Predicate Devices

Reference Devices

No reference devices were used in this submission.

AI/ML (Artificial Intelligence / Machine Learning)

Yes
The device description explicitly states that it uses "pre-trained and expert curated machine learning models" for automatic contouring (segmentation).

Therapeutic

No
This device is a software-only medical device intended for use in radiation treatment planning by providing optimal contours of anatomical structures. It does not directly provide therapy or treatment to a patient.

Diagnostic

No

The device is intended for "deriving optimal contours for input to radiation treatment planning" and "localization and definition of healthy anatomical structures," which aids in treatment planning rather than diagnosing a disease or condition.

SaMD (Software as a Medical Device)

Yes

The device description explicitly states "Limbus Contour is a stand-alone software medical device" and "The Limbus Contour Software assists in the following scenarios... Operates in conjunction with radiation treatment planning systems or DICOM viewing systems to load, save, and display medical images and contours...". It also clarifies that it "does not display or store DICOM images and relies on existing radiotherapy treatment planning systems (TPS) and DICOM image viewers for display and modification of generated segmentations." This confirms it is a software application that processes data from other systems and outputs data, without including any hardware components.

IVD (In Vitro Diagnostic)

Based on the provided information, this device is not an IVD (In Vitro Diagnostic).

Here's why:

Intended Use: The intended use clearly states that the software is for deriving optimal contours for input to radiation treatment planning. This is a process related to medical imaging and treatment delivery, not the examination of specimens derived from the human body for the purpose of providing information for diagnosis, monitoring, or compatibility testing.
Device Description: The device description reinforces that it's a software for automatic contouring of medical images (CT/MRI) for radiation treatment planning. It processes images, not biological samples.
Lack of Biological Specimen Handling: There is no mention of the device interacting with or analyzing biological specimens (like blood, urine, tissue, etc.), which is a core characteristic of IVD devices.
Focus on Image Processing: The device description and intended use heavily emphasize image processing and segmentation.
Intended User: The intended users are radiation oncologists, dosimetrists, and physicists, who are involved in radiation therapy planning and delivery, not typically in vitro diagnostic testing.

In summary, the Limbus Contour software is a medical device used in the field of radiation oncology for image processing and treatment planning, not for performing diagnostic tests on biological samples.

PCCP (Predetermined Change Control Plan) Authorized

No
The letter does not contain any explicit statement that the FDA has reviewed and approved or cleared a Predetermined Change Control Plan (PCCP) for this device.

Overview

Intended Use / Indications for Use

Limbus Contour is a software-only medical device intended for use by trained radiation oncologists, dosimetrists and physicists to derive optimal contours for input to radiation treatment planning.

Supported image modalities are Computed Tomography and Magnetic Resonance. The Limbus Contour Software assists in the following scenarios:

Operates in conjunction with radiation treatment planning systems or DICOM viewing systems to load, save, and display medical images and contours for treatment evaluation and treatment planning.
Creation, transformation, and modification of contours for applications including, but not limited to: transferring contours to radiotherapy treatment planning systems, aiding adaptive therapy and archiving contours for patient follow-up.
Localization and definition of healthy anatomical structures.
Limbus Contour is not intended for use with digital mammography.

Product codes

QKB

Device Description

Limbus Contour is a stand-alone software medical device. It is a single purposes cross-platform application for automatic contouring (segmentation) of CT/MRI DICOM images via pre-trained and expert curated machine learning models. The software is intended to be used by trained medical professionals to derive contours for input to radiation treatment planning. The Limbus Contour software segments normal tissues using machine learning models and further post-processing on machine learning model prediction outputs. Limbus Contour does not display or store DICOM images and relies on existing radiotherapy treatment planning systems (TPS) and DICOM image viewers for display and modification of generated segmentations. Limbus Contour interfaces with the user's operating system (importing DICOM image .dcm files and exporting segmented DICOM RT-Structure Set .dcm files).

