Search Results

The HiRise is intended to be used for 3-D imaging of the upper and the lower extremities and pelvis of adult and pediatric patients weighing from 40 to 450 lbs.
The device is to be operated in a professional healthcare environment by qualified health care professionals only.

The HiRise is a Cone Beam Computed Tomography Imaging Device that acquires 360-degree rotational projection sequences which are reconstructed into 3D volumetric images of the examined anatomical region. The device uses a gantry assembly, which is comprised of an X- ray source, image detector, and a motorized gantry. The gantry facilitates the acquisition of a full X-ray projection sequence by the acquisition software. For non-weight bearing scans of the lower extremity, a patient positioner accessory allows the patient to sit into a position where he/she can comfortably place his/her anatomy into the imaging bore.
The gantry assembly is mounted on vertical actuators and can travel vertically to capture weight-bearing anatomy at various heights ranging from the feet to the pelvis regions. The HiRise provides total vertical travel of 37 inches to accommodate patients of various sizes. Images produced by the HiRise can be sent electronically to a DICOM complaint image viewing software.

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

1. Acceptance Criteria and Reported Device Performance

• New tube voltage range
• New tube current range
• New scan times
• New max exposure times
• New imaging volumes
• Changes in protocol resolution
• Updated body parts imaged (removal of humerus, forearm protocol configurations)
• New scanner weight
• New power requirements
• New scatter numbers.
  This indicates the device met this criterion. |
  | Internal Verification (X-Ray Circuitry) | Passes all internal verification with no anomalies causing increased risk. | All applicable test plans passed with no anomalies increasing risk. This indicates the device met this criterion. |
  | Third-Party Testing (Safety & Compliance) | Passing all third-party testing to all applicable requirements. | Completed 3rd party testing with no outstanding failures for IEC 60601-1 and all applicable collateral and particular standards (specifically 60601-1-2, 60601-1-3, and 60601-2-44). This indicates the device met this criterion. |
  | Increased Weight | (No specific numerical criterion, implied that negligible increase is acceptable) | Weight changed to 900 lbs (from 850 lbs for scanner). Acknowledged as negligible increase (approx. 50 pounds). Labeling updated. This indicates the change was handled and deemed acceptable. |
  | Minor Material Changes (Biocompatibility) | Biocompatibility assessed in alignment with 10993 and do not impact the safety of the device. | Biocompatibility assessed in alignment with 10993 and do not impact the safety of the device. This indicates the device met this criterion. |

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

The document does not specify the sample size used for the test set.

The data provenance is not explicitly stated in terms of country of origin. The study appears to be retrospective based on the nature of "clinical review of images," implying existing images were evaluated rather than prospectively acquired data specifically for this study. The context suggests it was an internal validation of changes to an existing device (HiRise V2 compared to HiRise V1).

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

"fellowship-trained MSK radiologist."

The document states one fellowship-trained MSK radiologist was used. The specific number of years of experience is not provided, but "fellowship-trained" indicates advanced specialization in Musculoskeletal Radiology.

4. Adjudication Method

The document does not describe an explicit adjudication method (e.g., 2+1, 3+1). Since only one radiologist
is mentioned for the clinical review, it implies no formal multi-reader adjudication was performed for establishing the ground truth of the test set. The single expert's opinion served as the validation.

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

No, a multi-reader multi-case (MRMC) comparative effectiveness study was not reported. The validation involved a single radiologist reviewing images from the device's new protocols, focused on demonstrating diagnostic quality rather than comparing human reader performance with and without AI assistance.

6. Standalone (Algorithm Only) Performance

Yes, a standalone performance evaluation was implied through the "Internal Verification in alignment with all internal testing of previous HiRise" and "Completed 3rd party testing," which would assess the device's technical and imaging output characteristics independently of human interpretation.
However, the clinical acceptability for diagnostic quality was evaluated by the radiologist, which is human-in-the-loop, but the core technical performance before that step is standalone. The document does not describe an AI/algorithm where a standalone AI performance would be distinct from the device's imaging capabilities. This device is a CT X-ray system, not primarily an AI diagnostic algorithm.

7. Type of Ground Truth Used

The ground truth for assessing image quality and diagnostic acceptability was expert consensus from a single fellowship-trained MSK radiologist. The radiologist determined if the images were "clinically acceptable and adequate diagnostic quality."

