x 1024 | Up to 1024 x 1024 | Same |
| Reconstructed slice thickness | uCT 760: 0.625mm,1.25mm,2.5mm,5mm,10mm (axial), 0.625-10mm(helical); uCT 780: 0.5mm,0.625mm,1.25mm,2.5mm,5mm,10mm (axial), 0.5-10mm (helical) | 0.55mm,1.1mm,2.2mm,5.5mm,11mm (axial), 0.55-10mm(helical) | Note 15: More slice thickness options for predicate, but similar capabilities overall, not affecting safety/effectiveness. |
| Pitch | 0.1~2.0 | 0.1~2.0 | Same |
| Maximum continuous exposure time | Up to 100seconds | Up to 100seconds | Same |
| Electrical Safety | Comply with ES60601-1 | Comply with ES60601-1 | Same |
| EMC | Comply with IEC60601-1-2 | Comply with IEC60601-1-2 | Same |
| Biocompatibility | Patient Contact Materials were tested and demonstrated no cytotoxicity (ISO 10993-5), no evidence for irritation and sensitization (ISO 10993-10). | Patient Contact Materials were tested and demonstrated no cytotoxicity (ISO 10993-5), no evidence for irritation and sensitization (ISO 10993-10). | Same |
| Iterative noise reduction | KARL 3D | KARL 3D | Same |
| Adaptive Filter | Adaptive Filter | Adaptive Filter | Same |
| Metal artifact reduction | MAC | MAC | Same |
2. Sample Size Used for the Test Set and Data Provenance
The document mentions "Clinical Evaluation for sample clinical images evaluation" and "sample clinical images for proposed devices are provided in Section 36 Clinical Evaluation. Electronic file for each image are provide in MISC Folder." However, it does not specify the sample size of the test set for image evaluation or the data provenance (e.g., country of origin, retrospective/prospective). It only refers to "sample clinical images."
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications
The document does not provide information on the number of experts used, their qualifications, or how ground truth was established for the "sample clinical images."
4. Adjudication Method for the Test Set
The document does not describe any adjudication method for a test set.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
The document does not mention a MRMC comparative effectiveness study or any effect size of AI assistance for human readers. This device is a CT scanner, not an AI-assisted interpretation tool in the context of this submission. The "KARL iterative denoising reconstruction algorithm and MAC Metal artifact correction algorithm" are features of the CT system itself, not necessarily AI for reader improvement.
6. Standalone (Algorithm Only Without Human-in-the-Loop) Performance
As mentioned above, the device is a CT scanner. Its performance verification focuses on technical specifications (e.g., Image Spatial Resolution, Image Noise). The "KARL iterative denoising reconstruction algorithm" and "MAC Metal artifact correction algorithm" are integral parts of the image reconstruction process, not standalone interpretive algorithms. Therefore, a separate "standalone" performance for an interpretive algorithm is not applicable in the context described.
7. Type of Ground Truth Used
For the "sample clinical images evaluation," the type of ground truth is not specified. It refers to "sample clinical images," implying they are real patient images, but how their "truth" was confirmed (e.g., pathology, subsequent expert consensus) is not detailed. For the technical performance parameters (e.g., image noise, spatial resolution), the "ground truth" is typically established through phantom studies and objective measurements according to industry standards.
8. Sample Size for the Training Set
The document does not mention a training set sample size. This submission is for a CT scanner and its image reconstruction algorithms, not a deep learning AI model that typically requires a large training set of labeled data for classification or detection tasks. While iterative reconstruction algorithms have parameters that might be "trained" or optimized, the document does not discuss this in the context of a "training set" of patient images.
9. How the Ground Truth for the Training Set Was Established
Since no training set and its ground truth are mentioned (as per point 8), this information is not provided.
In summary: The provided document is a 510(k) summary focused on demonstrating substantial equivalence of a new CT system to a predicate device based on technical specifications and functional capabilities. It does not contain the detailed clinical study design, acceptance criteria (beyond substantial equivalence to predicate), and performance metrics typically associated with AI-driven diagnostic software, especially concerning human reader performance or defined ground truth for clinical datasets. The "Performance Verification" section lists standards compliance and "Clinical Evaluation for sample clinical images evaluation" but lacks the specifics requested.
Ask a specific question about this device
Page 1 of 1