Mentions image processing

Yes

Mentions AI, DNN, or ML

Yes

Input Imaging Modality

Computed Tomography, Magnetic Resonance

Anatomical Site

Pelvis, Thorax, Abdomen, Head Neck, All (for VB's)

Indicated Patient Age Range

Not Found

Intended User / Care Setting

trained radiation oncologists, dosimetrists and physicists

Description of the training set, sample size, data source, and annotation protocol

The architecture for the neural network models used in our device borrows its primary structure from the U-Net (Ronneberger 2015) and ResUNet (Diakogiannisa 2020). We use the Adam optimization algorithm (Kingma 2014) and the Sørensen-Dice coefficient loss function (Sørensen 1948) to train the network. The models are trained with examples of medical images and the corresponding human-generated contours for the region of interest. During training, an image is shown to the model, and the model generates a contour. The contour generated by the model is compared to the ground truth human-generated contour. The error between the generated and ground truth contours is used to adjust the parameters (weights) of the model. Backpropagation is the algorithm that uses the error to adjust the model parameters. The scans used in training and validating the models were collected from a variety of anonymized and pseudo-anonymized datasets that included human-generated contours. Datasets are from publicly available clinical trials, and from the company's clinical and research partners. All data is acquired, stored, and managed in accordance with the company's dataset management procedure. The fundamental criteria for including a dataset for training and validation is that the scans were generated for use in the course of radiotherapy treatment. Training dataset ground truth contours were reviewed and edited when necessary by our in-house clinicians and radiation oncologist to ensure consistency with established standards and guidelines for contouring. To reduce bias in the training data, the following selection criteria are in place: Datasets from multiple clinical sites (institutions) are included. Datasets include clinical sites of multiple country origin. Including but not limited to: United States, Canada, United Kingdom, France, Germany, Italy, Netherlands, Switzerland, Australia, New Zealand, Singapore. Ground truth annotations are consistent with recognized consensus contouring guidelines (RTOG 1106, RTOG 0848, EMBRACE II, DAHANCA, NRG, ESTRO, ACROP, EPTN). Datasets include different makes and models of imaging devices (CT/MR scans). Sampling data from many different institutions and countries ensures a variety of imaging devices are captured. Imaging devices included in training datasets include the following manufacturers: GE, Siemens, Phillips, Toshiba, Elekta. All imaging data was captured as part of standard clinic protocols for image acquisition in the course of radiation treatment planning. All imaging data is DICOM format and conforms to the DICOM standard (version 3.0 and newer). The scans and ground truth contours in the datasets are from the general population of patients receiving radiotherapy treatment. The datasets are not restricted on the basis of age, ethnicity, race, gender, or disease states. A mix of healthy patients with disease were used as training data. By sampling a large volume of data from a variety of institutions, countries, and including publically available datasets; a general population of patients receiving radiotherapy treatment is created. Multiple scan locations were used in the training data. For example, the scans used to train the parotid gland model were head and neck scans, while the scans used to train the bladder model were pelvic scans. The scans come from many different makes and models of imaging devices for both the CT and MRI scans. The total number of unique scans included in training datasets exceeds 10,000 scans. Example sample sizes for specific structures range from 35 to 1698 training scans and 4 to 140 validation scans.

Description of the test set, sample size, data source, and annotation protocol

For each structure, a set of at least 10 patient scans is used for testing. A single scan may be used to test multiple structures. These test scans are randomly selected from our total pool of patient scans that contain the structure and are not included in training or validation of the models. Our total pool of patient scans that contain the structures were selected to reflect the general population of patients receiving radiation treatments. Each patient test scan has a corresponding DICOM RT-Structure file which contains ground truth contours for some structures. The ground truth contours are from multiple experts at multiple institutions. They have all been reviewed by a board certified radiation oncologist to ensure structures are labeled properly and contoured according to clinical trial guidelines.

Summary of Performance Studies (study type, sample size, AUC, MRMC, standalone performance, key results)

Automatic Contouring - Validation Test (Benchtop performance testing)
Sample Size: For each structure, a set of at least 10 patient scans was used for testing. The smallest number of scans used for a test was 10, and the largest was 28.
Standalone Performance: Contours generated by Limbus Contour are compared to ground truth expert contours using the Sørensen-Dice Similarity Coefficient (DSC).
Key Results: For each structure, the Limbus Mean DSC, Limbus DSC Std Dev, Number of Scans, Limbus DSC lower 95% conf edge, and Test DSC Threshold were reported. All structures passed the performance testing, meaning the lower edge of the 95% confidence interval for Limbus DSC was greater than or equal to the Test DSC Threshold.

