Search Results

Magnetic Resonance Imaging (MRI) is a noninvasive technique used for diagnostic imaging. MRI with its soft tissue contrast capability enables the healthcare professional to differentiate between various soft tissues, for example, fat, water, and muscle, but can also visualize bone structures.

Depending on the region of interest, contrast agents may be used.

The MR system may also be used for imaging during interventional procedures and radiation therapy planning.

The PET images and measures the distribution of PET radiopharmaceuticals in humans to aid the physician in determining various metabolic (molecular) and physiologic functions within the human body for evaluation of diseases and disorders such as, but not limited to, cardiovascular disease, neurological disorders, and cancer.

The integrated system utilizes the MRI for radiation-free attenuation correction maps for PET studies. The integrated system provides inherent anatomical reference for the fused MR and PET images due to precisely aligned MR and PET image coordinate systems.

BIOGRAPH One with software Syngo MR XB10 includes new and modified hardware and software compared to the predicate device, Biograph mMR with software syngo MR E11P-AP01. A high level summary of the new and modified hardware and software is provided below:

Hardware

New Hardware

• Gantry offset phantom
• SDB (Smart Distribution Box)

New Coils

• BM Contour XL Coil
• BM Head/Neck Pro PET-MR Coil
• BM Spine Pro PET-MR Coil
• Transfer of up-to-date RF coils from the reference device MAGNETOM Vida.

Modified Hardware

• Main components such as:
  • Detector cassettes / DEA
  • Phantom holder
  • Gantry tube
  • Backplane
  • Magnet and cabling
  • Gradient coil
  • MaRS (measurement and reconstruction system)
  • MI MARS
  • PET electronics
  • RF transmitter TBX3 3T (TX Box 3)
• Other components such as:
  • Cover
  • Filter plate
  • Patient table
  • RFCEL_TEMP

Modified Coils

• Body coil
• Transfer of up-to-date RF coils from the reference device MAGNETOM Vida with some improvements.

Software

New Features and Applications

• Fast Whole-Body workflows
• Fast Head workflow
• myExam PET-MR Assist
• CS-Vibe
• myExam Implant Suite
• DANTE blood suppression
• SMS Averaging for TSE
• SMS Averaging for TSE_DIXON
• SMS without diffusion function
• BioMatrix Motion Sensor
• RF pulse optimization with VERSE
• Deep Resolve Boost for FL3D_VIBE and SPACE
• Deep Resolve Sharp for FL3D_VIBE and SPACE
• Preview functionality for Deep Resolve Boost
• EP2D_FID_PHS
• EP_SEG_FID_PHS
• ASNR recommended protocols for imaging of ARIA
• Open Workflow
• Ultra HD-PET
• "MTC Mode"
• OpenRecon 2.0
• Deep Resolve Boost for TSE
• GRE_PC
• The following functions have been migrated for the subject device without modifications from MAGNETOM Skyra Fit and MAGNETOM Sola Fit:
  • 3D Whole Heart
• Ghost reduction (Dual polarity Grappa (DPG))
• Fleet Reference Scan
• AutoMate Cardiac (Cardiac AI Scan Companion)
• Complex Averaging
• SPACE Improvement: high bandwidth IR pulse
• SPACE Improvement: increase gradient spoiling
• The following function has been migrated for the subject device without modifications from MAGNETOM Free.Max:
  • myExam Autopilot Spine
• The following functions have been migrated for the subject device without modifications from MAGNETOM Sola:
  • myExam Autopilot Brain
  • myExam Autopilot Knee
• Transfer of further up-to-date SW functions from the reference devices.

New Software / Platform

• PET-Compatible Coil Setup
• Select&GO
• PET-MR components communication

Modified Features and Applications

• HASTE_CT
• FL3D_VIBE_AC
• PET Reconstruction
• Transfer of further up-to-date SW functions from the reference devices with some improvements.

Modified Software / Platform

• Several software functions have been improved. Which are:
  • PET Group
  • PET Viewing
  • PET RetroRecon
  • PET Status and Tune-up/QA

Other Modifications and / or Minor Changes

• Indications for use
• Contraindications
• SAR parameter
• Off-Center Planning Support
• Flip Angle Optimization (Lock TR and FA)
• Inline Image Filter
• Marketing bundle "myExam Companion"
• ID Gain
• Automatic System Shutdown (ASS) sensor (Smoke Detector)
• Patient data display (PDD)

The FDA 510(k) Clearance Letter for BIOGRAPH One refers to several AI/Deep Learning features. However, the provided document does not contain explicit acceptance criteria for these AI features in a table format, nor does it detail a comparative effectiveness study (MRMC study) for human readers. It primarily focuses on demonstrating non-inferiority to the predicate device through various non-clinical tests.

Below is an attempt to extract and synthesize the information based on the provided text, while acknowledging gaps in the information regarding specific acceptance criteria metrics and clinical studies.

Acceptance Criteria and Study Details for BIOGRAPH One AI Features

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state numerical acceptance criteria in a dedicated table format. Instead, it describes performance in terms of achieving "convergence of the training" and "improvements compared to conventional parallel imaging," or confirming "very similar metrics" to the predicate. The "acceptance criteria" are implied by these statements and the successful completion of the described tests.

2. Sample Sizes Used for Test Set and Data Provenance

The document primarily describes a validation dataset which serves as the "test set" for the AI models during development, and an additional "test dataset" for specific evaluations.

• Deep Resolve Boost for FL3D_VIBE and SPACE:

  • Test Set Description: The "collaboration partners (testing)" data is mentioned as the source for testing, implying an external, independent test set. No specific number for this test set is provided beyond the 1265 measurements for training/validation.
  • Sample Size (Validation/Training): 27,679 3D patches from 1265 measurements.
  • Data Provenance: "in-house measurements (training and validation) and collaboration partners (testing)." The country of origin is not specified but is likely Germany (Siemens Healthineers AG) and/or China (Siemens Shenzhen Magnetic Resonance LTD.) where the manufacturing is listed.
  • Retrospective/Prospective: "Input data was retrospectively created from the ground truth by data manipulation and augmentation." This indicates retrospective data use.

• Deep Resolve Sharp for FL3D_VIBE and SPACE:

  • Test Set Description: The document states, "The high-resolution datasets were split to 70% training and 30% validation datasets before training to ensure independence of them." This implies the 30% validation dataset is used as the test set.
  • Sample Size (Validation/Training): 27,679 3D patches from 1265 measurements (split into 70% training and 30% validation).
  • Data Provenance: "in-house measurements (training and validation) and collaboration partners (testing)."
  • Retrospective/Prospective: "Input data was retrospectively created from the ground truth by data manipulation." This indicates retrospective data use.

• Deep Resolve Boost for TSE (First Mention - General Performance):

  • Test Set Description: The "evaluation on the test dataset" is mentioned. The validation set is 30% of the 500 measurements.
  • Sample Size (Validation/Training): Approximately 13,000 high resolution 3D patches from 500 measurements (split into 70% training and 30% validation).
  • Data Provenance: "in-house measurements."
  • Retrospective/Prospective: "Input data was retrospectively created from the ground truth by data manipulation." This indicates retrospective data use.

• Deep Resolve Boost for TSE (Second Mention - Banding Artifacts):

  • Test Set Description: "Additional test dataset for banding artifact reduction: more than 2000 slices." This dataset was acquired after the release of the predicate network.
  • Sample Size: More than 2000 slices.
  • Data Provenance: "in-house measurements and collaboration partners."
  • Retrospective/Prospective: Not explicitly stated for this specific additional dataset, but the training/validation data for the predicate was retrospective.

3. Number of Experts and Qualifications for Ground Truth

• Radiologist Evaluation for Deep Resolve Boost for TSE (Second Mention): The document mentions "the radiologist evaluation revealed no difference in suitability for clinical diagnostics."

