(25 days)
Not Found
Yes
The summary explicitly states that the measurements generated from the device are an "inferred reflection of the performance of the deep learning algorithm," and describes the training and testing of this algorithm.
No
The device is an image processing system for visualization and analysis, not for treatment or therapy. It assists in diagnosis and analysis by providing measurements and visualizations but explicitly states that "Patient management decisions should not be made based solely on the results of TumorSight Viz."
No
The device is described as an "image processing system designed to assist in the visualization and analysis" of imaging studies, performing functions like image registration, subtractions, measurements, 3D renderings, and reformats. It explicitly states that "Patient management decisions should not be made based solely on the results of TumorSight Viz," indicating it provides supportive information rather than definitive diagnostic conclusions.
Yes
The device description explicitly states that the TumorSight system consists of proprietary software hosted on a cloud-based platform and accessed on an off-the-shelf computer, indicating it is a software-only medical device.
Based on the provided information, this device is not an In Vitro Diagnostic (IVD).
Here's why:
- IVDs analyze samples taken from the human body. The intended use and device description clearly state that TumorSight Viz analyzes magnetic resonance imaging (MRI) studies, which are images of the body, not samples taken from it (like blood, tissue, or urine).
- The device processes and visualizes imaging data. Its functions are focused on image processing, analysis, and visualization of existing MRI scans.
- The output is image-based analysis and measurements. The results are presented as processed images, measurements derived from those images, and visualizations.
While the device is used in the context of diagnosing and managing breast cancer, its function is to assist in the interpretation of medical images, not to perform a test on a biological sample.
No
The letter explicitly states "Not Found" under "Control Plan Authorized (PCCP) and relevant text". Therefore, there is no indication that the FDA has reviewed or cleared a PCCP for this specific device.
Intended Use / Indications for Use
TumorSight Viz is intended to be used in the visualization and analysis of breast magnetic resonance imaging (MRI) studies for patients with biopsy proven early-stage or locally advanced breast cancer. TumorSight Viz supports evaluation of dynamic MR data acquired from breast studies during contrast administration. TumorSight Viz performs processing functions (such as image registration, subtractions, measurements, 3D renderings, and reformats).
TumorSight Viz also includes user-configurable features for visualizing findings in breast MRI studies. Patient management decisions should not be made based solely on the results of TumorSight Viz.
Product codes
QIH
Device Description
TumorSight Viz is an image processing system designed to assist in the visualization and analysis of breast DCE-MRI studies.
TumorSight reads DICOM magnetic resonance images. TumorSight processes and displays the results on the TumorSight web application.
Available features support:
. Visualization (standard image viewing tools, MIPs, and reformats)
. Analysis (registration, subtractions, kinetic curves, parametric image maps, segmentation and 3D volume rendering)
. Communication and storage (DICOM import, retrieval, and study storage)
The TumorSight system consists of proprietary software developed by SimBioSys, Inc. hosted on a cloud-based platform and accessed on an off-the-shelf computer.
Mentions image processing
Yes
Mentions AI, DNN, or ML
The measurements generated from the device result directly from the segmentation methodology and are an inferred reflection of the performance of the deep learning algorithm.
Input Imaging Modality
magnetic resonance imaging (MRI) studies
Anatomical Site
breast
Indicated Patient Age Range
Not Found
Intended User / Care Setting
Not Found
Description of the training set, sample size, data source, and annotation protocol
DCE-MRI were obtained from eight hundred thirty-three (833) patients (corresponding to 916 samples when accounting for bilateral disease) from more than (9) clinical sites in the U.S. for use in training and tuning the device. All patients had pathologically confirmed invasive, early stage or locally advanced breast cancer.
Data was collected to ensure adequate coverage of MRI manufacturer and field strength, and to ensure similarity with the broader population of early-stage and locally advanced breast cancer patients in the U.S. Specifically, patient age at diagnosis, breast cancer subtype, T stage, histologic subtype, and race/ethnicity all reflect the broader U.S. population.
Training Dataset (n=676 samples)
Age: 70 (3.1%)
Race/Ethnicity: Black (10.2%), Asian and Pacific Islander (4.6%), White (69.7%), American Indian or Alaska Native (0.0%), Other (1.3%), Hispanic (0.0%). Missing/Unknown (14.2%).
