(113 days)
Yes
The device description explicitly states that it uses "deep learning algorithms" to generate organ contours. Deep learning is a subset of machine learning and artificial intelligence.
No.
The device is a post-processing software that generates contours for radiation therapy planning, and its output requires user verification and editing. It does not exert any direct therapeutic effect on the patient.
No
The device is a post-processing software that generates contours of organs-at-risk for radiation therapy planning and does not provide a diagnosis of a patient's medical condition. It is explicitly stated as a "workflow tool for initial segmentation" and its output is intended to be verified and edited by users.
Yes
The device description explicitly states "Auto Segmentation is a post-processing software designed to automatically generate contours..." and "The application is intended as a workflow tool...". It processes existing CT images and outputs a DICOM RTSS file, which is a software format. There is no mention of accompanying hardware or hardware components being part of the device itself.
Based on the provided information, this device is not an IVD (In Vitro Diagnostic).
Here's why:
- Intended Use: The intended use is to generate radiotherapy structure sets (RTSS) for radiation therapy planning. This is a clinical workflow tool used in conjunction with medical imaging, not for analyzing biological samples or providing diagnostic information about a patient's health status based on in vitro tests.
- Device Description: The device processes CT images to create contours of organs at risk. It's a post-processing software for medical images, not a device that performs tests on biological specimens.
- Input: The input is CT images, which are medical images, not biological samples.
- Output: The output is a DICOM RTSS, which is a data format used in radiotherapy planning, not a diagnostic result from an in vitro test.
- Intended User: The intended users are radiotherapy practitioners (dosimetrists, medical physicists, and radiation oncologists), who use the output for treatment planning, not for diagnosing diseases based on laboratory tests.
IVD devices are used to examine specimens derived from the human body (like blood, urine, tissue) to provide information for diagnosis, monitoring, or screening. This device operates on medical images and assists in the planning of a medical treatment (radiotherapy), which falls outside the scope of IVD regulation.
No
The input explicitly states "Control Plan Authorized (PCCP) and relevant text: Not Found", meaning the letter does not contain the required language for a PCCP authorized device.
Intended Use / Indications for Use
Auto Segmentation generates a Radiotherapy Structure Set (RTSS) DICOM with segmented organs at risk which can be used by dosimetrists, medical physicists, and radiation oncologists as initial contours to accelerate workflow for radiation therapy planning. It is the responsibility of the user to verify the processed output contours and user-defined labels for each organ at risk and correct the contours/labels as needed. Auto Segmentation may be used with images acquired on CT scanners, in adult patients.
Product codes
QKB
Device Description
Auto Segmentation is a post-processing software designed to automatically generate contours of organ(s) at risk (OARs) from Computed Tomography (CT) images in the form of a DICOM Radiotherapy Structure Set (RTSS) series. The application is intended as a workflow tool for initial segmentation of OARs to streamline the process of organ at risk delineation. The Auto Segmentation is intended to be used by radiotherapy (RT) practitioners after review and editing, if necessary, and confirming the accuracy of the contours for use in radiation therapy planning.
Auto Segmentation uses deep learning algorithms to generate organ at risk contours for the head and neck, thorax, abdomen and pelvis regions from CT images across 40 organ subregion(s). The automatically generated organ at risk contours are networked to predefined DICOM destination(s), such as review workstations supporting RTSS format, for review and editing, as needed.
The organ at risk contours generated with the Auto Segmentation are designed to improve the contouring workflow by automatically creating contours for review by the intended users. The application is compatible with CT DICOM images with single energy acquisition modes and may be used with both GE and non-GE CT scanner acquired images (contrast), in adult patients.
Mentions image processing
Yes
Mentions AI, DNN, or ML
Yes
Input Imaging Modality
CT scanners
Anatomical Site
head and neck, thorax, abdomen and pelvis regions
Indicated Patient Age Range
Adult (18 - 89 years old)
Intended User / Care Setting
dosimetrists, medical physicists, and radiation oncologists; radiotherapy (RT) practitioners
Description of the training set, sample size, data source, and annotation protocol
The Auto Segmentation algorithms were developed and trained using a dataset of 911 different CT exams from several clinical sites from multiple countries. The original development and training data was used for radiotherapy planning, and so is representative of typical clinical practice for the subject device.