  • Number of Experts: Not specified (singular "radiologist" used, but typically multiple are implied for such evaluations).
  • Qualifications: "Radiologist." No specific years of experience or subspecialty are mentioned.

• Other features: For Deep Resolve Boost/Sharp for FL3D_VIBE and SPACE, and Deep Resolve Boost for TSE (first mention), the ground truth is derived directly from acquired image data (see section 7). No independent human expert ground truth establishment for these.

4. Adjudication Method (for Test Set)

• Radiologist Evaluation for Deep Resolve Boost for TSE (Second Mention): The adjudication method is not specified in the document (e.g., 2+1, 3+1). It only states "the radiologist evaluation."

• Other features: Adjudication methods are not applicable as human experts were not establishing ground truth for objective metrics.

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

• Was an MRMC study done? No, the document does not describe an MRMC comparative effectiveness study where human readers' performance with and without AI assistance is compared. The evaluation for Deep Resolve Boost for TSE mentions "radiologist evaluation" but not in a comparative MRMC study context.
• Effect Size: Not applicable, as no MRMC study was performed.

6. Standalone (Algorithm Only) Performance

• Was standalone performance done? Yes, the performance testing for all Deep Resolve features (Boost and Sharp for FL3D_VIBE, SPACE, and TSE) was conducted "algorithm only" by evaluating metrics like PSNR, SSIM, MSE, and LPIPS, and then visual inspection/radiologist evaluation. These refer to the algorithm's direct output performance.

7. Type of Ground Truth Used

• Deep Resolve Boost for FL3D_VIBE and SPACE: "The acquired datasets (as described above) represent the ground truth for the training and validation."
• Deep Resolve Sharp for FL3D_VIBE and SPACE: "The acquired datasets represent the ground truth for the training and validation." Input data was manipulated (cropped k-space) to create low-resolution input and high-resolution output/ground truth from the same dataset.
• Deep Resolve Boost for TSE (First Mention): "The acquired datasets represent the ground truth for the training and validation." Input data was manipulated (cropped k-space) to create low-resolution input and high-resolution output/ground truth from the same dataset.
• Deep Resolve Boost for TSE (Second Mention): "The acquired training/validation datasets... represent the ground truth for the training and validation." Input data was manipulated by undersampling k-space, adding noise, and mirroring k-space.
• Summary: The ground truth for all AI features was derived from acquired, high-resolution original image data (retrospectively manipulated to simulate inputs). For Deep Resolve Boost for TSE (second mention), there was also an implicit "expert consensus" or "expert reading" component for the "radiologist evaluation" regarding clinical suitability.

8. Sample Size for the Training Set

• Deep Resolve Boost for FL3D_VIBE and SPACE: 81% of 1265 measurements (for 27,679 3D patches).
• Deep Resolve Sharp for FL3D_VIBE and SPACE: 70% of 1265 measurements (for 27,679 3D patches).
• Deep Resolve Boost for TSE (First Mention): 70% of 500 measurements (for approx. 13,000 high resolution 3D patches).
• Deep Resolve Boost for TSE (Second Mention): More than 23,250 slices (93% of the combined training/validation dataset from K213693).

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

• Deep Resolve Boost for FL3D_VIBE and SPACE: The "acquired datasets" represent the ground truth. "Input data was retrospectively created from the ground truth by data manipulation and augmentation. This process includes further undersampling of the data by discarding k-space lines as well as creating sub-volumes of the acquired data."
• Deep Resolve Sharp for FL3D_VIBE and SPACE: The "acquired datasets represent the ground truth." "Input data was retrospectively created from the ground truth by data manipulation. k-space data has been cropped such that only the center part of the data was used as input. With this method corresponding low-resolution data as input and high-resolution data as output / ground truth were created for training and validation."
• Deep Resolve Boost for TSE (First Mention): Similar to Deep Resolve Sharp for FL3D_VIBE and SPACE: "The acquired datasets represent the ground truth for the training and validation. Input data was retrospectively created from the ground truth by data manipulation. k-space data has been cropped such that only the center part of the data was used as input. With this method corresponding low-resolution data as input and high-resolution data as output / ground truth were created for training and validation."
• Deep Resolve Boost for TSE (Second Mention): "The acquired training/validation datasets... represent the ground truth for the training and validation. Input data was retrospectively created from the ground truth by data manipulation and augmentation. This process includes further undersampling of the data by discarding k-space lines, lowering of the SNR level by addition of noise and mirroring of k-space data."

In summary, for all AI features, the ground truth for training was established by using high-quality, originally acquired MRI data that was then retrospectively manipulated (e.g., undersampled, cropped, noise added) to create synthetic "lower quality" input data for the AI model to learn from, with the original high-quality data serving as the target output or ground truth.

Ask a specific question about this device

Spectrum Dynamics Medical's VERITON system is intended for use by trained healthcare professionals to aid in the detection, localization, diagnosis, staging and restaging of lesions, diseases, and organ function. For evaluating diseases and disorders such as cardiovascular disease, neurological disorders, and trauma. System outcomes can be used to plan, guide, and monitor therapy.

SPECT: The SPECT component is intended to detect or image the distribution of radionuclides in the body or organ (physiology), using the following techniques: whole body and tomographic imaging.

CT: The CT component is intended to produce cross-sectional images of the body by computer reconstruction of x-ray transmission data (anatomy) from either the same axial plane taken at different angles or spiral planes take at different angles.

SPECT +CT: The SPECT and CT components used together acquire SPECT/CT images. The SPECT images can be corrected for attenuation with the CT images, and can be combined (image registration) to merge the patient's physiological (SPECT) and anatomical (CT) images.

The VERITON CT 300/400 Digital SPECT/CT System is a hybrid imaging system combining SPECT and multi-slice CT imaging for anatomical and functional assessment. The subject device introduces a software-only modification to the cleared system by adding VERITAS.AI Noise Reduction, an optional deep-learning post-processing feature integrated into the VERITON-CT Operator's Console. The AI module uses a convolutional neural network (CNN) to reduce noise in reconstructed SPECT images without altering acquisition parameters, hardware performance, radiation-emitting components, or quantitative reconstruction values.

Three workflow-specific models are included: Bone-IQ.AI, Thera-IQ.AI, and MIBG-IQ.AI.

All hardware, electrical safety characteristics, EMC characteristics, and imaging subsystems are unchanged from the predicate device.

This FDA 510(k) clearance letter describes the VERITON CT Digital SPECT/CT System with an added VERITAS.AI Noise Reduction feature. Here's a breakdown of the acceptance criteria and the study proving the device meets them:

Acceptance Criteria and Reported Device Performance

Device: VERITON CT Digital SPECT/CT System with VERITAS.AI Noise Reduction (Software-only modification)
Purpose of AI Module: Reduce noise in reconstructed SPECT images.
Models included: Bone-IQ.AI, Thera-IQ.AI, and MIBG-IQ.AI.

Study Details Proving Acceptance Criteria

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

   • Total Patients in Dataset: 106 (multi-center, international)
   • Evaluation Set (Test Set): 30 patients
     • 13 Bone (Tc-99m)
     • 17 Soft Tissue (10 Lu-177, 7 mIBG)
   • Number of Samples (SPECT bed-position scans) in Evaluation Set: 30 images (one per patient).
   • Data Provenance: Multi-center, international. The submission does not specify if the data was retrospective or prospective.

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

   • For Quantitative Analysis (Lesion ROIs): Not explicitly stated how many experts defined ROIs, but it mentions Lesion ROIs were defined in MIM (K233620) for quantitative analysis.
   • For Visual Assessment: Two independent, board-certified nuclear medicine physicians. The experience level is not specified beyond being "board-certified."

3. Adjudication Method for the Test Set:

   • For visual assessment, the two physicians rated images independently. Inter-reader agreement was calculated (92–100%), but no explicit adjudication method (like 2+1 or 3+1) is described for resolving disagreements. The high agreement suggests disputes were minimal or not a primary focus of the reporting.

4. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, and the effect size of how much human readers improve with AI vs. without AI assistance:

   • A formal MRMC comparative effectiveness study, comparing human readers with AI assistance vs. without AI assistance, was not explicitly described.
   • The study focused on the performance of the AI noise reduction software by evaluating image quality for human readers and quantitative metrics. While human readers assessed the images, the study did not measure their diagnostic performance improvement with or without AI assistance.

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

   • Yes, performance metrics like Signal Preservation (R² and Slope for lesion maximum voxel intensity) and Noise Suppression (Background SD/Mean ratio reduction) were evaluated as standalone algorithm performance before human visual assessment.
   • The AI module "operates only on reconstructed SPECT images" and "produces secondary enhanced images while preserving original images," indicating a standalone processing step.

6. The Type of Ground Truth Used:

   • For quantitative analysis (Signal Preservation, Noise Suppression): Non-processed clinical images (paired with NR-processed for direct comparison) served as the reference standard. Lesion Regions of Interest (ROIs) were defined for quantitative analysis.
   • For visual assessment: The "reference standard" for comparison was the paired non-processed clinical images that the two nuclear medicine physicians compared against the NR-processed images. This is a comparative qualitative ground truth established by expert consensus.

7. The Sample Size for the Training Set:

   • Training/Tuning Set: 76 patients
     • 39 Bone
     • 37 Lu-177
   • Total SPECT bed-position (BP) scans: 369 (from 106 patients total dataset)
   • Note: The submission mentions "Each patient contributed multiple BP scans (Bone: typically, 3 BPs; Lu-177: typically, 6 BPs)." If all 76 training patients contributed multiple BPs, the actual number of training images is significantly higher than 76.

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

   • The document states that the evaluation set was "not used in training/tuning," implying that the training set also utilized real patient data.
   • The method for establishing ground truth for the training set is not explicitly detailed in the provided text. However, for a noise reduction algorithm, the ground truth typically involves the "original" (non-noise-reduced) image for training the model to predict a "cleaner" version. The quantitative metrics (signal preservation, noise suppression) tested on the evaluation set suggest that similar quantitative measures or comparisons with the original images were likely used during training and tuning.

Ask a specific question about this device

The AnyScan 3.0 NM Scanner Family is intended for use by appropriately trained health care professionals to aid in detecting, localizing, diagnosing, staging and restaging of lesions, tumors, disease and organ function for the evaluation of diseases and disorders such as, but not limited to, cardiovascular disease, neurological disorders and cancer. The images produced by the system can also be used by the physician to aid in radiotherapy treatment planning or additional uses.

SPECT: The SPECT subsystem is intended to provide projection and cross-sectional images through computer reconstruction of the data, representing radioisotope distribution in the patient body or in a specific organ using planar and tomographic scanning modes for isotopes with energies up to 588 keV.

CT: CT component is intended to provide cross sectional images of the body by computer reconstruction of x-ray transmission data providing anatomical information.

PET: The PET component is intended to provide cross- sectional images representing the distribution of tomographic scanning modes.

SPECT+CT: The SPECT and CT components used together acquire SPECT/CT images. The SPECT images can be corrected for attenuation with the CT images, and can be combined (image registration) to merge the patient's physiological (SPECT) and anatomical (CT) images.

PET+CT: The PET and CT components used together acquire PET/CT images. The PET images can be corrected for attenuation with the CT images, and can be combined (image registration) to merge the patient's physiological (PET) and anatomical (CT) images.

The system maintains independent functionality of the SPECT, CT and PET components, allowing for single modality SPECT, CT and/ or PET diagnostic imaging.

Software: The Nucline software is an acquisition, display and analysis package intended to aid the clinician to extract diagnostic information supported by image assessment tools, image enhancement features and image quantification of pathologies in images produced from SPECT, CT, PET and other imaging modalities.

This CT system can be used for low dose lung cancer screening in high risk populations.*

*As defined by professional medical societies. Please refer to clinical literature, including the results of the National Lung Screening Trial (N Engl J Med 2011; 365:395-409) and subsequent literature, for further information.

The AnyScan 3.0 NM Scanner Family will enable clinicians to utilize the device to perform separate studies in SPECT-CT, PET-CT, SPECT, PET and multi-slice CT modalities.

The AnyScan 3.0 NM Scanner Family includes the following products:
AnyScan 3.0 NM Scanner Family

The partial product names 'TRIO' and 'DUO' only differentiate the number of built-in SPECT detectors.

The partial product names 'IQMAX' and 'TheraMAX' only differentiate the type of built-in SPECT detector. The SPECT gamma camera generates nuclear medicine images based on the uptake of radioisotope tracers in a patient's body, and supports integration with CT's anatomical detail for precise reference of the location of the metabolic activity.

The CT component produces cross-sectional images of the body by computer reconstruction of x-ray transmission data from either the same axial plane taken at different angles or spiral planes taken at different angles.

The PET component images and measures the distribution of PET radiopharmaceuticals in humans for the purpose of determining various metabolic (molecular) and physiologic functions within the human body and utilizes the CT for fast attenuation correction maps for PET studies and precise anatomical reference for the fused PET and CT images.

The combination of SPECT, CT, and PET in a single device has several benefits. The SPECT subsystem images biochemical function while the CT subsystem images anatomy. The combination enables scans that not only indicate function, e.g., how active a tumor is, but precise localization, e.g., the precise location of that tumor in the body.

Combined SPECT and CT subsystems are intended for SPECT imaging enhanced with spatially registered CT image-based corrections, anatomical localization of tracer uptake and anatomical mapping. CT can be used to correct for the attenuation in SPECT acquisitions. Attenuation in SPECT is an unwanted side effect of the gamma rays scattering and being absorbed by tissue. This can lead to errors in the final image. The CT directly measures attenuation and can be used to create a 3D attenuation map of the patient which can be used to correct the SPECT images. The SPECT-CT scanner can be used to image and track how much dose was delivered to both the target and the surrounding tissue. The system maintains independent functionality of the CT and SPECT subsystems.

Combined PET and CT subsystems are intended for PET imaging enhanced with spatially registered CT image-based corrections, anatomical localization of tracer uptake and anatomical mapping. system maintains independent functionality of the CT and PET subsystems, allowing for single modality CT and/or PET diagnostic imaging.

A patient positioning light marker is generated by a low-power (Class II per IEC 60825-1) red laser.

Nucline software is installed on acquisition workstation to perform patient management, data management, scan control, image reconstruction and image archival and evaluation. All images conform to DICOM imaging format requirements.

The systems also include display equipment, data storage devices, patient and equipment supports, software, and accessories.

InterView XP; InterView FUSION (K221984) and software is integrated for DICOM image visualization and post-processing.

ClariCT (K212074) software is integrated for DICOM CT de-noising.

N/A

Ask a specific question about this device

The device is a diagnostic imaging system that combines Positron Emission Tomography (PET) and X-ray Computed Tomography (CT) systems. The CT component produces cross-sectional images of the body by computer reconstruction of X-ray transmission data. The PET component images the distribution of PET radiopharmaceuticals in the patient body. The PET component utilizes CT images for attenuation correction and anatomical reference in the fused PET and CT images.

This device is to be used by a trained health care professional to gather metabolic and functional information from the distribution of the radiopharmaceutical in the body for the assessment of metabolic and physiologic functions. This information can assist in the evaluation, detection, localization, diagnosis, staging, restaging, follow-up, therapeutic planning and therapeutic outcome assessment of (but not limited to) oncological, cardiovascular, neurological diseases and disorders. Additionally, this device can be operated independently as a whole body multi-slice CT scanner.