Tuning Dataset (n=240 samples)
Age: 70 (8.0%)
Race/Ethnicity: Black (20.5%), Asian and Pacific Islander (3.0%), White (63.3%), American Indian or Alaska Native (0.0%), Other (1.1%), Hispanic (2.3%). Missing/Unknown (0.8%).
Description of the test set, sample size, data source, and annotation protocol
DCE-MRI were obtained for two hundred sixteen (216) patients (corresponding to 217 samples when accounting for bilateral disease) from more than (7) clinical sites in the U.S. for use in validating the device. All patients had pathologically confirmed invasive, early stage or locally advanced breast cancer.
Data was collected to ensure adequate coverage of MRI manufacturer and field strength, and to ensure similarity with the broader population of early-stage and locally advanced breast cancer patients in the U.S. Specifically, patient age at diagnosis, breast cancer subtype, T stage, histologic subtype, and race/ethnicity all reflect the broader U.S. population.
Validation Dataset (n=217 samples)
Age: 70 (6.0%)
Race/Ethnicity: Black (12.0%), Asian and Pacific Islander (9.7%), White (69.1%), American Indian or Alaska Native (0.0%), Other (1.8%), Hispanic (4.6%). Missing/Unknown (2.8%).
Images were acquired from sites that utilize standard of care dynamic contrast enhanced MR protocols from GE, Philips, and Siemens scanners with both 1.5T and 3T field strength magnets.
Three (3) U.S. Board Certified radiologists reviewed 217 validation samples to establish the ground truth for the dataset according to predefined guidelines. For each case, two radiologists measured various characteristics about the cancer including longest dimensions along three axes and tumor to landmark (chest, nipple, skin) distances. Each study was reviewed by two radiologists to determine if the candidate segmentation was appropriate. In cases where the two radiologists did not agree on whether the segmentation was appropriate, a third radiologist provided an additional opinion and established a ground truth by majority consensus.
Independence of validation data from training data was ensured by confirming there was no overlap of patients between training/tuning and validation datasets.
Summary of Performance Studies (study type, sample size, AUC, MRMC, standalone performance, key results)
Performance Tests: SimBioSys has completed performance testing on an independent dataset to ensure TumorSight Viz meets clinically acceptable levels.
Validation Testing (Sample Size: Tumor Volume = 184, Tumor-to-breast volume ratio = 184, Tumor longest dimension = 202, Tumor-to-nipple distance = 200, Tumor-to-skin distance = 202, Tumor-to-chest distance = 202, Tumor center of mass = 184, Tumor segmentation = 184 for both metrics).
Mean Absolute Error (Validation Testing):
Tumor Volume: 5.22 ± 15.58 cubic centimeters (cc)
Tumor-to-breast volume ratio: 0.51 ± 1.48 %
Tumor longest dimension: 1.60 ± 1.93 centimeters (cm)
Tumor-to-nipple distance: 1.20 ± 1.37 centimeters (cm)
Tumor-to-skin distance: 0.63 ± 0.61 centimeters (cm)
Tumor-to-chest distance: 0.91 ± 1.14 centimeters (cm)
Tumor center of mass: 0.72 ± 1.42 centimeters (cm)
Mean ± Std. Dev (Validation Testing) for Tumor segmentation:
Volumetric Dice: 0.75 ± 0.24
Surface Dice: 0.88 ± 0.24
Performance Comparison (Predicate/TumorSight Viz, TumorSight Viz/Ground Truth, Predicate/Ground Truth, Inter-radiologist Variability):
Longest Dimension (N=197):
Abs. Distance Error: Predicate/TumorSight Viz (1.33 cm ± 1.80 cm), TumorSight Viz/Ground Truth (1.59 cm ± 1.93 cm), Predicate/Ground Truth (1.27 cm ± 1.34 cm), Inter-radiologist Variability (1.30 cm ± 1.34 cm)
Tumor to Skin (N=197):
Abs. Distance Error: Predicate/TumorSight Viz (0.24 cm ± 0.39 cm), TumorSight Viz/Ground Truth (0.61 cm ± 0.60 cm), Predicate/Ground Truth (0.55 cm ± 0.48 cm), Inter-radiologist Variability (0.51 cm ± 0.48 cm)
Tumor to Chest (N=197):
Abs. Distance Error: Predicate/TumorSight Viz (0.64 cm ± 1.13 cm), TumorSight Viz/Ground Truth (0.89 cm ± 1.12 cm), Predicate/Ground Truth (0.69 cm ± 0.88 cm), Inter-radiologist Variability (0.97 cm ± 1.16 cm)
Tumor to Nipple (N=195):
Abs. Distance Error: Predicate/TumorSight Viz (0.89 cm ± 1.03 cm), TumorSight Viz/Ground Truth (1.15 cm ± 1.30 cm), Predicate/Ground Truth (1.01 cm ± 1.23 cm), Inter-radiologist Variability (1.03 cm ± 1.30 cm)
Tumor Volume (N=197):
Abs. Volume Error: Predicate/TumorSight Viz (4.42 cc ± 11.03 cc), TumorSight Viz/Ground Truth (5.22 cc ± 15.58 cc), Predicate/Ground Truth (6.50 cc ± 21.40 cc)
Key Results: All tests met the acceptance criteria, demonstrating adequate performance for the intended use. The errors in measurements are consistent with the variability between expert readers. The differences in error between the mean absolute errors (MAE) for the predicate and subject device are clinically acceptable because they are on the order of one to two voxels for the mean voxel size in the dataset and are clinically insignificant.