8. Sample Size for the Training Set

The document does not provide information on a training set sample size. Since this submission is for changes to an existing CT X-ray system (HiRise V2) and its imaging protocols, and not for a new AI/Machine Learning algorithm, the concept of a "training set" in the context of AI development is not directly applicable here. The device's performance relies on its hardware and reconstruction algorithms, which are validated through technical performance standards and clinical review, not typically "trained" on data in the same way an AI model would be.

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

As noted in point 8, the concept of a training set is not directly applicable in the context of this device's submission as described. Therefore, there is no information on how ground truth for a training set was established.

Ask a specific question about this device

The HiRise is intended to be used for 3-D imaging of the upper extremities and pelvis of adult and pediatric patients weighing from 40 to 450 lbs.

The device is to be operated in a professional healthcare environment by qualified health care professionals only.

The HiRise is a Cone Beam Computed Tomography Imaging Device that acquires 360-degree rotational projection sequences which are reconstructed into 3D volumetric images of the examined anatomical region. The device uses a gantry assembly, which is comprised of an X- ray source, image detector, and a motorized gantry. The gantry facilitates the acquisition of a full X-ray projection sequence by the acquisition software. For non-weight bearing scans of the lower extremity, a patient positioner accessory allows the patient to sit into a position where he/she can comfortably place his/her anatomy into the imaging bore.

The gantry assembly is mounted on vertical actuators and can travel vertically to capture weight-bearing anatomy at various heights ranging from the feet to the pelvis regions. The HiRise provides total vertical travel of 37 inches to accommodate patients of various sizes. Images produced by the HiRise can be sent electronically to a DICOM complaint image viewing software.

The provided text describes the HiRise, a Cone Beam Computed Tomography (CBCT) X-ray system, and states that it has been determined substantially equivalent to its predicate device, the CurveBeam LineUP.

Here's an analysis of the acceptance criteria and the study information based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for the HiRise device are primarily established through its demonstrated substantial equivalence to the predicate device, the CurveBeam LineUP. The reported device performance is presented in the context of this comparison and in studies confirming diagnostic quality and safety.

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

The document does not specify a distinct "test set" in terms of patient cases for a statistical performance study comparing the HiRise to a ground truth or predicate quantitatively. Instead, it relies on two primary methods:

• Phantom Scans: "image quality phantoms were scanned in the HiRise and evaluated by a medical physicist." The sample size of phantoms is not specified.
• Clinical Image Review: "datasets of the humerus, elbow, forearm, hand, wrist, pelvis/hip, femur, knee, shin (lower leg or tib/fib) and foot/ankle were reviewed by a board-certified radiologist and found to be of diagnostic quality." The number of patient cases (datasets) reviewed is not specified.
  The data provenance is not explicitly stated (e.g., country of origin, retrospective or prospective). However, the "clinical review of images" suggests actual patient data, likely retrospective.

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

• Number of experts: At least one board-certified radiologist.
• Qualifications: "board-certified radiologist." Specific years of experience are not mentioned.
• Additionally, a "medical physicist" evaluated image quality phantoms.

4. Adjudication method

The adjudication method for the clinical image review is not explicitly described. It states that "a board-certified radiologist... found [images] to be of diagnostic quality." This implies a single expert's assessment without mentioning a consensus process or multiple readers.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done

No, a multi-reader multi-case (MRMC) comparative effectiveness study comparing human readers with AI assistance versus without AI assistance was not conducted or described. The HiRise device is a Computed Tomography X-Ray System, which directly acquires images, and the provided documentation focuses on its imaging performance versus a predicate device, not on AI-assisted interpretation.

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

Yes, an assessment of the device's standalone performance was done. The "performance testing by a medical physicist" and "clinical review of images by a radiologist" evaluating the HiRise images themselves indicate a standalone assessment of the device's output quality. There's no mention of a human-in-the-loop component being evaluated for the device's image acquisition functionality.

7. The type of ground truth used

The ground truth for evaluating the HiRise's performance was established through:

• Expert Consensus/Opinion: For clinical images, it was the "diagnostic quality" determined by a "board-certified radiologist."
• Bench Testing/Phantom Standards: For image quality phantoms, evaluation by a "medical physicist" against established image quality requirements, and verification of "optimal X-Ray tube voltage" and "increased tube current" through bench testing. Additionally, "Detector performance testing" verified image quality met requirements.

8. The sample size for the training set

The document describes the HiRise as an imaging device (hardware and associated software for image acquisition and reconstruction), not a machine learning or AI algorithm in the context of image interpretation. Therefore, there is no "training set" for an AI model mentioned in this submission. The "training set" concept is not applicable here as the device's functionality doesn't appear to be based on a trained AI for image analysis.