AiCE-i for PET is intended to improve image quality and reduce image noise for FDG whole body data by employing deep learning artificial neural network methods which can explore the statistical properties of the signal and noise of PET data. The AiCE algorithm can be applied to improve image quality and denoising of PET images.

Deviceless PET Respiratory gating system, for use with Cartesion Prime PET-CT system, is intended to automatically generate a gating signal from the list-mode PET data. The generated signal can be used to reconstruct motion corrected PET images affected by respiratory motion. In addition, a single motion corrected volume can automatically be generated. Resulting motion corrected PET images can be used to aid clinicians in detection, localization, evaluation, diagnosis, staging, restaging, follow-up of diseases and disorders, radiotherapy planning, as well as their therapeutic planning, and therapeutic outcome assessment. Images of lesions in the thorax, abdomen and pelvis are mostly affected by respiratory motion. Deviceless PET Respiratory gating system may be used with PET radiopharmaceuticals, in patients of all ages, with a wide range of sizes, body habitus and extent/type of disease.

The Cartesion Prime (PCD-1000A/3) V10.21 combines a high-end CT and a high-throughput PET designed to acquire CT, PET and fusion images.

The high-end CT system is a multi-slice helical CT scanner with a gantry aperture of 780 mm and a maximum scan field of view (FOV) of 700 mm. The high-throughput PET system has a digital PET detector utilizing SiPM sensors with temporal resolution of < 250 ps (238 ps typical). Cartesion Prime (PCD-1000A/3) V10.21 is intended to acquire PET images of any desired region of the whole body and CT images of the same region (to be used for attenuation correction or image fusion), to detect the location of positron emitting radiopharmaceuticals in the body with the obtained images. This device is used to gather the metabolic and functional information from the distribution of radiopharmaceuticals in the body for the assessment of metabolic and physiologic functions. This information can assist research, detection, localization, evaluation, diagnosis, staging, restaging, follow-up of diseases and disorders, as well as their therapeutic planning, and therapeutic outcome assessment. This device can also function independently as a whole body multi-slice CT scanner.

The subject device incorporates the latest reconstruction technology, AiCE-i for PET (Advanced Intelligent Clear-IQ Engine- integrated), intended to improve image quality and reduce image noise for FDG whole body data by employing deep learning artificial neural network methods which can more fully explore the statistical properties of the signal and noise of PET data. The AiCE algorithm will be able to better differentiate signal from noise and can be applied to improve image quality and denoising of PET images compared to conventional PET imaging reconstruction.

A Deviceless PET Respiratory gating system has been implemented for use with the subject device. With this subject device, respiration is extracted using a pre-trained neural network. Respiratory-gated reconstruction is performed at a speed equal to or faster than that with "Normal".

Here's a breakdown of the acceptance criteria and study details for the Cartesion Prime PET-CT System, based on the provided FDA 510(k) clearance letter:

Acceptance Criteria and Device Performance for Cartesion Prime PET-CT System (K251370)

The submission describes two primary feature enhancements: AiCE-i for PET (AiCE2) and Deviceless PET Respiratory gating system (DRG2).

1. Table of Acceptance Criteria and Reported Device Performance

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

For AiCE-i for PET (AiCE2) - Clinical Images:

• Sample Size: 10 PET DICOM clinical 18F-FDG whole body cases.
• Data Provenance: Not explicitly stated, but the submission notes "selected to cover characteristics common to the intended U.S. patient population." The training data for AiCE2 is mentioned to have over half acquired from the U.S.

For Deviceless PET Respiratory Gating (DRG2) - Clinical Images:

• Sample Size: 10 patients.
• Data Provenance: Not explicitly stated, but the submission notes "selected to cover characteristics common to the intended U.S. patient population." The training data for DRG2 was acquired entirely from the U.S.

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

For AiCE-i for PET (AiCE2) - Clinical Images:

• Number of Experts: Three (3) physicians.
• Qualifications: At least 20 years of experience in nuclear medicine.

For Deviceless PET Respiratory Gating (DRG2) - Clinical Images:

• Number of Experts: Three (3) physicians.
• Qualifications: At least 20 years of experience in nuclear medicine.

4. Adjudication Method for the Test Set

The adjudication method is not explicitly stated as 2+1, 3+1, or none. However, for both clinical image evaluations, it states that "All three physicians reported/determined that..." This implies a consensus-based adjudication among the three experts was used to reach the conclusions. It does not indicate individual disagreements were arbitrated by a fourth reader.

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

A formal MRMC comparative effectiveness study, designed to quantify the effect size of human readers improving with AI assistance, was not explicitly described in the provided text. The clinical image evaluations involved expert review and comparison, but the focus was on the algorithm's performance and image quality, not a direct measurement of human reader improvement with vs. without AI assistance.

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

Yes, standalone performance was extensively evaluated for both features:

• AiCE-i for PET:
  • Bench tests for pediatric use (CRC, BGV, CNR, artifacts, SUVmean equivalence).
  • Bench tests for image intensity (SUV max/mean accuracy, tumor volume improvement).
  • Phantom testing (NEMA NU-2, Adult and Pediatric NEMA phantoms, Small Pool phantom) comparing AiCE2 to AiCE1 and conventional methods across quantitative metrics (SUVmean, BGV, CRC, SNR) and for artifact absence.
• Deviceless PET Respiratory Gating:
  • Bench tests for AI operational mode (equivalence to external device gating, improvements in SUV max/mean, tumor volume).
  • Evaluation against predicate DRG1 using reconstructed clinical raw data and quantified outputs.

7. The Type of Ground Truth Used

• For AiCE-i for PET (AiCE2):
  • Phantom Studies: Objective, physics-based ground truth (e.g., known sphere sizes, activity concentrations) for quantitative metrics like SUV, CRC, BGV, SNR.
  • Clinical Image Evaluation: Expert consensus/opinion of three nuclear medicine physicians for subjective assessments like diagnostic quality, image sharpness, and noise levels.
• For Deviceless PET Respiratory Gating (DRG2):
  • Bench Tests/Comparison to DRG1: Quantitative measurements of SUV (max and mean) and tumor volume from reconstructed data, likely compared against a known or established ground truth from reference reconstructions.
  • Clinical Image Evaluation: Expert consensus/opinion of three nuclear medicine physicians for subjective assessments related to diagnostic quality and motion correction effectiveness.

8. The Sample Size for the Training Set

• For AiCE-i for PET (AiCE2): Subset assembled from FDG studies of sixteen (16) cancer patients.
• For Deviceless PET Respiratory Gating (DRG2): FDG studies of 27 cancer patients.

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

The text indicates that both AI algorithms (AiCE2 and DRG2) use deep learning artificial neural network methods. The ground truth for training these networks is implicitly derived from the input PET data itself, with the algorithms learning statistical properties of signal and noise or motion patterns.

• For AiCE-i for PET: The algorithm was "trained to automatically adapt to different noise levels to produce consistently high-quality images." This suggests the training data contained examples of both "noisy" input and perhaps "ideal" or "denoised" outputs (or features that guided the network to achieve denoised outputs with improved image quality), where the "ground truth" was likely the desired image characteristics or underlying signal.
• For Deviceless PET Respiratory Gating: The neural network was "trained on FDG studies... to extract motion information from acquired PET data and to generate a corresponding gating signal." This implies the "ground truth" for training involved identifying and characterizing respiratory motion within the raw PET data, possibly using external motion tracking data if available during training, or highly curated datasets where experts delineated motion patterns. The text does not explicitly state how this ground truth was established, only that it was trained on these patient studies.

Ask a specific question about this device

The PennPET Explorer PET system is a diagnostic imaging device that, together with the co-located IQon CT scanner, combines Positron Emission Tomography (PET) and X-ray Computed Tomography (CT) systems. The IQon CT system images anatomical cross-sections by computer reconstruction of X-ray transmission data. The PET system images the distribution of PET anatomy-specific radiopharmaceuticals in the patient. Together, these systems are used for the purposes of detecting, localizing, diagnosing, staging, re-staging, and follow-up for monitoring therapy response of various diseases in oncology, cardiology, and neurology.