Examples of Passed Results:

A_Aorta (CT, Thorax): Limbus Mean DSC: 0.909095, Limbus DSC Std Dev: 0.0455771, Number of Scans: 10, Limbus DSC lower 95% conf edge: 0.87649337, Test DSC Threshold: 0.81. Result: Passed.
Brain (CT, Head Neck): Limbus Mean DSC: 0.992205, Limbus DSC Std Dev: 0.00251205, Number of Scans: 16, Limbus DSC lower 95% conf edge: 0.99078444, Test DSC Threshold: 0.988. Result: Passed.
Prostate (MRI, Pelvis): Limbus Mean DSC: 0.915164, Limbus DSC Std Dev: 0.03096645, Number of Scans: 10, Limbus DSC lower 95% conf edge: 0.89301348, Test DSC Threshold: 0.8. Result: Passed.
SpinalCord (CT, Head Neck/Thorax): Limbus Mean DSC: 0.87788679, Limbus DSC Std Dev: 0.06353613, Number of Scans: 28, Limbus DSC lower 95% conf edge: 0.8507265, Test DSC Threshold: 0.722. Result: Passed.

Key Metrics (Sensitivity, Specificity, PPV, NPV, etc.)

Sørensen-Dice Similarity Coefficient (DSC)

Predicate Device(s)

K230575

Reference Device(s)

No reference devices were used in this submission.

Predetermined Change Control Plan (PCCP) - All Relevant Information

Not Found

Regulation Number and Section

§ 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).

Summary

0

October 9, 2024

Image /page/0/Picture/1 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 consists of a blue square with the letters "FDA" in white, followed by the words "U.S. FOOD & DRUG ADMINISTRATION" in blue.

Limbus AI Inc. Jonathan Giambattista Director - Software 2431 Glamis Place Regina, SK S4N3K9 Canada

Re: K241837

Trade/Device Name: Limbus Contour Regulation Number: 21 CFR 892.2050 Regulation Name: Medical Image Management And Processing System Regulatory Class: Class II Product Code: QKB Dated: July 15, 2024 Received: July 15, 2024

Dear Jonathan Giambattista:

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 (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 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"

1

(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 (OS) 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-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 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-regulatory

2

assistance/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,

Locoa Werchner

Lora D. Weidner, Ph.D. Assistant Director Radiation Therapy Team 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

Submission Number (if known)

K241837

Device Name

Limbus Contour

Indications for Use (Describe)

Limbus Contour is a software-only medical device intended for use by trained radiation oncologists, dosimetrists and physicists to derive optimal contours for input to radiation treatment planning.

Supported image modalities are Computed Tomography and Magnetic Resonance. The Limbus Contour Software assists in the following scenarios:

Operates in conjunction with radiation treatment planning systems or DICOM viewing systems to load, save, and display medical images and contours for treatment evaluation and treatment planning.

Creation, transformation, and modification of contours for applications including, but not limited to: transferring contours to radiotherapy treatment planning systems, aiding adaptive therapy and archiving contours for patient follow-up.

Localization and definition of healthy anatomical structures.

Limbus Contour is not intended for use with digital mammography.

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."

4

K241837

510(k) SUMMARY

This summary of 510(k) safety and effectiveness information is submitted in accordance with the requirements of 21 CFR §807.92:

SUBMITTER -Limbus Al Inc. 2431 Glamis Pl Regina, Saskatchewan, Canada, S4V1A5 Tel: 1-306-502-5982

Contact Person:	Jon Giambattista
Date Prepared:	October 9, 2024

II. DEVICE
Name of Device:	Limbus Contour
Classification Name:	Radiological Image Processing System
Regulation:	21 CFR §892.2050
Regulatory Class:	Class II
Product Classification Code:	QKB

III. PREDICATE DEVICE

Predicate Manufacturer:	Limbus AI, Inc.
Predicate Trade Name:	Limbus Contour
Predicate 510(k):	K230575

No reference devices were used in this submission.

IV. DEVICE DESCRIPTION

Limbus Contour is a stand-alone software medical device. It is a single purposes cross-platform application for automatic contouring (segmentation) of CT/MRI DICOM images via pre-trained and expert curated machine learning models. The software is intended to be used by trained medical professionals to derive contours for input to radiation treatment planning. The Limbus Contour software segments normal tissues using machine learning models and further post-processing on machine learning model prediction outputs. Limbus Contour does not display or store DICOM images and relies on existing radiotherapy treatment planning systems (TPS) and DICOM image viewers for display and modification of generated segmentations. Limbus Contour interfaces with the user's operating system (importing DICOM image .dcm files and exporting segmented DICOM RT-Structure Set .dcm files).