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

As there is no "training set" for an AI model mentioned, this question is not applicable based on the provided text.

Ask a specific question about this device

CubeVue serves as an accessory to Cone Beam CT extremity imaging devices with the intended use to retrieve, display, and distribute 2D and 3D volumetric image displaying component allows users to manipulate the images to aid in diagnosis and treatment planning, including rotating through 3D renderings and 2D MPR slices, adjusting display settings, and making measurements.

It is the User's responsibility to ensure monitor quality and ambient light conditions are consistent with the clinical application.

CubeVue serves as an accessory to Cone Beam CT extremity imaging devices with the intended use to retrieve, display, and distribute 2D and 3D volumetric image data.

CubeVue provides a list of patient scans that have been sent to the its image database through its DICOM interface or imported locally by the user can browse, search, and sort the patient list to select a patient and open his or her image data.

The main screen displays a 3D rendering of the image in addition to axial, sagittal, and coronal slices. In the slices, the user can navigate through the volume by paging and rotating. The user can also adjust the window level, zoom, and pan of the 2D slices. In the 3D volume, the user can rotate the volume, cut through a plane, and change the displayed tissue density threshold and rendering style.

The user can make measurements on the image including distances, angles, and density values.

The user can export patient data to a file or media, with the option to anonymize patient demographic information. It supports DICOM and JPEG for image communication.

The provided text is a 510(k) summary for the CubeVue device, which is a Picture Archiving Communications System (PACS). The submission aims to demonstrate substantial equivalence to the predicate device, Osirix MD.

Here's an analysis of the acceptance criteria and the study information:

1. Table of Acceptance Criteria and Reported Device Performance:

The document doesn't explicitly define "acceptance criteria" in a quantitative manner (e.g., target accuracy percentages or specific error rates) for the CubeVue device itself. Instead, the performance testing focuses on demonstrating that CubeVue's functionalities are equivalent to the predicate device, Osirix MD, for the subset of applications CubeVue supports (extremity Cone Beam CT and extremity X-Ray data).

The "Functionality" table provided in the 510(k) summary lists features and directly states whether both CubeVue and Osirix MD possess them. The "Reported Device Performance" is implied by the "Yes" entries in for CubeVue, indicating it meets the described functionalities.

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

The document mentions "Verification and Validation testing" and "Testing Results" but does not specify the sample size of image data used for testing. It also does not explicitly state the country of origin of the data or whether it was retrospective or prospective.

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

The document does not describe the establishment of a "ground truth" for the test set or the involvement of experts for this purpose. The testing primarily involved software validation/verification against system requirements and comparison of functionalities to the predicate device.

4. Adjudication Method for the Test Set:

No adjudication method is described as the testing was focused on software functionality rather than diagnostic performance against a ground truth assessed by experts.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Effect Size of Improvement:

No MRMC comparative effectiveness study is mentioned. This type of study is typically performed for AI/CADe devices that directly assist in diagnostic tasks, which is not the primary claim of this PACS system (CubeVue's role is display and manipulation of images for diagnosis and planning, not direct AI-based interpretation).

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

The testing appears to be "standalone" in the sense that it evaluates the software's functionality independently. However, CubeVue is a PACS accessory to Cone Beam CT devices, and its intended use is to aid users (humans) in diagnosis and treatment planning. The software itself is not making diagnostic decisions without human oversight. The testing described is functional testing of the software.

7. The Type of Ground Truth Used:

As noted above, no specific "ground truth" (like expert consensus, pathology, or outcomes data) is mentioned as being used for performance evaluation in the context of diagnostic accuracy. The ground truth for the functional testing would be the expected behavior of the software based on its design specifications and standard requirements (DICOM, JPEG).

8. The Sample Size for the Training Set:

This device is not an AI/ML algorithm that requires a training set in the conventional sense for learning patterns from data. It is a PACS system for displaying and manipulating images. Therefore, the concept of a "training set" is not applicable here.