Key Metrics (Sensitivity, Specificity, PPV, NPV, etc.)
Mean Absolute Error, Volumetric Dice, Surface Dice.
Predicate Device(s)
Reference Device(s)
Not Found
Predetermined Change Control Plan (PCCP) - All Relevant Information
Not Found
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).
0
Image /page/0/Picture/0 description: The image contains the logo of the U.S. Food and Drug Administration (FDA). On the left is the Department of Health & Human Services logo. To the right of that is the FDA logo, which is a blue square with the letters "FDA" in white. To the right of the blue square is the text "U.S. FOOD & DRUG ADMINISTRATION" in blue.
October 25, 2024
SimBioSys, Inc. Kimberly Oleson Senior Vice President of Regulatory Affairs 320 N Sangamon, Suite 700 Chicago, Illinois 60607
Re: K243189 Trade/Device Name: TumorSight Viz Regulation Number: 21 CFR 892.2050 Regulation Name: Medical Image Management And Processing System Regulatory Class: Class II Product Code: QIH Dated: September 30, 2024 Received: September 30, 2024
Dear Kimberly Oleson:
We have reviewed your section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (the Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. Although this letter refers to your product as a device, please be aware that some cleared products may instead be combination products. The 510(k) Premarket Notification Database available at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm identifies combination product submissions. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading.
If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of
1
Page 2
Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.
Additional information about changes that may require a new premarket notification are provided in the FDA guidance documents entitled "Deciding When to Submit a 510(k) for a Change to an Existing Device" (https://www.fda.gov/media/99812/download) and "Deciding When to Submit a 510(k) for a Software Change to an Existing Device" (https://www.fda.gov/media/99785/download).
Your device is also subject to, among other requirements, the Quality System (QS) regulation (21 CFR Part 820), which includes, but is not limited to, 21 CFR 820.30, Design controls; 21 CFR 820.90, Nonconforming product; and 21 CFR 820.100, Corrective and preventive action. Please note that regardless of whether a change requires premarket review, the QS regulation requires device manufacturers to review and approve changes to device design and production (21 CFR 820.30 and 21 CFR 820.70) and document changes and approvals in the device master record (21 CFR 820.181).
Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting of medical device-related adverse events) (21 CFR Part 803) for devices or postmarketing safety reporting (21 CFR Part 4, Subpart B) for combination products (see https://www.fda.gov/combination-products/guidance-regulatory-information/postmarketing-safety-reportingcombination-products); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820) for devices or current good manufacturing practices (21 CFR Part 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR Parts 1000-1050.
All medical devices, including Class I and unclassified devices and combination product device constituent parts are required to be in compliance with the final Unique Device Identification System rule ("UDI Rule"). The UDI Rule requires, among other things, that a device bear a unique device identifier (UDI) on its label and package (21 CFR 801.20(a)) unless an exception or alternative applies (21 CFR 801.20(b)) and that the dates on the device label be formatted in accordance with 21 CFR 801.18. The UDI Rule (21 CFR 830.300(a) and 830.320(b)) also requires that certain information be submitted to the Global Unique Device Identification Database (GUDID) (21 CFR Part 830 Subpart E). For additional information on these requirements, please see the UDI System webpage at https://www.fda.gov/medical-devices/device-advicecomprehensive-regulatory-assistance/unique-device-identification-system-udi-system.
Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to https://www.fda.gov/medical-device-safety/medical-device-reportingmdr-how-report-medical-device-problems.
For comprehensive regulatory information about medical devices and radiation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/medicaldevices/device-advice-comprehensive-regulatory-assistance) and CDRH Learn (https://www.fda.gov/training-and-continuing-education/cdrh-learn). Additionally, you may contact the
2
Division of Industry and Consumer Education (DICE) to ask a question about a specific regulatory topic. See the DICE website (https://www.fda.gov/medical-device-advice-comprehensive-regulatoryassistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).
Sincerely,
Daniel M. Krainak, Ph.D. Assistant Director DHT8C: Division of Radiological Imaging and Radiation Therapy Devices OHT8: Office of Radiological Health Office of Product Evaluation and Quality Center for Devices and Radiological Health
Enclosure
3
Indications for Use
510(k) Number (if known) K243189
Device Name TumorSight Viz
Indications for Use (Describe)
TumorSight Viz is intended to be used in the visualization and analysis of breast maging (MRI) studies for patients with biopsy proven early-stage or locally advanced breast cancer. TumorSight Viz supports evaluation of dynamic MR data acquired from breast studies during contrast administration. TumorSight Viz performs processing functions (such as image registration, subtractions, measurements, 3D renderings, and reformats).
TumorSight Viz also includes user-configurable features for visualizing findings in breast MRI studies. Patient management decisions should not be made based solely on the results of TumorSight Viz.
| Type of Use (Select one or both, as applicable) | |
|---|---|
| ------------------------------------------------- | -- |
X Prescription Use (Part 21 CFR 801 Subpart D)
| Over-The-Counter Use (21 CFR 801 Subpart C)
CONTINUE ON A SEPARATE PAGE IF NEEDED.
This section applies only to requirements of the Paperwork Reduction Act of 1995.
DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW.
The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:
Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff(@fda.hhs.gov
"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."
4
510(k) Summarv
Submitter Details
SimBioSys, Inc. 320 N Sangamon, Suite 700, Chicago IL 60607 United States Contact: Kimberly Oleson Contact Telephone: (612) 803-2610 Contact Email: kim.oleson@simbiosys.com Date of Preparation: September 30, 2024
Details of the Submitted Device
Proprietary Name: TumorSight Viz Common Name: Medical image management and processing system Classification Name: System, Image Processing, Radiological Regulation Number: 892.2050 Product Code: QIH Committee/Panel: Radiology Device Class: II
Type of 510(k) Submission:
Special 510(k)
Identification of the Legally Marketed Predicate Device
Predicate #: K231130
Predicate Trade Name: TumorSight Viz
Product Code: QIH
Device Description
TumorSight Viz is an image processing system designed to assist in the visualization and analysis of breast DCE-MRI studies.
TumorSight reads DICOM magnetic resonance images. TumorSight processes and displays the results on the TumorSight web application.
Available features support:
- . Visualization (standard image viewing tools, MIPs, and reformats)
- . Analysis (registration, subtractions, kinetic curves, parametric image maps, segmentation and 3D volume rendering)
- . Communication and storage (DICOM import, retrieval, and study storage)
The TumorSight system consists of proprietary software developed by SimBioSys, Inc. hosted on a cloud-based platform and accessed on an off-the-shelf computer.
5
Intended Use and Indications for Use
TumorSight Viz is intended to be used in the visualization and analysis of breast magnetic resonance imaging (MRI) studies for patients with biopsy proven early-stage or locally advanced breast cancer. TumorSight Viz supports evaluation of dynamic MR data acquired from breast studies during contrast administration. TumorSight Viz performs processing functions (such as image registration, subtractions, measurements, 3D renderings, and reformats).
TumorSight Viz also includes user-configurable features for visualizing findings in breast MRI studies. Patient management decisions should not be made based solely on the results of TumorSight Viz.
Indications for Use Comparison
TumorSight Viz has the same Indications for Use as the predicate.