Description of the test set, sample size, data source, and annotation protocol
Performance testing to evaluate the device's performance in segmenting organs-at-risk was performed using a database of 302 retrospective CT radiation therapy planning exams, from multiple clinical sites in North America, Asia, and Europe, that is representative of the clinical scenarios where Auto Segmentation is intended to be used. This bench testing dataset was segregated, completely independent and not used in any stage of algorithm development, including training.
The data was acquired using a variety of CT scanners and scanner protocols from different manufacturers. The demographic distribution of the dataset consists of:
- -Gender: 87 Female, 160 Male, 55 Unknown
- -Age: Adult (18 - 89 years old)
- Ethnicity: Dataset was collected from 9 global sources, including USA, EU, and Asia. -
Note: due to anonymization, gender, age, and ethnicity information is not available for all exams.
Ground truth annotations were established following RTOG and DAHANCA clinical guidelines manually by three independent, qualified radiotherapy practitioners. The annotation process was designed to reflect segmentation practices following international clinical guidelines.
Summary of Performance Studies (study type, sample size, AUC, MRMC, standalone performance, key results)
The evaluation used the DICE similarity coefficient (DSC) as a primary metric to compare 2552 Auto Segmentation generated contours to ground truth contours, and evaluate performance against predefined acceptance DICE values on a per organ basis. These 2552 contours were generated from the 302 unique patient exams in the bench testing dataset.
A qualitative preference study evaluation comparing to a Likert scale was conducted using a database of sample clinical CT images to demonstrate that the contours generated by the Auto Segmentation application are adequate for radiotherapy planning use. Each contour used in the evaluation was generated using the Auto Segmentation application, and reviewed by three qualified radiotherapy practitioners, who provided an assessment of the adequacy of the subject device generated contours. The evaluators completed their assessments independently and were blinded to the results of the other evaluators' assessments. The results of the algorithm clinical testing shows that the Auto Segmentation generated organ contours are adequate for use in radiotherapy planning.
The subject device performance was superior for all organs with atlas-based predicates. The evaluation of the Dice mean for the Auto Segmentation algorithms demonstrates that the algorithm performance is in line with the performance of the predicate, as well as state of the art, recently cleared similar automated contouring devices.
A subgroup analysis found that the algorithms' performance is consistent across multiple CT system vendors, pixel spacing, slice distance, gender, and geographical subgroups. Known limitations are described in the user documentation.
The results of the algorithm testing demonstrate that Auto Segmentation performs as expected.
Key Metrics (Sensitivity, Specificity, PPV, NPV, etc.)
Dice Mean, Lower CI95
Predicate Device(s)
Reference Device(s)
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
May 4, 2023
Image /page/0/Picture/1 description: The image shows 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 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.
GE Medical Systems, LLC % Niki Mavrodieva Regulatory Affairs Leader 3000 N. Grandview Blvd. WAUKESHA WI 53188
Re: K230082
Trade/Device Name: Auto Segmentation Regulation Number: 21 CFR 892.2050 Regulation Name: Medical image management and processing system Regulatory Class: Class II Product Code: QKB Dated: April 7, 2023 Received: April 7, 2023
Dear Niki Mavrodieva:
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 (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 located 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 Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.
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 803) for
1
devices or postmarketing safety reporting (21 CFR 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 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.
Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR Part 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 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.
Image /page/1/Picture/5 description: The image shows a digital signature. The signature is from Lora D. Weidner -S. The date of the signature is 2023.05.04, and the time is 13:08:39-04'00'.
Lora D. Weidner, 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
2
Indications for Use
510(k) Number (if known) K230082
Device Name
Auto Segmentation
Indications for Use (Describe)
Auto Segmentation generates a Radiotherapy Structure Set (RTSS) DICOM with segmented organs at risk which can be used by dosimetrists, medical physicists, and radiation oncologists as initial contours to accelerate workflow for radiation therapy planning. It is the responsibility of the user to verify the processed output contours and user-defined labels for each organ at risk and correct the contours/labels as needed. Auto Segmentation may be used with images acquired on CT scanners, in adult patients.
| Type of Use (Select one or both, as applicable) | |
|---|---|
| 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."