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

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

Ask a specific question about this device

The LineUP is intended to be used for 3-D imaging of the foot, knee, hand, and elbow regions to visualize and assess the osseous and certain soft tissue structures, including joint spaces, bone angles and fractures.
It is also intended to capture 2-D images (standard plain x-ray projections) of the foot, knee, hand, and elbow regions.
This modality is anticipated to be applicable to pediatric* cases as well as adults, when appropriate diagnosis of a given condition is considered necessary. Patient parameters: 50 lbs to 400 lbs
*2D Imaging not intended for pediatric use

The LineUP is a Cone Beam Computed Tomography Imaging Device that acquires 360-degree rotational projection sequences which are reconstructed into 3D volumetric images of the examined anatomical region. It can also capture 2D plain X-Ray projections. The device uses a gantry assembly, which is comprised of an X- ray source, image detector, and a motorized gantry facilitates the acquisition of a full X-ray projection sequence by the acquisition software. For hand and elbow scans, a patient transporter accessory allows the patient to sit and then recline into a position where he/she can comfortably place his/her anatomy into the imaging bore. The transporter can also be configured to permit a seated non-weight bearing scan of the feet.
The gantry assembly is mounted on vertical actuators and can travel vertically to capture weight-bearing anatomy at various heights ranging from the feet to knee regions. The LineUP provides total vertical travel of 17 inches to accommodate patients of various sizes. Images produced by the LineUP can be sent electronically to a DICOM complaint image viewing software.

The provided text is a 510(k) Summary for a medical device called "LineUP," a Computed Tomography X-ray system. Here's a breakdown of the acceptance criteria and supporting study details:

1. Table of Acceptance Criteria and Reported Device Performance:

The document doesn't explicitly state "acceptance criteria" in a quantified, pass/fail manner. Instead, it describes performance in terms of diagnostic quality and equivalence to predicate/reference devices.

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

• Sample Size for Test Set: The document does not provide specific numerical sample sizes for the "datasets of foot, knee, hand, and elbow regions" or "anatomic hand, foot, elbow, and knee phantom scans" that were reviewed by radiologists. It refers to these as "datasets" and "anatomic phantom scans" without quantification. For 2D X-ray images, it states "X-Ray images of the foot, knee, hand, and elbow regions" were reviewed, again without a specific number.
• Data Provenance: Not explicitly stated. The context implies that the data would be internal to CurveBeam's testing, likely from phantom studies and potentially some clinical volunteers for the specific anatomies scanned. It does not mention country of origin or whether it's retrospective or prospective.

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

• Number of Experts: Multiple instances refer to "a board-certified radiologist" (singular) reviewing images for diagnostic quality. It also mentions "a medical physicist" for technical image quality assessments.
• Qualifications of Experts: For clinical image review, the expert is identified as "a board-certified radiologist." For technical performance, the expert is "a medical physicist." No specific years of experience are provided.

4. Adjudication method for the test set:

• The document describes a sole expert review (a single board-certified radiologist and a single medical physicist). There is no mention of an adjudication process (such as 2+1, 3+1) because only one expert appears to have been involved in each specific type of review.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, if so, what was the effect size of how much human readers improve with AI vs without AI assistance:

• No, an MRMC comparative effectiveness study was not conducted or described. The LineUP device is an imaging system, not an AI-assisted diagnostic tool for interpretation. The study focused on assessing the diagnostic quality of images produced by the device itself and its equivalence to predicate devices, not on comparing human reader performance with or without AI assistance.

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

• Yes, the performance evaluations described are essentially "standalone" in the context of the device itself.
  • For 3D CT imaging, the device's image quality was assessed by a medical physicist using phantoms and by a radiologist reviewing datasets and anatomical phantom scans for diagnostic quality.
  • For 2D X-ray imaging, technical performance was bench-tested against industry standards and reviewed by a medical physicist and a radiologist.
  • The focus is on the device's ability to produce diagnostic-quality images, not on its ability to interpret those images or interact with a human reader's workflow.

7. The type of ground truth used:

• The ground truth for image quality and diagnostic utility appears to be primarily expert consensus/opinion by a board-certified radiologist for clinical aspects and a medical physicist for technical aspects.
• For technical performance (e.g., meeting IEC standards, detector performance, reconstruction algorithm performance), the ground truth is established through bench testing against established industry standards and internal performance requirements.

8. The sample size for the training set:

• The document does not provide information regarding a specific "training set" or its sample size. This is typical for a 510(k) submission for an imaging device, which primarily demonstrates image quality and safety, rather than an AI/ML algorithm that requires training data.

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

• Since no specific training set is mentioned in the document, there is no information on how ground truth for such a set was established.