The system is intended to image the whole body, heart, brain, lung, gastrointestinal, bone, lymphatic, and other major organs for a wide range of patient types, sizes, and extent of diseases. The CT scanner can also be operated as fully functional, independent diagnostic tool, including for use in radiation therapy planning.

The PennPET Explorer is based on the PET technology of its predicate device, the Philips Vereos PET/CT scanner, but follows the model of its reference device, the previous Philips Gemini TF PET/CT by having co-located—yet separated—PET and CT scanners served by a common patient table. The PennPET Explorer uses a newly designed 142 cm axial field-of-view (AFOV) PET gantry and is intended to be used with a co-located Philips IQon multi-energy CT and patient table.

The PennPET Explorer PET gantry is a modular system comprising six PET detector rings stacked axially, yielding a 142 cm axial FOV. This allows imaging of the human head, torso, and upper legs in a single frame without moving the patient. The entire imaging chain of components from the detectors to the data acquisition computers is supplied by Philips and consists of components that are used in the Vereos PET scanner. The mechanical structure and data processing software have been modified and developed to handle the additional data from all six PET rings simultaneously.

Each of the six detector rings is substantially equivalent to a Philips Vereos PET scanner.

N/A

Ask a specific question about this device

Hybrid Viewer is a software application for nuclear medicine and radiology. Based on user input, Hybrid Viewer processes, displays and analyzes nuclear medicine and radiology imaging data and presents the results to the user. The results can be stored for future analysis.

Hybrid Viewer is equipped with dedicated workflows which have predefined settings and layouts optimized for specific nuclear medicine investigations.

The software application can be configured based on user needs.

The investigation of physiological or pathological states using measurement and analysis functionality provided by Hybrid Viewer is not intended to replace visual assessment. The information obtained from viewing and/or performing quantitative analysis on the images is used, in conjunction with other patient related data, to inform clinical management.

Hybrid Viewer is a software application which provides 2D and 3D viewing, processing and analysis for nuclear medicine investigations.

The studies can be loaded from patient browsers (e.g., GOLD) or PACS (Picture Archiving and Communication System).

Hybrid Viewer provides general tools which include scrolling, zooming, panning, filtering, motion correction, fusion, registration, triangulation drawing regions of interest, synchronization of studies and performing mathematical operations. Specific investigation areas for Hybrid Viewer are Neurology (BRASS), Cardiology, Gastroenterology, Hepatology, Pneumology, Osteology and Nephrology.

N/A

Ask a specific question about this device

The ACUSON Sequoia and Sequoia Select ultrasound imaging systems are intended to provide images of, or signals from, inside the body by an appropriately trained healthcare professional in a clinical setting for the following applications: Fetal, Abdominal, Pediatric, Neonatal Cephalic, Small Parts, OB/GYN (useful for visualization of the ovaries, follicles, uterus and other pelvic structures), Cardiac, Transesophageal, Pelvic, Vascular, Adult Cephalic, Musculoskeletal and Peripheral Vascular applications.

The system supports the Ultrasonically-Derived Fat Fraction (UDFF) measurement tool to report an index that can be useful as an aid to a physician managing adult and pediatric patients with hepatic steatosis.

The system also provides the ability to measure anatomical structures for fetal, abdominal, pediatric, small organ, cardiac, transrectal, transvaginal, peripheral vessel, musculoskeletal and calculation packages that provide information to the clinician that may be used adjunctively with other medical data obtained by a physician for clinical diagnosis purposes.

The ACUSON Origin and Origin ICE ultrasound imaging systems are intended to provide images of, or signals from, inside the body by an appropriately trained healthcare professional in a clinical setting for the following applications: Fetal, Abdominal, Pediatric, OB/GYN (useful for visualization of the ovaries, follicles, uterus and other pelvic structures), Cardiac, Transesophageal, Intracardiac, Vascular, Adult Cephalic, and Peripheral Vascular applications.

The catheter is intended for intracardiac and intra-luminal visualization of cardiac and great vessel anatomy and physiology as well as visualization of other devices in the heart of adult and pediatric patients. The catheter is intended for imaging guidance only, not treatment delivery, during cardiac interventional percutaneous procedures.

The system also provides the ability to measure anatomical structures for fetal, abdominal, pediatric, cardiac, peripheral vessel, and calculation packages that provide information to the clinician that may be used adjunctively with other medical data obtained by a physician for clinical diagnosis purposes.

The ACUSON Sequoia, Sequoia Select, Origin, and Origin ICE Diagnostic Ultrasound Systems (software version VC10) are multi-purpose, mobile, software-controlled, diagnostic ultrasound systems with an on-screen display of thermal and mechanical indices related to potential bio- effect mechanisms. The function of these ultrasound systems is to transmit, receive, process ultrasound echo data (distance and intensities information about body tissue) in various modes of operation and display it as ultrasound imaging, anatomical and quantitative measurements, calculations, analysis of the human body and fluid flow, etc. These ultrasound systems use a variety of transducers to provide imaging in all standard acquisition modes and also have comprehensive networking and DICOM capabilities.

The provided FDA 510(k) clearance letter and summary discuss the ACUSON Sequoia, Sequoia Select, Origin, and Origin ICE Diagnostic Ultrasound Systems. This document indicates a submission for software feature enhancements and workflow improvements, including an "AI Measure and AI Assist workflow efficiency feature" and "Liver Elastography optimization."

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

Acceptance Criteria and Reported Device Performance

The submission focuses on enhancements to existing cleared devices rather than a de novo AI device. Therefore, the "acceptance criteria" discussed are primarily related to the performance of the Liver Elastography optimization using phantom testing.

Study Details for Liver Elastography Optimization (SWE Performance Testing)

The primary study mentioned in the document for performance evaluation is related to the Liver Elastography optimization.

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

   • Test Set: Calibrated elasticity phantoms. The specific number of phantoms used is not stated beyond "calibrated elasticity phantoms."
   • Data Provenance: Not explicitly stated, but implies laboratory testing using commercially available or manufacturer-certified phantoms. Transducers listed were DAX, 5C1, 9C2, 4V1, and 10L4.

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

   • Ground Truth Establishment: The ground truth for the test set (phantom stiffness) was established by the phantom manufacturer, as they were "calibrated elasticity phantoms certified by the phantom manufacturer."
   • Number/Qualifications of Experts: The document does not specify the number or qualifications of experts involved in the phantom's certification process or in the actual testing of the Siemens device. The testing appears to be objective, relying on the calibrated properties of the phantoms.

3. Adjudication Method for the Test Set:

   • Adjudication Method: Not applicable. Phantom testing typically relies on quantitative measurements against known phantom properties, not human adjudication of results.

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

   • MRMC Study: No, an MRMC comparative effectiveness study was not conducted according to this document. The submission focuses on technical enhancements and phantom validation for elastography, and system safety/effectiveness.

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

   • Standalone Performance: The "SWE Performance Testing" with phantoms could be considered a form of standalone performance assessment as it evaluates the device's measurement capabilities against a known standard. However, the AI Measure and AI Assist features are described as "workflow efficiency features" where measurements are "automatically launched" after classification, implying an interaction with a human user rather than a fully standalone diagnostic output. No specific standalone performance metrics for the AI Measure/Assist components are provided.

6. The Type of Ground Truth Used:

   • Ground Truth: For the elastography testing, the ground truth was the known stiffness values of the calibrated elasticity phantoms.

7. The Sample Size for the Training Set:

   • Training Set Sample Size: The document does not provide information about a training set size for the AI Measure and AI Assist features or the elastography optimization. This type of 510(k) submission typically focuses on validation and verification of changes to an already cleared product, rather than detailing the initial development or training data for AI algorithms.