3
Image /page/3/Picture/1 description: The image shows the General Electric (GE) logo. The logo consists of the letters 'GE' in a stylized script, enclosed within a blue circle. There are decorative flourishes resembling water droplets or stylized leaves surrounding the circle, adding a touch of elegance to the design. The logo is simple, recognizable, and represents the multinational conglomerate.
K230082 510(k) SUMMARY
This 510(k) summary of Safety and Effectiveness information is submitted in accordance with the requirement of 21 CFR Part 807.92:
- Date: January 06, 2023 GE Medical Systems, LLC Submitter: 3000 North Grandview Blvd Waukesha, WI 53188 Primary Contact: Niki Mavrodieva
- Regulatory Affairs Leader GE Healthcare Phone: +1 (902) 302-3913 Email: nikolina.mavrodieva@ge.com
- Camille Vidal Secondary Contacts: Senior Director Regulatory Affairs GE Healthcare Phone: +1 (240) 280-5356 Email: camille.vidal@ge.com
| Subject Device Name: | Auto Segmentation |
|---|---|
| Device Classification | Class II |
| Regulation Number: | 21 CFR 892.2050 Medical image management and processing system |
| Product Code: | QKB |
Predicate Device Information
| Device Name: | AccuContour |
|---|---|
| Manufacturer: | Xiamen Manteia Technology LTD. |
| 510(k) Number: | K191928 |
| Regulation Number: | 21 CFR 892.2050 Medical image management and processing system |
| Product Code: | QKB |
Reference Devices Information
| Device Name: | AdvantageSim MD With CT Atlas-Based Contouring and Re-Planning Options |
|---|---|
| Manufacturer: | GE Hungary KFT |
| 510(k) Number: | K132045 |
| Regulation Number: | 21 CFR 892.5840 Radiation therapy simulation system |
| Product Code: | KPQ |
4
Image /page/4/Picture/1 description: The image shows the General Electric (GE) logo. The logo consists of the letters 'GE' in a stylized script, enclosed within a blue circle. The circle is surrounded by several white, teardrop-shaped elements, giving the impression of motion or energy emanating from the center.
Device Description
Auto Segmentation is a post-processing software designed to automatically generate contours of organ(s) at risk (OARs) from Computed Tomography (CT) images in the form of a DICOM Radiotherapy Structure Set (RTSS) series. The application is intended as a workflow tool for initial segmentation of OARs to streamline the process of organ at risk delineation. The Auto Segmentation is intended to be used by radiotherapy (RT) practitioners after review and editing, if necessary, and confirming the accuracy of the contours for use in radiation therapy planning.
Auto Segmentation uses deep learning algorithms to generate organ at risk contours for the head and neck, thorax, abdomen and pelvis regions from CT images across 40 organ subregion(s). The automatically generated organ at risk contours are networked to predefined DICOM destination(s), such as review workstations supporting RTSS format, for review and editing, as needed.
The organ at risk contours generated with the Auto Segmentation are designed to improve the contouring workflow by automatically creating contours for review by the intended users. The application is compatible with CT DICOM images with single energy acquisition modes and may be used with both GE and non-GE CT scanner acquired images (contrast), in adult patients.
Intended Use
Auto Segmentation is intended to be used as a workflow tool for initial anatomy segmentation of organs at risk on CT images as an aid in radiation therapy planning after user confirmation.
Indications for Use
Auto Segmentation generates a Radiotherapy Structure Set (RTSS) DICOM with segmented organs at risk which can be used by dosimetrists, medical physicists, and radiation oncologists as initial contours to accelerate workflow for radiation therapy planning. It is the responsibility of the user to verify the processed output contours and user-defined labels for each organ at risk and correct the contours/labels as needed. Auto Segmentation may be used with images acquired on CT scanners, in adult patients.