Technological Characteristics
Visualization of dynamic magnetic resonance imaging (MRI) studies is the technological principle for both the subject and predicate devices. It is based on the use of dynamic MRI images in DICOM format which are to be viewed and analyzed by a skilled physician. Both the subject and predicate devices perform the following same technological features:
- . Standard Image Viewing Tools (zoom, pan, window/level)
- . Image Post Processing (MIPs, reformats, image registration)
- . Parametric Maps
- Kinetic Curves
- . Automatic Volume Segmentation
- Automatic Linear Measurements (distance to nipple, chest, and closest skin surface)
- . DICOM Image Import
The following technological features differ between the subject and predicate devices:
- . Automated DICOM image import
- . Updated segmentation model
6
Performance Tests
SimBioSys has completed performance testing on an independent dataset to ensure TumorSight Viz meets clinically acceptable levels.
DCE-MRI were obtained from eight hundred thirty-three (833) patients (corresponding to 916 samples when accounting for bilateral disease) from more than (9) clinical sites in the U.S. for use in training and tuning the device. DCE-MRI were obtained for two hundred sixteen (216) patients (corresponding to 217 samples when accounting for bilateral disease) from more than (7) clinical sites in the U.S. for use in validating the device. All patients had pathologically confirmed invasive, early stage or locally advanced breast cancer.
Data was collected to ensure adequate coverage of MRI manufacturer and field strength, and to ensure similarity with the broader population of early-stage and locally advanced breast cancer patients in the U.S. Specifically, patient age at diagnosis, breast cancer subtype, T stage, histologic subtype, and race/ethnicity all reflect the broader U.S. population.
| | Training Dataset
(n=676 samples) | Tuning Dataset
(n=240 samples) | Validation Dataset
(n=217 samples) |
|--------------------------------------|-------------------------------------|-----------------------------------|---------------------------------------|
| Age | | | |
| 70 | 21 (3.1%) | 21 (8.0%) | 13 (6.0%) |
| Missing | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) |
| Race/Ethnicity | | | |
| Black† | 69 (10.2%) | 54 (20.5%) | 26 (12.0%) |
| Asian and Pacific Islander† | 31 (4.6%) | 8 (3.0%) | 21 (9.7%) |
| White† | 471 (69.7%) | 167 (63.3%) | 150 (69.1%) |
| American Indian or Alaska
Native† | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) |
| Other | 9 (1.3%) | 3 (1.1%) | 4 (1.8%) |
| Hispanic | 0 (0.0%) | 6 (2.3%) | 10 (4.6%) |
| Missing/Unknown | 96 (14.2%) | 2 (0.8%) | 6 (2.8%) |
- Non-Hispanic
7
The following subgroups present in the dataset were comparable to the U.S. population: cancer subtype, grade, histology, T stage, and N stage.
Images were acquired from sites that utilize standard of care dynamic contrast enhanced MR protocols from GE, Philips, and Siemens scanners with both 1.5T and 3T field strength magnets.
Three (3) U.S. Board Certified radiologists reviewed 217 validation samples to establish the ground truth for the dataset according to predefined guidelines. For each case, two radiologists measured various characteristics about the cancer including longest dimensions along three axes and tumor to landmark (chest, nipple, skin) distances. Each study was reviewed by two radiologists to determine if the candidate segmentation was appropriate. In cases where the two radiologists did not agree on whether the segmentation was appropriate, a third radiologist provided an additional opinion and established a ground truth by majority consensus.
Independence of validation data from training data was ensured by confirming there was no overlap of patients between training/tuning and validation datasets.
The validation samples were tested using both TumorSight Viz and the predicate device.
The measurements generated from the device result directly from the segmentation methodology and are an inferred reflection of the performance of the deep learning algorithm. For example, the distance from chest or skin is calculated after the deep learning segmentation identifies the region of interest and then the resulting measurement is output.
The mean absolute error and variability between the automated measurements (Yalidation Testing) and ground truth for tumor volume (measured in cc) and landmark distances (measured in cm) was similar to the variability between device-to-radiologist measurements and inter-radiologist variability. This demonstrates that the error in measurements is consistent to the variability between expert readers. Performance data for the automated measurements is summarized below:
| Measurement Description | Units | Validation Testing
(Mean Abs. Error ± Std. Dev.) |
|--------------------------------------|------------------------|-----------------------------------------------------|
| Tumor Volume (n=184) | cubic centimeters (cc) | 5.22 ± 15.58 |
| Tumor-to-breast volume ratio (n=184) | % | 0.51 ± 1.48 |
| Tumor longest dimension (n=202) | centimeters (cm) | 1.60 ± 1.93 |
| Tumor-to-nipple distance (n=200) | centimeters (cm) | 1.20 ± 1.37 |
| Tumor-to-skin distance (n=202) | centimeters (cm) | 0.63 ± 0.61 |
| Tumor-to-chest distance (n=202) | centimeters (cm) | 0.91 ± 1.14 |
| Tumor center of mass (n=184) | centimeters (cm) | 0.72 ± 1.42 |
The tumor segmentation was assessed using the Dice coefficient, utilizing both the volumetric and surface Dice coefficients, which together validate the location, volume, and surface agreement with a reference standard.