V. INDICATIONS FOR USE

Limbus Contour is a software-only medical device intended for use by trained radiation oncologists, dosimetrists and physicists to derive optimal contours for input to radiation treatment planning.

Supported image modalities are Computed Tomography and Magnetic Resonance. The Limbus Contour Software assists in the following scenarios:

Operates in conjunction with radiation treatment planning systems to load, save, and ● display medical images and contours for treatment evaluation and treatment planning.
. Creation, transformation, and modification of contours for applications including, but not limited to: transferring contours to radiotherapy treatment planning systems, aiding adaptive therapy and archiving contours for patient follow-up.

5

. Localization and definition of anatomical structures.
Limbus Contour is not intended for use with digital mammography.

VI. COMPARISON OF TECHNOLOGICAL CHARACTERISTICS WITH THE PREDICATE DEVICE

The following characteristics were compared between the subject device and the predicate device in order to demonstrate substantial equivalence:

Indications for Use The predicate and subject device are identical with the exception . that the predicate has support for MacOS while the subject device does not and the predicate has less structures for automatic contouring available.
. Materials – The predicate and subject device are software-only devices and do not inherently contain material.
Design - The predicate and subject device have equivalent designs.
. Energy Source – The predicate and subject device are software-only devices, powered by the computer system.
. Performance Testing – The predicate and subject device were both validated using an automatic contouring test to ensure the contours were accurate.

| Item | Limbus Contour v1.8 | Limbus Contour v1.7 - K230575 | Similarities /
Differences |
|-----------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------|
| Classification
Regulation | 892.2050 - Medical image
management and processing
system | 892.2050 - Medical image
management and processing
system | Same |
| Product Code | QKB | LLZ | Similar; Both
product codes
refer to the
same CFR
892.2050 -
Medical image
management
and
processing
system |
| Indications for
Use | Limbus Contour is a software only
medical device intended for use
by trained radiation oncologists,
dosimetrists and physicists to
derive optimal contours for input
to radiation treatment planning.

Supported image modalities are
Computed Tomography and
Magnetic Resonance. The Limbus
Contour Software assists in the
following
scenarios:
• Operates in conjunction with
radiation treatment planning
systems or DICOM viewing
systems to load, save, and display
medical images and contours for
treatment
evaluation and treatment planning.
• Creation, transformation, and
modification of contours for | Limbus Contour is a software only
medical device intended for use
by trained radiation oncologists,
dosimetrists and physicists to
derive optimal contours for input to
radiation treatment planning.