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

   • Training Set Ground Truth: Not applicable, as information on a specific training set is not provided in this document.

Summary regarding AI components:

While the document mentions "AI Measure" and "AI Assist" as workflow efficiency features (e.g., launching relevant measurements after cardiac view classification), it does not provide detailed performance metrics, test set sizes, ground truth establishment, or clinical study information specifically for these AI components. The 510(k) emphasizes that these are "software feature enhancements and workflow improvements" that, along with other changes, do not raise new questions of safety and effectiveness, leading to substantial equivalence with the predicate device. The only detailed "performance testing" described is for the Liver Elastography optimization using phantoms. This suggests that the AI features themselves might have been validated through internal software verification and validation activities that are not detailed in this public summary, or their impact on diagnostic performance was considered incremental and not requiring specific clinical comparative studies for this particular submission.

Ask a specific question about this device

PHAROS is a dedicated PET scanner intended to obtain Positron Emission Tomography (PET) images of parts of human body that fit in the patient aperture (brain, breast, arms and legs) to detect abnormal patterns of distribution of radioactivity after injection of a positron emitting radiopharmaceutical. This information can assist in diagnosis, therapeutic planning and therapeutic outcome assessment.

PHAROS is a specialized high-sensitivity and high-resolution PET system designed for imaging specific organs, such as the brain, breast, arms and legs.

Positron emission tomography (PET) captures images by detecting the distribution of internal radioactivity in human organs, utilizing radioactive pharmaceuticals. This technology reconstructs the body's internal biochemical and metabolic processes, producing high-resolution 3D visualizations. The method involves measuring a pair of simultaneous gamma rays, each with an energy of 511 keV, resulting from the annihilation of positrons. By labeling the positron emitter with a tracer and using a ring-shaped gamma ray detector, the spatial location of positron-emitting nuclides within the body is visualized.

PHAROS features four different scanning modes, each tailored for specific types of imaging:

1. Brain Scan Mode (Sitting Position):
   This mode is designed for brain imaging while the patient is seated.

2. Brain Scan Mode (Lying Position):
   This mode is designed for brain imaging while the patient lies down on a bed.

3. Breast Scan Mode:
   This mode is designed for breast imaging while the patient lies in a prone position.

4. Periphery Scan Mode:
   This mode is designed for imaging the periphery of the body, including the arms, hands, legs, and knees.

For both upper and lower extremity imaging, the height of detector head can be adjusted to ensure optimal patient comfort and accurate positioning. Aside from the physical height adjustment of the detector head, there is no difference in image acquisition method or image generation algorithm between upper and lower extremity scans.

Here's a summary of the acceptance criteria and study information for the PHAROS device, based on the provided FDA 510(k) clearance letter:

1. Table of Acceptance Criteria and Reported Device Performance

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

The document indicates that "a total of five images were obtained, including those from both patients and a normal control group" for the clinical evaluation.

The provenance of this data (e.g., country of origin, retrospective or prospective) is not explicitly stated in the provided text.

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

Ground truth for the clinical acceptability of the five images was established by "a nuclear medicine physician". The exact number of physicians is not explicitly stated beyond "a physician," implying one. No specific years of experience or other qualifications are provided for this expert. It does not mention experts establishing a "ground truth" for the NEMA phantom performance tests, as these are objective measurements.

4. Adjudication method for the test set

The document states that the images were "assessed by a nuclear medicine physician for clinical acceptability." This implies a direct assessment by this physician. There is no mention of an adjudication method such as 2+1 or 3+1, suggesting a single expert's assessment without a formal adjudication process involving multiple readers for this specific clinical evaluation.

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

No, a multi-reader multi-case (MRMC) comparative effectiveness study evaluating human reader performance with and without AI assistance was not done or reported. The study appears to be a standalone performance evaluation of the device against objective phantom criteria and a limited clinical acceptability assessment.

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

The "Performance Testing – Bench" section, which evaluates the device against NEMA NU2:2018 and NEMA NU4:2008 standards (e.g., spatial resolution, scatter fraction, peak NECR, sensitivity, energy resolution, time resolution), represents a standalone evaluation of the device's intrinsic image acquisition and reconstruction capabilities. This can be considered a standalone performance assessment of the system. The clinical images were also reviewed by a physician, but the NEMA tests are purely objective, algorithm-only type performance.

7. The type of ground truth used

• For the bench tests (NEMA standards): The ground truth is established by the physical properties of the phantoms used in the NEMA NU2:2018 and NEMA NU4:2008 standards, and the adherence to these quantitative metrics. This is an objective, standardized ground truth.
• For the clinical evaluation: The ground truth for the five clinical images was based on the "clinical acceptability" determined by a nuclear medicine physician. This is a form of expert consensus/assessment, though it's not explicitly detailed how this "acceptability" was defined or if it referenced other diagnostic findings or pathology.

8. The sample size for the training set

The provided document does not mention a training set sample size. This 510(k) pertains to a PET scanner hardware device, not an AI/ML software device that typically requires a large training dataset for model development. While the device utilizes algorithms for image reconstruction, the focus here is on the system's physical performance and output quality rather than an AI model's training.

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

As no training set is mentioned in the context of an AI/ML model for this device, this information is not applicable based on the provided text. The "ground truth" related to the device's fundamental function is based on established physics principles for PET imaging and standardized phantom measurements.

Ask a specific question about this device

The Siemens PET/CT systems are combined X-Ray Computed Tomography (CT) and Positron Emission Tomography (PET) scanners that provide registration and fusion of high resolution physiologic and anatomic information.

The CT component produces cross-sectional images of the body by computer reconstruction of X-Ray transmission data from either the same axial plane taken at different angles or spiral planes taken at different angles. The PET subsystem images and measures the distribution of PET radiopharmaceuticals in humans for the purpose of determining various metabolic (molecular) and physiologic functions within the human body and utilizes the CT for fast attenuation correction maps for PET studies and precise anatomical reference for the fused PET and CT images.

The system maintains independent functionality of the CT and PET devices, allowing for single modality CT and/or PET diagnostic imaging.

These systems are intended to be utilized by appropriately trained health care professionals to aid in detecting, localizing, diagnosing, staging and restaging of lesions, tumors, disease and organ function for the evaluation of diseases and disorders such as, but not limited to, cardiovascular disease, neurological disorders and cancer. The images produced by the system can also be used by the physician to aid in radiotherapy treatment planning and interventional radiology procedures.

This system can be used for low dose lung cancer screening in high risk populations.*

*As defined by professional medical societies. Please refer to clinical literature, including the results of the National Lung Screening Trial (N Engl J Med 2011; 365:395-409) and subsequent literature, for further information.

Biograph Trinion PET/CT systems are combined multi-slice X-Ray Computed Tomography and Positron Emission Tomography scanners. This system is designed for whole body oncology, neurology and cardiology examinations. Biograph Trinion PET/CT systems provide registration and fusion of high-resolution metabolic and anatomic information from the two major components of each system (PET and CT). Additional components of the system include a patient handling system and acquisition and processing workstations with associated software.

Biograph Trinion VK20 software is a command-based program used for patient management, data management, scan control, image reconstruction and image archival and evaluation. All images conform to DICOM imaging format requirements.

Biograph PET/CT systems, which are the subject of this application, are substantially equivalent to the commercially available Biograph Trinion VK10 family of PET/CT systems (K233677). Differences compared to the commercially available Biograph Trinion systems include:

• The commercially available SOMATOM go.All and go.Top systems with VB10 (K233650) software have been incorporated into the Biograph Trinion VK20 systems, including commercially available CT features.

• Additional PET axial field of view (FoV) systems allowing for more scalability.

• Additional patient communication and comfort features.

• PET respiratory gating with an external gating device has been implemented.

The Biograph Trinion models may also use the names Biograph Mission, Biograph Wonder, Biograph Ambition and Biograph Devotion for marketing purposes.