8
The surface Dice coefficient is particularly useful as a proxy for the accuracy of 3D rendering and surfaceto-surface distances. Additionally, to further assess the tumor segmentation localization accuracy, we used the distance between the centers of mass of the reference standards and device-generated regions.
Results of Dice and surface Dice are summarized below:
| Performance Measurement | Metric | Validation Testing
(Mean ± Std. Dev.) |
|----------------------------|-----------------|------------------------------------------|
| Tumor segmentation (n=184) | Volumetric Dice | 0.75 ± 0.24 |
| Tumor segmentation (n=184) | Surface Dice | 0.88 ± 0.24 |
We found that all tests met the acceptance criteria, demonstrating adequate performance for our intended use.
Risk Management
The device risks were managed and controlled following the requirements of ISO 14971 standard. The device hazards were identified, their risk levels were evaluated and mitigation measures were taken to reduce the risk levels. The benefits of the TumorSight Viz software, outweigh the device residual risks.
Substantial Equivalence
TumorSight Viz is comparable to the predicate in terms of intended use, technological characteristics, and principle of operation.
| Predicate Device Comparison | ||
|---|---|---|
| Predicate Device | TumorSight Viz | |
| 510(k) | K231130 | K243189 |
| Manufacturer | SimBioSys, Inc. | SimBioSys, Inc. |
| Regulation Number | 892.2050 | 892.2050 |
| Regulation Name | Medical image management and | |
| processing system | Medical image management and | |
| processing system | ||
| Classification | 2 | 2 |
| Device Common Name | Image Processing System | Image Processing System |
| Product Code | QIH | QIH |
| Functions | - Extract dynamic contrast | |
| enhanced MRI sequence from | ||
| MRI images for the 3D display | ||
| and visualization of the anatomy | ||
| of patient's breast | - Extract dynamic contrast | |
| enhanced MRI sequence from | ||
| MRI images for the 3D display | ||
| and visualization of the anatomy | ||
| of patient's breast |
A table comparing the key features of the subject and predicate devices is provided below:
9
| Intended Use | TumorSight Viz is intended to
be used in the visualization and
analysis of breast magnetic
resonance imaging (MRI)
studies for patients with biopsy
proven early-stage or locally
advanced breast cancer.
TumorSight Viz supports
evaluation of dynamic MR data
acquired from breast studies
during contrast administration.
TumorSight Viz performs
processing functions (such as
image registration, subtractions,
measurements, 3D renderings,
and reformats).
TumorSight Viz also includes
user-configurable features for
visualizing and analyzing
findings in breast MRI studies.
Patient management decisions
should not be made based solely
on the results of TumorSight
Viz. | TumorSight Viz is intended to
be used in the visualization and
analysis of breast magnetic
resonance imaging (MRI)
studies for patients with biopsy
proven early-stage or locally
advanced breast cancer.
TumorSight Viz supports
evaluation of dynamic MR data
acquired from breast studies
during contrast administration.
TumorSight Viz performs
processing functions (such as
image registration, subtractions,
measurements, 3D renderings,
and reformats).
TumorSight Viz also includes
user-configurable features for
visualizing and analyzing
findings in breast MRI studies.