Ask a specific question about this device

The In Reach is intended to be used for 3-D imaging of the hand, wrist, elbow, knee, foot, and ankle regions, to visualize and assess the osseous and certain soft tissue structures, bone angles and fractures. This modality is anticipated to be applicable to pediatric* cases as well as adults*, when appropriate diagnosis of a given hand, wrist, elbow, knee, foot or ankle condition is considered necessary.

• Patient parameters: 50 lbs to 400 lbs

The In Reach is a dedicated X-Ray imaging device that acquires a 360 degree rotational X-ray projection sequence and reconstructs a three-dimensional volume of the examined anatomical region, comprised of a stack of slices of specified thickness. The device uses a gantry assembly, which is comprised of an Xray source, image detector, and motorized gantry. The gantry facilitates the acquisition of a full X-ray projection sequence by the acquisition software.

The gantry assembly is mounted on vertical rails and can travel up and down those rails depending on the anatomical region of interest and patient scanning position. The vertical travel allows placement of the field-of-view opening (bore) at the desired height depending on the extremity to be scanned. The In Reach provides total vertical travel of 23.5 inches to facilitate placement of the leg or arm extremity into the bore, one limb at a time. The In Reach provides software tools to measure distances and angles on slices and 3D renderings. Images produced by the In Reach can be printed or optical media, or sent electronically. The In Reach software also displays the selected set of reconstructed slices and 3D renderings on the workstation monitor for viewing.

The provided document is a 510(k) summary for a medical device called "In Reach," a Computed Tomography X-Ray System. It does not contain information about acceptance criteria or a study that directly proves the device meets those criteria in terms of clinical performance or specific diagnostic accuracy metrics, as would be expected for an AI/ML-based device.

Instead, the document focuses on demonstrating substantial equivalence to a predicate device (CurveBeam PedCat) by comparing technological characteristics, intended use similarities, and performance through phantom testing to ensure image quality and safety.

Here's an analysis based on the provided text, addressing your questions where possible:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly present a table of acceptance criteria with corresponding performance metrics. The "performance" described is largely comparative to the predicate device and related to image quality and safety assessments using phantoms.

2. Sample Size for the Test Set and Data Provenance

• Sample Size: The test set appears to consist of various phantoms: a resolution and density phantom, a water equivalent material phantom, and anatomic hand, foot, and knee phantoms (skeleton enclosed in soft tissue equivalent material). The exact "sample size" in terms of number of phantoms or scans is not specified, but it implies a single instance of each type of phantom being scanned and evaluated.
• Data Provenance: Not applicable in the context of phantom testing. These are controlled, manufactured phantoms, not patient data from specific geographical regions. The testing was performed by CurveBeam, LLC.

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

• Number of Experts: One board-certified radiologist.
• Qualifications: Board-certified radiologist. No specific experience length is provided.

4. Adjudication Method for the Test Set

N/A. For the phantom testing, a single board-certified radiologist evaluated the anatomical phantom scans. There is no mention of multiple readers or an adjudication process for these technical performance evaluations.

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

No. The document describes technical phantom testing and comparison to a predicate device, not a human reader study with or without AI assistance.

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

This device itself is the imaging system, not an AI algorithm. The performance evaluation focuses on the image quality and physical characteristics of the CT scanner. Therefore, a "standalone algorithm performance" is not applicable in the typical AI/ML sense. The device's performance stands alone as the output of the CT scan.

7. The Type of Ground Truth Used

For the phantom testing:

• For spatial resolution, HU uniformity, and HU values, the ground truth is the known physical properties and characteristics of the manufactured phantoms (e.g., the specified line pairs, the known density of water, the calibrated densities of materials in the phantom).
• For anatomical phantom scans, the "ground truth" for evaluating osseous details and joint spaces was based on the expert opinion of a board-certified radiologist.

8. The Sample Size for the Training Set

N/A. This document pertains to a medical imaging device (CT scanner), not an AI/ML algorithm that would involve a training set. The device's software components use standard FDK back-projection algorithms for reconstruction, not machine learning that requires a "training set" in the common sense.

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

N/A (as per point 8).

Ask a specific question about this device

The PedCAT is intended to be used for 3-D imaging of the foot & ankle region, to visualize and assess the osseous and certain soft tissue structures, including joint spaces, bone angles and fractures. This modality is anticipated to be applicable to pediatric* cases as well as adults*, when appropriate diagnosis of a given foot condition is considered necessary.