The provided FDA 510(k) clearance letter for the Biograph Trinion PET/CT system primarily focuses on demonstrating substantial equivalence to a predicate device and adherence to recognized performance standards. It indicates that "all performance testing met the predetermined acceptance values," but does not provide specific numerical acceptance criteria or reported device performance for an AI/algorithm component, nor does it detail a study proving the device meets AI-specific acceptance criteria. The context suggests the "performance testing" refers to general PET/CT system performance, not AI-driven diagnostic assistance.

Therefore, many of the requested details, particularly those related to a standalone AI algorithm's performance, human-in-the-loop studies, dataset characteristics (sample size, provenance), and ground truth establishment methods for an AI component, are not available in the provided text.

Based on the information available in the document, here's what can be extracted and inferred, with explicit notes where information is missing or not applicable in the context of an AI study.

Acceptance Criteria and Reported Device Performance

The document states that "all performance testing met the predetermined acceptance values." However, it does not specify what those acceptance values were or the precise reported performance metrics beyond this general statement. The tests conducted were primarily related to the physical performance of the PET/CT system as per NEMA NU 2:2024 and NEMA XR 25:2019 standards, not specifically an AI component for diagnostic aid.

Table of Acceptance Criteria and Reported Device Performance (Based on available information for the PET/CT system):

Study Details (Focusing on AI-related aspects where applicable, and general system testing otherwise)

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

   • For System Performance (NEMA tests): The document does not specify a "test set" in terms of patient data. NEMA tests typically involve phantom studies rather than patient data. Thus, sample size and data provenance are not applicable in the traditional sense for these tests.
   • For AI Component: The document does not provide any information on a test set (patient cases, images) or data provenance (e.g., country of origin, retrospective/prospective) for validating an AI component for diagnostic assistance. The descriptions are entirely about the physical PET/CT system.

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

   • For System Performance: Ground truth for NEMA tests is established by physical measurements and calibration standards, not human experts.
   • For AI Component: This information is not provided in the document as there's no mention of an AI-driven diagnostic aid requiring expert-established ground truth.

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

   • For System Performance: Not applicable.
   • For AI Component: This information is not provided in the document.

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:

   • The document does not indicate that an MRMC study was performed for an AI component. The focus is on the substantial equivalence of the PET/CT hardware and software to a predicate device, and compliance with performance standards for the imaging system itself.

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

   • The document does not detail any standalone algorithm performance testing. The performance testing described is for the integrated PET/CT system's physical and functional characteristics.

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

   • For System Performance: Ground truth for NEMA tests involves physical phantoms and established measurement protocols.
   • For AI Component: This information is not provided in the document.

7. The sample size for the training set:

   • This information is not provided in the document, as there is no mention of an AI model that undergoes a separate training process requiring a distinct training set.

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

   • This information is not provided in the document, as there is no mention of an AI model's training set.

Summary of Device and Performance Information from Document:

The provided 510(k) clearance letter for the Biograph Trinion is for a PET/CT imaging system, not an AI-based diagnostic software. The "performance testing" described in the document pertains to the physical and functional aspects of the PET/CT scanner (e.g., spatial resolution, sensitivity, image quality) as measured against industry standards (NEMA NU 2:2024). The clearance is based on proving substantial equivalence to a predicate device and adherence to these well-established performance standards for imaging hardware.

Therefore, the detailed questions regarding AI acceptance criteria, AI test set characteristics, human-in-the-loop studies, and AI ground truth establishment are not addressed in this document because the device being cleared is the imaging system itself, not an AI software component for image analysis or diagnostic support. The document implies that the system can be used for certain clinical applications (like lung cancer screening), but it doesn't describe an automated AI system within the device that requires separate clinical validation with reader studies or large patient datasets.

Ask a specific question about this device

The uMI Panvivo is a PET/CT system designed for providing anatomical and functional images. The PET provides the distribution of specific radiopharmaceuticals. CT provides diagnostic tomographic anatomical information as well as photon attenuation information for the scanned region. PET and CT scans can be performed separately. The system is intended for assessing metabolic (molecular) and physiologic functions in various parts of the body. When used with radiopharmaceuticals approved by the regulatory authority in the country of use, the uMI Panvivo system generates images depicting the distribution of these radiopharmaceuticals. The images produced by the uMI Panvivo are intended for analysis and interpretation by qualified medical professionals. They can serve as an aid in detection, localization, evaluation, diagnosis, staging, re-staging, monitoring, and/or follow-up of abnormalities, lesions, tumors, inflammation, infection, organ function, disorders, and/or diseases, in several clinical areas such as oncology, cardiology, neurology, infection and inflammation. The images produced by the system can also be used by the physician to aid in radiotherapy treatment planning and interventional radiology procedures.

The CT system can be used for low dose CT lung cancer screening for the early detection of lung nodules that may represent cancer. The screening must be performed within the established inclusion criteria of programs / protocols that have been approved and published by either a governmental body or professional medical society.

The proposed device uMI Panvivo combines a 295/235 mm axial field of view (FOV) PET and 160-slice CT system to provide high quality functional and anatomical images, fast PET/CT imaging and better patient experience. The system includes PET system, CT system, patient table, power distribution unit, control and reconstruction system (host, monitor, and reconstruction computer, system software, reconstruction software), vital signal module and other accessories.

The uMI Panvivo has been previously cleared by FDA via K243538. The main modifications performed on the uMI Panvivo (K243538) in this submission are due to the addition of Deep MAC(also named AI MAC), Digital Gating(also named Self-gating), OncoFocus(also named uExcel Focus and RMC), NeuroFocus(also named HMC), DeepRecon.PET (also named as HYPER DLR or DLR), uExcel DPR (also named HYPER DPR or HYPER AiR)and uKinetics. Details about the modifications are listed as below:

• Deep MAC, Deep Learning-based Metal Artifact Correction (also named AI MAC) is an image reconstruction algorithm that combines physical beam hardening correction and deep learning technology. It is intended to correct the artifact caused by metal implants and external metal objects.

• Digital Gating (also named Self-gating, cleared via K232712) can automatically extract a respiratory motion signal from the list-mode data during acquisition which called data-driven (DD) method. The respiratory motion signal was calculated by tracking the location of center-of-distribution(COD) in body cavity mask. By using the respiratory motion signal, system can perform gate reconstruction without respiratory capture device.

• OncoFocus (also named uExcel Focus and RMC, cleared via K232712) is an AI-based algorithm to reduce respiratory motion artifacts in PET/CT images and at the same time reduce the PET/CT misalignment.

• NeuroFocus (also named HMC) is head motion correction solution, which employs a statistics-based head motion correction method that correct motion artifacts automatically using the centroid-of-distribution (COD) without manual parameter tuning to generate motion free images.

• DeepRecon.PET (also named as HYPER DLR or DLR, cleared via K193210) uses a deep learning technique to produce better SNR (signal-to-noise-ratio) image in post-processing procedure.

• uExcel DPR (also named HYPER DPR or HYPER AiR, cleared via K232712) is a deep learning-based PET reconstruction algorithm designed to enhance the SNR of reconstructed images. High-SNR images improve clinical diagnostic efficacy, particularly under low-count acquisition conditions (e.g., low-dose radiotracer administration or fast scanning protocols).

• uKinetics(cleared via K232712) is a kinetic modeling toolkit for indirect dynamic image parametric analysis and direct parametric analysis of multipass dynamic data. Image-derived input function (IDIF) can be extracted from anatomical CT images and dynamic PET images. Both IDIF and populated based input function (PBIF) can be used as input function of Patlak model to generate kinetic images which reveal biodistribution map of the metabolized molecule using indirect and direct methods.