Patient management decisions
should not be made based solely
on the results of TumorSight
Viz. |
|--------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Data Source (Input) | MRI | MRI |
| Output/Accessibility | Graphic and text results of
breast anatomy are accessed via
a device with internet
connectivity | Graphic and text results of
breast anatomy are accessed via
a device with internet
connectivity |
| Physical Characteristics | "-non-invasive software package
-DICOM compatible" | "-non-invasive software package
-DICOM compatible" |
| Safety | Clinician review and assessment
of analysis prior to use in pre-
operative planning. | Clinician review and assessment
of analysis prior to use in pre-
operative planning. |
10
| Feature | Predicate Device | TumorSight Viz |
|---|---|---|
| Standard image viewing tools | Yes | Yes |
| MIPs | Yes | Yes |
| Reformats | Yes | Yes |
| Registration | Yes | Yes |
| Subtraction series | Yes | Yes |
| View 3D volume rendering | Yes | Yes |
| Kinetic curves | Yes | Yes |
| Parametric image maps | Yes | Yes |
| Manual DICOM import | Yes | Yes |
| Automated DICOM image import | No | Yes |
| Updated segmentation model | No | Yes |
| View finding volume | Yes | Yes |
| View finding location | Yes | Yes |
| View finding size | Yes | Yes |
| View kinetic curve with highest uptake | Yes | Yes |
| View finding distance to nipple | Yes | Yes |
| View finding distance to skin | Yes | Yes |
| View finding distance to chest | Yes | Yes |
| View adjusted finding size | No - Segmentation is not editable, but surgical margins are editable | No - Segmentation is not editable, but surgical margins are editable |
| Interactive rotation of 3D volume rendering | Yes | Yes |
Performance of TumorSight Viz was directly compared to that of the predicate for measurements including tumor longest dimension, tumor to skin distance, tumor to chest distance, and tumor to nipple distance. As summarized in the following table, these were comparable to inter-radiologist variability in the same measurements:
11
| Performance
Measurement | N | Metric | Predicate/
TumorSight
Viz | TumorSight
Viz/
Ground
Truth | Predicate/
Ground
Truth | Inter-
radiologist
Variability |
|----------------------------|-----|---------------------------|---------------------------------|---------------------------------------|-------------------------------|--------------------------------------|
| | | | (Mean ± Std. Dev.) | (Mean ± Std. Dev.) | (Mean ± Std. Dev.) | (Mean ± Std. Dev.) |
| Longest
Dimension | 197 | Abs.
Distance
Error | 1.33 cm ± 1.80 cm | 1.59 cm ± 1.93 cm | 1.27 cm ± 1.34 cm | 1.30 cm ± 1.34 cm |
| Tumor to Skin | 197 | Abs.
Distance
Error | 0.24 cm ± 0.39 cm | 0.61 cm ± 0.60 cm | 0.55 cm ± 0.48 cm | 0.51 cm ± 0.48 cm |
| Tumor to
Chest | 197 | Abs.
Distance
Error | 0.64 cm ± 1.13 cm | 0.89 cm ± 1.12 cm | 0.69 cm ± 0.88 cm | 0.97 cm ± 1.16 cm |
| Tumor to
Nipple | 195 | Abs.
Distance
Error | 0.89 cm ± 1.03 cm | 1.15 cm ± 1.30 cm | 1.01 cm ± 1.23 cm | 1.03 cm ± 1.30 cm |
| Tumor Volume | 197 | Abs.
Volume
Error | 4.42 cc ± 11.03 cc | 5.22 cc ± 15.58 cc | 6.50 cc ± 21.40 cc | NA |
The differences in error between the mean absolute errors (MAE) for the predicate and subject device are clinically acceptable because they are on the order of one to two voxels for the mean voxel size in the dataset. These differences are clinically insignificant.
Substantial Equivalence Conclusion
The comparison of the features and non-clinical bench performance testing described above demonstrates that TumorSight Viz is substantially equivalent to the predicate device in function. Furthermore, performance testing in an independent dataset of radiologist measurement ground truth demonstrates adequate performance for the intended use.
Additionally, TumorSight Viz measurement outputs were compared directly to the predicate device output for 197 cases, and both sets of measurements were directly compared to radiologist measurements. TumorSight Viz compared equivalently to the predicate device on all measurements including tumor longest dimension, tumor to skin distance, tumor to chest distance, and tumor to nipple distance.
Non-clinical bench testing, an independent assessment of device performance to radiologist ground truth, and a direct comparison to the predicate device demonstrate that TumorSight Viz is substantially equivalent to the predicate device.
12
Functions Not Subject to FDA Premarket Review
This medical device product has functions subject to FDA premarket review as well as functions that are not subject to FDA premarket review. For this application, if the product has functions that are not subject to FDA premarket review, FDA assessed those functions only to the extent that they either could adversely impact the safety and effectiveness of the functions subject to FDA premarket review or they are included as a labeled positive impact that was considered in the assessment of the functions subject to FDA premarket review.