Supported image modalities are
Computed Tomography and
Magnetic Resonance. The Limbus
Contour Software assists in the
following
scenarios:
• Operates in conjunction with
radiation treatment planning
systems or DICOM viewing
systems to load, save, and display
medical images and contours for
treatment
evaluation and treatment planning.
• Creation, transformation, and
modification of contours for | Same |
| Item | Limbus Contour v1.8 | Limbus Contour v1.7 - K230575 | Similarities /
Differences |
| | applications including, but not
limited to: transferring contours to
radiotherapy treatment planning
systems, aiding adaptive therapy
and archiving contours for patient
follow-up.
• Localization and definition of
healthy anatomical Structures.
Limbus Contour is not
intended for use with digital
mammography. | applications including, but not
limited to: transferring contours to
radiotherapy treatment planning
systems, aiding adaptive therapy
and archiving contours for patient
follow-up.
• Localization and definition of
healthy anatomical Structures.
Limbus Contour is not
intended for use with digital
mammography. | |
| Intended User | Healthcare providers | Healthcare providers | Same |
| Machine
Learning
Algorithm | Locked algorithm; Deep Learning
model | Locked algorithm; Deep Learning
model | Same |
| Contouring
Modes | Automatic | Automatic | Same |
| Supported
Image
Modalities | CT; MR | CT; MR | Same |
| Compatible
Scanner Models | No Limitation on scanner model,
DICOM 3.0 compliance required. | No Limitation on scanner model,
DICOM compliance required. | Same |
| Compatible
Treatment
Planning
System | No Limitation on TPS model | No Limitation on TPS model | Same |
| Result
Visualization | Limbus Contour has no data
visualization. Data processing is
automated and does not require user
interaction. A control interface is
provided for system administration
and configuration only.
Visualization software must be used
to facilitate the review and edit of the
generated contours. | Limbus Contour has no data
visualization. Data processing is
automated and does not require user
interaction. A control interface is
provided for system administration
and configuration only.
Visualization software must be used
to facilitate the review and edit of the
generated contours. | Same |
| Structures
Available for
Contouring | CT Structures
• A_Aorta
• A_Aorta_I
• A_Celiac
• A_LAD
• A_Mesenteric_S
• A_Pulmonary
• Bag_Bowel
• Bag_Bowel_Extend
• Bag_Bowel_Full
• Bag_Bowel_S
• Bladder
• Body
• Bone_Hyoid | CT Structures
• A_Aorta
• A_Aorta_Base
• A_Aorta_I
• A_Celiac
• A_LAD
• A_Mesenteric_S
• A_Pulmonary
• Atrium_L
• Atrium_R
• Bowel_Bag
• Bowel_Bag_Extend
• Bowel_Bag_Full
• Bowel_Bag_Superior | Similar; The
subject device
adds new
structures for
existing
supported
image
modalities
(CT/MR) |
| Item | Limbus Contour v1.8 | Limbus Contour v1.7 - K230575 | Similarities / Differences |
| | Bone_Ilium_L Bone_Ilium_R Bone_Ilium Bone_Mandible Bowel Bowel_Extend Bowel_Full Bowel_S BrachialPlex_L BrachialPlex_R BrachialPlexs Brain Brainstem Breast_L Breast_R Breasts Bronchus Canal_Anal CaudaEquina Cavity_Oral Chestwall_L Chestwall_R Chestwalls Clavicle_L Clavicle_R Cochlea_L Cochlea_R Colon_Sigmoid Cornea_L Cornea_R Duodenum Esophagus Eye_L Eye_R Eyes Femur_Head_L Femur_Head_R Femur_Heads Gallbladder Glnd_Lacrimal_L Glnd_Lacrimal_R Glnd_Submand_L Glnd_Submand_R Glnd_Thyroid GreatVes Heart Hippocampus_L Hippocampus_R Humerus_L Humerus_R Kidney_L Kidney_R Kidneys Larynx Lens_L Lens_R Lips Liver | Bowel Bowel_Extend Bowel_Full Bowel_Superior Bladder Body Body+Mask Bone_Hyoid Bone_Ilium_L Bone_Ilium_R Bone_Ilium Bone_Ischium_L Bone_Ischium_R Bone_Mandible Bone_Pelvic BoneMarrow_Pelvic BrachialPlex_L BrachialPlex_R BrachialPlexs Brain Brainstem Breast_Implant_L Breast_Implant_R Breast_L Breast_R Breasts Bronchus Canal_Anal Carina CaudaEquina Cavity_Oral Cerebellum Chestwall_L Chestwall_R Chestwall Clavicle_L Clavicle_R Cochlea_L Cochlea_R Colon_Sigmoid Cornea_L Cornea_R Duodenum Esophagus Eye_L Eye_R Eyes Femur_Head_L Femur_Head_R Femur_Heads Gallbladder Glnd_Lacrimal_L Glnd_Lacrimal_R Glnd_Submand_L Glnd_Submand_R Glnd_Thyroid GreatVes Heart | |