The PedCAT is a dedicated X-Ray imaging device that acquires a 360 degree rotational X-ray sequence, reconstructs a three-dimensional matrix of the examined volume and produces two dimensional views of this volume. The PedCAT can measure distances and thickness on two dimensional images. Images produced by the PedCAT can be printed or exported on magnetic and optical media. The PedCAT gantry is comprised of an X-ray source, image detector, and motorized gantry. The gantry facilitates the acquisition of a full Xray sequence by the software. The software receives the two dimensional images acquired by the detector, transforms them into three dimensional images and displays them on the workstation monitor for viewing.

This document describes the CurveBeam PedCAT, a computed tomography x-ray system intended for 3-D imaging of the foot & ankle region. The submission is a 510(k) summary, which generally focuses on demonstrating substantial equivalence to a predicate device rather than presenting detailed clinical study results against specific acceptance criteria for performance metrics.

The provided text focuses on establishing substantial equivalence to a predicate device (Imaging Sciences DVT Scanner, K051980) based on similar intended use and technological characteristics. The study that proves the device meets "acceptance criteria" in this context is primarily a comparison to the predicate device and phantom testing to show similar imaging performance.

Here's a breakdown of the requested information based on the provided document:

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

The document does not explicitly state "acceptance criteria" with numerical targets for clinical performance metrics (e.g., sensitivity, specificity for fracture detection). Instead, the performance evaluation aims to demonstrate that the PedCAT's imaging characteristics are "authentic and accurate representations of the imaged object, within the expected tolerance for a CBCT device, and very similar to the predicate device."

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

• Test Set Sample Size (Clinical): "A large collection of patients have been scanned at 2 clinical sites." The exact number of patients is not specified.
• Test Set Sample Size (Phantom):
  • ACR phantom
  • Water equivalent material phantom
  • Anatomic foot phantom (foot skeleton enclosed in soft tissue equivalent material)
• Data Provenance:
  • Clinical: Retrospective or Prospective is not explicitly stated, but it was performed "under IRB framework," suggesting a structured approach to clinical data collection. Country of origin not specified.
  • Phantom: Bench testing. Specific country not specified, but the submission is to the FDA in the USA.

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

The document does not specify the number of experts or their qualifications used to establish ground truth for the clinical data. The statement "the results have been found to be substantially similar to the predicate device, in terms of osseous details and joint spaces" implies a comparison by qualified personnel, but details are not provided. For phantom studies, the "ground truth" is the known characteristics of the phantoms themselves.

4. Adjudication method for the test set

The document does not describe a formal adjudication method for the clinical test set (e.g., 2+1, 3+1). The evaluation appears to be a qualitative comparison of image characteristics ("substantially similar... in terms of osseous details and joint spaces") rather than an adjudication of specific findings against a consensus ground truth.

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 explicitly mentioned or described.
• AI Assistance: The PedCAT is described as an "X-Ray imaging device" and the software focuses on image reconstruction and display. There is no mention of Artificial Intelligence (AI) or AI-assisted reading features, so the concept of human reader improvement with/without AI assistance is not applicable to this submission.

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

Yes, a standalone performance assessment was done, primarily through phantom studies. The characteristics of the resultant volume on the ACR phantom and the water equivalent phantom were evaluated for spatial resolution, uniformity of Hounsfield Units, and HU values of various density materials. The anatomic foot phantom also assessed osseous details and joint spaces. These evaluations are based on the output of the device's imaging and reconstruction algorithms without human interpretation in the loop. The clinical evaluation implicitly assesses the algorithm's output as well, but in a physician-reviewed context.

7. The type of ground truth used

• Phantom Studies: The ground truth was the known physical properties and structures of the phantoms (ACR phantom, water equivalent material phantom, anatomic foot phantom).
• Clinical Studies: The implicit ground truth for comparison was the image quality and diagnostic information provided by the predicate device, particularly regarding "osseous details and joint spaces." There is no mention of pathology, surgery, or long-term outcomes data used as ground truth. This is typical for 510(k) submissions focusing on substantial equivalence in imaging.

8. The sample size for the training set

The document does not mention a training set, as the PedCAT is an imaging device and not described as using machine learning or AI models that require a separate training set. The "software domain" section describes image capture tools and reconstruction algorithms (FDK back-projection), which are not typically "trained" in the machine learning sense. Statistical models that might be 'trained' or parameterized for image reconstruction are established algorithms, not typically requiring a distinct "training set" in a regulatory submission like this.

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

Not applicable, as no training set for machine learning or AI was described.

Ask a specific question about this device