Technology:
The proposed device, Auto Segmentation, employs similar fundamental scientific technology as its predicate device.
Comparisons
The Auto Segmentation software is substantially equivalent to the predicate device, AccuContour (K191928). The proposed device is based on the same fundamental technology as the predicate device using deep learning algorithms for organ at risk segmentation. The proposed device is intended for automatic segmentation only, while the predicate has manual contouring, registration, and other general capabilities. The table below summarizes the substantive feature/technological similarities and differences between the predicate device and the proposed device:
5
GE Healthcare 510(k) Premarket Notification Submission – Auto Segmentation
Image /page/5/Picture/1 description: The image shows the General Electric (GE) logo. The logo consists of the letters 'G' and 'E' intertwined in a stylized script, enclosed within a blue circle. The circle has decorative white swirls around the edges, giving it a classic and recognizable appearance.
| Specification | Predicate Device
AccuContour (K191928) | Proposed Device
Auto Segmentation |
|--------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Indications for Use | It is used by radiation oncology
department to register
multimodality images and segment
(non-contrast) CT images, to
generate needed information for
treatment planning, treatment
evaluation and treatment
adaptation. | Auto Segmentation generates a Radiotherapy
Structure Set (RTSS) DICOM with segmented
organs at risk which can be used by dosimetrists,
medical physicists, and radiation oncologists as
initial contours to accelerate workflow for
radiation therapy planning. It is the responsibility
of the user to verify the processed output contours
and user-defined labels for each organ at risk and
correct the contours/labels as needed. Auto
Segmentation may be used with images acquired
on CT scanners, in adult patients. |
| Contra-indications | None | Same |
| Patient Population | Adults only | Same |
| Algorithm | Deep Learning | Same |
| Compatible Modality | Non-contrast CT images | CT (contrast and non-contrast) images |
| OAR Segmentation
Anatomic Regions | Head & Neck
Thorax
Abdomen
Pelvis | Same |
| Workflow | Automated | Same |
| User Interface | Basic result preview of automatic
segmentation results. Manual
segmentation is possible.
Configuration menu. | Automated execution of the software with no user
interaction, other than configuration settings.
Generated contours are automatically transmitted
to review workstation(s) supporting RTSS objects
for review and editing, as needed. |
| Compatible Scanner
Models | No Limitation on scanner model,
DICOM 3.0 compliance required | Same |
| Deployment Platform | Cloud and server-based
deployment | Server-based deployment |
5-3
6
Image /page/6/Picture/1 description: The image shows the General Electric (GE) logo. The logo consists of the letters 'GE' in a stylized script, enclosed within a blue circle. The circle is surrounded by decorative swirls, also in blue, giving the logo a classic and recognizable appearance.
Determination of Substantial Equivalence
Summary of Non-Clinical Testing
Auto Segmentation has successfully completed the design control testing per GE Healthcare's quality system. It was designed under the Quality System Regulations of 21CFR 820 and ISO 13485. No new questions of safety and effectiveness and no unexpected test results were observed.
The following quality assurance measures have been applied to the development of the system:
- Requirement Definition
- Risk Analysis and Control
- Technical Design Reviews
- Formal Design Reviews
- . Software Development Lifecycle
- Safety Testing (Verification)
- Performance Testing (Verification, Validation)
- Software Release
Auto Segmentation has been successfully verified. The testing and results did not raise any new issues of safety and effectiveness. Software documentation provided is for a "Major" Level of Concern.
The Auto Segmentation algorithms were developed and trained using a dataset of 911 different CT exams from several clinical sites from multiple countries. The original development and training data was used for radiotherapy planning, and so is representative of typical clinical practice for the subject device.
Performance testing to evaluate the device's performance in segmenting organs-at-risk was performed using a database of 302 retrospective CT radiation therapy planning exams, from multiple clinical sites in North America, Asia, and Europe, that is representative of the clinical scenarios where Auto Segmentation is intended to be used. This bench testing dataset was segregated, completely independent and not used in any stage of algorithm development, including training.