The provided FDA 510(k) clearance letter describes the uMI Panvivo PET/CT System and mentions several new software functionalities (Deep MAC, Digital Gating, OncoFocus, NeuroFocus, DeepRecon.PET, uExcel DPR, and uKinetics). The document includes performance data for four of these functionalities: DeepRecon.PET, uExcel DPR, OncoFocus, and DeepMAC.

The following analysis focuses on the acceptance criteria and study details for these four AI-based image processing/reconstruction algorithms as detailed in the document. The document presents these as "performance verification" studies.

Overview of Acceptance Criteria and Device Performance (for DeepRecon.PET, uExcel DPR, OncoFocus, DeepMAC)

The document details the evaluation of four specific software functionalities: DeepRecon.PET, uExcel DPR, OncoFocus, and DeepMAC. Each of these has its own set of acceptance criteria and reported performance results, detailed below.

1. Table of Acceptance Criteria and Reported Device Performance

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

• DeepRecon.PET:

  • Phantoms: NEMA IQ phantoms.
  • Clinical Patients: 20 volunteers.
  • Data Provenance: "collected from various clinical sites" and explicitly stated to be "different from the training data." The document does not specify country of origin or if it's retrospective/prospective, but "volunteers were enrolled" suggests prospective collection for the test set.

• uExcel DPR:

  • Phantoms: Two NEMA IQ phantom datasets, two uniform cylindrical phantom datasets.
  • Clinical Patients: 19 human subjects.
  • Data Provenance: "derived from uMI Panvivo and uMI Panvivo S," "collected from various clinical sites and during separated time periods," and "different from the training data." "Study cohort" and "human subjects" imply prospective collection for the test set.

• OncoFocus:

  • Clinical Patients: 50 volunteers.
  • Data Provenance: "collected from general clinical scenarios" and explicitly stated to be "on cases different from the training data." "Volunteers were enrolled" suggests prospective collection for the test set.

• DeepMAC:

  • Phantoms: PMMA phantom datasets.
  • Clinical Patients: 20 human subjects.
  • Data Provenance: "from uMI Panvivo and uMI Panvivo S," "collected from various clinical sites" and explicitly stated to be "different from the training data." "Volunteers were enrolled" suggests prospective collection for the test set.

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

The document does not explicitly state that experts established "ground truth" for the quantitative metrics (e.g., SUV, CNR, BV, CR, ∆Volume, ∆SUVmax, HU differences) for the test sets. These seem to be derived from physical measurements on phantoms or calculations from patient image data using established methods.

• For qualitative evaluation/clinical diagnosis assessment:

  • DeepRecon.PET: Two American Board of Radiologists certified physicians.
  • uExcel DPR: Two American board-certified nuclear medicine physicians.
  • OncoFocus: Two American Board of Radiologists-certified physicians.
  • DeepMAC: Two American Board of Radiologists certified physicians.

  The exact years of experience for these experts are not provided, only their board certification status.

4. Adjudication Method for the Test Set

The document states that the radiologists/physicians evaluated images "independently" (uExcel DPR) or simply "were evaluated by" (DeepRecon.PET, OncoFocus, DeepMAC). There is no mention of an adjudication method (such as 2+1 or 3+1 consensus) for discrepancies between reader evaluations for any of the functionalities. The evaluations appear to be separate assessments, with no stated consensus mechanism.

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

• The document describes qualitative evaluations by radiologists/physicians comparing the AI-processed images to conventionally processed images (OSEM/no motion correction/no MAC). These are MRMC comparative studies in the sense that multiple readers evaluated multiple cases.
• However, these studies were designed to evaluate the image quality (e.g., diagnostic sufficiency, noise, contrast, sharpness, lesion detectability, artifact reduction) of the AI-processed images compared to baseline images, rather than to measure an improvement in human reader performance (e.g., diagnostic accuracy, sensitivity, specificity, reading time) when assisted by AI vs. without AI.
• Therefore, the studies were not designed as comparative effectiveness studies measuring the effect size of human reader improvement with AI assistance. They focus on the perceived quality of the AI-processed images themselves.

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

• Yes, for DeepRecon.PET, uExcel DPR, OncoFocus, and DeepMAC, quantitative (phantom and numerical) evaluations were conducted that represent the standalone performance of the algorithms in terms of image metrics (e.g., SUV bias, BV, SNR, CNR, CR, COV, ∆Volume, ∆SUVmax, HU differences). These quantitative results are directly attributed to the algorithm's output without human intervention for the measurement/calculation.
• The qualitative evaluations by the physicians (described in point 3 above) also assess the output of the algorithm, but with human interpretation.

7. The Type of Ground Truth Used

• For Quantitative Evaluations:

  • Phantoms: The "ground truth" for phantom studies is implicitly the known physical properties and geometry of the NEMA IQ and PMMA phantoms, allowing for quantitative measurements (e.g., true SUV, true CR, true signal-to-noise).
  • Clinical Data (DeepRecon.PET, uExcel DPR): For these reconstruction algorithms, "ground-truth images were reconstructed from fully-sampled raw data" for the training set. For the test set, comparisons seem to be made against OSEM with Gaussian filtering or full-count OSEM images as reference/comparison points, rather than an independent "ground truth" established by an external standard.
  • Clinical Data (OncoFocus): Comparisons are made relative to "no motion correction images" (∆Volume and ∆SUVmax), implying these are the baseline for comparison, not necessarily an absolute ground truth.
  • Clinical Data (DeepMAC): Comparisons are made to a "control image" without metal artifacts for quantitative assessment of HU differences.

• For Qualitative Evaluations:

  • The "ground truth" is based on the expert consensus / qualitative assessment by the American Board-certified radiologists/nuclear medicine physicians, who compared images for attributes like noise, contrast, sharpness, motion artifact reduction, and diagnostic sufficiency. This suggests a form of expert consensus, although no specific adjudication is described. There's no mention of pathology or outcomes data as ground truth.

8. The Sample Size for the Training Set

The document provides the following for the training sets:

• DeepRecon.PET: "image samples with different tracers, covering a wide and diverse range of clinical scenarios." No specific number provided.
• uExcel DPR: "High statistical properties of the PET data acquired by the Long Axial Field-of-View (LAFOV) PET/CT system enable the model to better learn image features. Therefore, the training dataset for the AI module in the uExcel DPR system is derived from the uEXPLORER and uMI Panorama GS PET/CT systems." No specific number provided.
• OncoFocus: "The training dataset of the segmentation network (CNN-BC) and the mumap synthesis network (CNN-AC) in OncoFocus was collected from general clinical scenarios. Each subject was scanned by UIH PET/CT systems for clinical protocols. All the acquisitions ensure whole-body coverage." No specific number provided.
• DeepMAC: Not explicitly stated for the training set. Only validation dataset details are given.

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

• DeepRecon.PET: "Ground-truth images were reconstructed from fully-sampled raw data. Training inputs were generated by reconstructing subsampled data at multiple down-sampling factors." This implies that the "ground truth" for training was derived from high-quality, fully-sampled (and likely high-dose) PET data.
• uExcel DPR: "Full-sampled data is used as the ground truth, while corresponding down-sampled data with varying down-sampling factors serves as the training input." Similar to DeepRecon.PET, high-quality, full-sampled data served as the ground truth.
• OncoFocus:
  • For CNN-BC (body cavity segmentation network): "The input data of CNN-BC are CT-derived attenuation coefficient maps, and the target data of the network are body cavity region images." This suggests the target (ground truth) was pre-defined body cavity regions.
  • For CNN-AC (attenuation map (umap) synthesis network): "The input data are non-attenuation-corrected (NAC) PET reconstruction images, and the target data of the network are the reference CT attenuation coefficient maps." The ground truth was "reference CT attenuation coefficient maps," likely derived from actual CT scans.
• DeepMAC: Not explicitly stated for the training set. The mention of pre-trained neural networks suggests an established training methodology, but the specific ground truth establishment is not detailed.

Ask a specific question about this device