The data was acquired using a variety of CT scanners and scanner protocols from different manufacturers. The demographic distribution of the dataset consists of:
- -Gender: 87 Female, 160 Male, 55 Unknown
- -Age: Adult (18 - 89 years old)
- Ethnicity: Dataset was collected from 9 global sources, including USA, EU, and Asia. -
Note: due to anonymization, gender, age, and ethnicity information is not available for all exams.
Ground truth annotations were established following RTOG and DAHANCA clinical guidelines manually by three independent, qualified radiotherapy practitioners. The annotation process was designed to reflect segmentation practices following international clinical guidelines.
The evaluation used the DICE similarity coefficient (DSC) as a primary metric to compare 2552 Auto Segmentation generated contours to ground truth contours, and evaluate performance against predefined acceptance DICE values on a per organ basis. These 2552 contours were generated from the 302 unique patient exams in the bench testing dataset.
7
Image /page/7/Picture/1 description: The image shows the logo for General Electric (GE). The logo consists of the letters "GE" in a stylized script, enclosed within a blue circle. There are also two white swirls on either side of the circle.
The acceptance criteria were defined individually for each organ as the target quantitative performance requirement that the segmentation model must reach in order to establish the performance for that specific organ. Acceptance criteria were established from:
- -Published performance metrics from deep-learning based or atlas-based FDA cleared products;
- -Estimation of expected performance based on organ specific characteristics and clinical justification, where a DICE value for deep-learning or atlas-based FDA cleared devices was not accessible.
The Auto Segmentation device performance results are shown in Table 1. The following is a summary of the overall performance evaluation:
| OAR | Auto Segmentation
(subject device) | | Acceptance Criteria | |
|----------------------|---------------------------------------|------------|---------------------|-----------|
| | Dice Mean | Lower C195 | Type | Dice Mean |
| Adrenal Left | 78.68% | 76.63% | Estimated | 68.0% |
| Adrenal Right | 72.48% | 69.78% | Estimated | 68.0% |
| Bladder | 81.50% | 78.33% | Deep learning | 80.0% |
| Body | 99.50% | 99.38% | Atlas-based | 98.1% |
| Brainstem | 87.69% | 87.15% | Deep learning | 88.4% |
| Chiasma | 43.81% | 41.03% | Atlas-based | 11.7% |
| Esophagus | 81.69% | 80.38% | Atlas-based | 45.8% |
| Eye Left | 91.32% | 89.77% | Deep learning | 90.1% |
| Eye Right | 90.25% | 88.23% | Deep learning | 89.9% |
| Femur Left | 97.65% | 97.18% | Atlas-based | 71.6% |
| Femur Right | 97.92% | 97.78% | Atlas-based | 70.8% |
| Kidney Left | 92.53% | 90.30% | Deep learning | 86.8% |
| Kidney Right | 94.82% | 93.48% | Deep learning | 85.6% |
| Lacrimal Gland Left | 59.79% | 57.65% | Deep learning | 50.0% |
| Lacrimal Gland Right | 58.09% | 55.81% | Deep learning | 50.0% |
| Lens Left | 76.86% | 74.80% | Deep learning | 73.3% |
| Lens Right | 79.09% | 77.40% | Deep learning | 75.6% |
| Liver | 94.28% | 92.27% | Deep learning | 91.1% |
| Lung Left | 97.70% | 97.38% | Deep learning | 97.4% |
| Lung Right | 97.99% | 97.81% | Deep learning | 97.8% |
| Mandible | 92.70% | 92.36% | Deep learning | 94.0% |
Table 1: Summary of Auto Segmentation performance
8
GE Healthcare 510(k) Premarket Notification Submission – Auto Segmentation
Image /page/8/Picture/1 description: The image shows the logo for General Electric (GE). The logo is a blue circle with the letters "GE" in a stylized font in the center. The letters are also blue. There are three white swirls around the letters, evenly spaced around the circle.
| OAR | Auto Segmentation
(subject device) | Auto Segmentation
(subject device) | Acceptance Criteria | |
|-------------------------------------------------|---------------------------------------|---------------------------------------|---------------------|-----------|
| | Dice Mean | Lower CI95 | Type | Dice Mean |
| Optic Nerve Left | 79.22% | 77.99% | Deep learning | 71.1% |
| Optic Nerve Right | 80.20% | 78.94% | Deep learning | 71.2% |
| Oral Cavity | 87.43% | 86.20% | Deep learning | 91.0% |
| Pancreas | 80.34% | 78.50% | Estimated | 73.0% |
| Parotid Left | 84.35% | 83.27% | Deep learning | 65.0% |
| Parotid Right | 85.55% | 84.48% | Deep learning | 65.0% |
| Proximal Bronchial Tree
(PBtree) | 84.94% | 83.71% | Atlas-based | 54.8% |
| Inferior PCM (Pharyngeal
Constrictor Muscle) | 70.51% | 68.72% | Estimated | 68.0% |
| Middle PCM | 67.09% | 65.21% | Estimated | 68.0% |
| Superior PCM | 59.57% | 57.85% | Estimated | 50.0% |
| Pericardium | 93.58% | 92.00% | Atlas-based | 84.4% |
| Pituitary | 75.62% | 74.12% | Deep learning | 78.0% |
| Prostate | 79.67% | 77.60% | Atlas-based | 52.1% |
| Spinal Cord | 88.55% | 87.43% | Deep learning | 87.0% |
| Submandibular Left | 86.85% | 85.95% | Deep learning | 77.0% |
| Submandibular Right | 85.70% | 84.79% | Deep learning | 78.0% |
| Thyroid | 85.37% | 84.27% | Deep learning | 83.0% |
| Trachea | 91.02% | 90.47% | Atlas-based | 69.2% |
| Whole Brain | 98.53% | 98.46% | Estimated | 93.0% |
Note: in "Type" column above, Deep learning means similar device using deep learning technology, while Atlas-based means similar devices using Atlas-based technology. Estimated means acceptance criteria were estimated based on organ specific characteristics and clinical justification.
The subject device performance was superior for all organs with atlas-based predicates. The evaluation of the Dice mean for the Auto Segmentation algorithms demonstrates that the algorithm performance is in line with the performance of the predicate, as well as state of the art, recently cleared similar automated contouring devices.
A subgroup analysis found that the algorithms' performance is consistent across multiple CT system vendors, pixel spacing, slice distance, gender, and geographical subgroups. Known limitations are described in the user documentation.
The results of the algorithm testing demonstrate that Auto Segmentation performs as expected.
9
Image /page/9/Picture/1 description: The image shows the General Electric (GE) logo. The logo consists of the letters 'GE' in a stylized script, enclosed within a blue circle. The circle has decorative swirls or flourishes around the letters, giving it a classic and recognizable appearance.
Summary of Clinical Testing
A qualitative preference study evaluation comparing to a Likert scale was conducted using a database of sample clinical CT images to demonstrate that the contours generated by the Auto Segmentation application are adequate for radiotherapy planning use. Each contour used in the evaluation was generated using the Auto Segmentation application, and reviewed by three qualified radiotherapy practitioners, who provided an assessment of the adequacy of the subject device generated contours. The evaluators completed their assessments independently and were blinded to the results of the other evaluators' assessments. The results of the algorithm clinical testing shows that the Auto Segmentation generated organ contours are adequate for use in radiotherapy planning.
Substantial Equivalence Conclusion
Auto Segmentation and the predicate have substantially equivalent indications for use, and represent equivalent technological characteristics, including the use of deep learning algorithms.
Auto Segmentation was developed under GE Healthcare's quality system. Design verification and validation, along with bench testing and the clinical reader study provided in this submission demonstrate that the Auto Segmentation software is substantially equivalent and, hence, as safe and effective as the legally marketed predicate device. GE Healthcare's quality system design, verification, and risk management processes did not identify any unexpected results or new questions of safety and effectiveness.
GE Healthcare believes that Auto Segmentation is substantially equivalent to the predicate device and, hence, is safe and effective for its intended use.