| Item | Limbus Contour v1.8 | Limbus Contour v1.7 - K230575 | Similarities /
Differences |
| | LN_Ax_Sclav_R LN_Ax_L1_L LN_Ax_L1_R LN_Ax_L2_L LN_Ax_L2_R LN_Ax_L3_L LN_Ax_L3_R LN_Sclav_L LN_Sclav_R LN_IMN_L LN_IMN_R LN_Neck_L LN_Neck_R LN_Neck_234_L LN_Neck_234_R LN_Neck_2347AB_L LN_Neck_2347AB_R LN_Neck_IA LN_Neck_IA6 LN_Neck_IB_L LN_Neck_IB_R LN_Neck_II_L LN_Neck_II_R LN_Neck_III_L LN_Neck_III_R LN_Neck_IV_L LN_Neck_IV_R LN_Neck_V_L LN_Neck_V_R LN_Neck_VI LN_Neck_VIIAB_L LN_Neck_VIIAB_R LN_Pelvis Lung_L Lung_R Lungs Musc_Constrict Musc_PecMinor_L Musc_PecMinor_R Musc_Sclmast_L Musc_Sclmast_R OpticChiasm OpticNrv_L OpticNrv_R Pancreas Parotid_L Parotid_R PelvisVessels PenileBulb Pituitary Prostate Prostate+SeminalVes PubicSymphys Rectum Retina_L Retina_R Ribs_L Ribs_R Ribs | Hippocampus_L Hippocampus_R Humerus_L Humerus_R InternalAuditoryCanal_L InternalAuditoryCanal_R Kidney_L Kidney_R Kidneys Larynx Lens_L Lens_R Lips Liver Lung_L Lung_R Lungs Mesorectum Musc_Constrict Musc_PecMinor_L Musc_PecMinor_R Musc_Sclmast_L Musc_Sclmast_R Optics OpticChiasm OpticNrv_L OpticNrv_R Pancreas Parotid_L Parotid_R PelvisVessels PenileBulb Pericardium Pericardium+A_Pulm Pituitary Prostate Prostate+SeminalVes ProstateBed PubicSymphys Rectum Retina_L Retina_R Ribs_L Ribs_R Ribs Sacrum SeminalVes Skin SpinalCanal SpinalCord Spleen Sternum Stomach Trachea Uterus+Cervix V_Venacava_I V_Venacava_S Vagina VB C1 | |
| Item | Limbus Contour v1.8 | Limbus Contour v1.7 - K230575 | Similarities / Differences |
| | Sacrum SeminalVes Skin SpinalCanal SpinalCord Spleen Sternum Stomach Trachea Uterus_Cervix V_Venacava_l V_Venacava_S Vagina Ventricle_L MR Structures Brainstem Cornea_L Cornea_R Eye_L Eye_R Hippocampus_L Hippocampus_R Optics PenileBulb Prostate Retina_L Retina_R SeminalVes | VB_C2 VB_C3 VB_C4 VB_C5 VB_C6 VB_C7 VB_L1 VB_L2 VB_L3 VB_L4 VB_L5 VB_T01 VB_T02 VB_T03 VB_T04 VB_T05 VB_T06 VB_T07 VB_T08 VB_T09 VB_T10 VB_T11 VB_T12 VBs Ventricle_L Ventricle_R Bladder_HDR Bowel_HDR Canal_Anal_HDR Colon_Sigmoid_HDR Rectum_HDR Urethra_HDR Bladder_CBCT Femur_Head_L_CBCT Femur_Head_R_CBCT LN_Pelvics_CBCT Prostate_CBCT Rectum_CBCT SeminalVes_CBCT MR Structures Bladder Brainstem Cornea_L Cornea_R Eye_L Eye_R Femur_Head_L Femur_Head_R Hippocampus_L Hippocampus_R Optics PenileBulb PubicSymphys Prostate Rectum Retina_L Retina_R | |
| Item | Limbus Contour v1.8 | Limbus Contour v1.7 - K230575 | Similarities /
Differences |
| | | • SeminalVes
• Bladder_HDR
• Bowel_HDR
• Canal_Anal_HDR
• Colon_Sigmoid_HDR
• Rectum_HDR
• Urethra_HDR | |
| Training
Characteristics
and datasets | The architecture for the neural
network models used in our device
borrows its primary structure from
the U-Net (Ronneberger 2015) and
ResUNet (Diakogiannisa 2020). We
use the Adam
optimization algorithm (Kingma
2014) and the Sørensen-Dice
coefficient loss function (Sørensen
1948) to train the network.
The models are trained with
examples of medical images and the
corresponding
human-generated contours for the
region of interest. During training, an
image is shown to the model, and
the model generates a contour. The
contour generated by the model is
compared to the ground truth
human-generated contour. The error
between the generated and ground
truth contours is used to adjust the
parameters (weights) of the model.
Backpropagation is the algorithm
that uses the error to adjust the
model parameters.
The scans used in training and
validating the models were collected
from a variety of anonymized and
pseudo-anonymized datasets that
included human-generated contours.
Datasets are from publicly available
clinical trials, and from the
company's clinical and research
partners.
Anatomical segmentation accuracy
investigation by Wong 2020
demonstrated that Limbus Contour
outputs are highly comparable to
expert contours (for the evaluated
Head and Neck and Prostate
treatment sites). Specifically, for | architecture for the neural network
models used in our device borrows
its primary structure from the U-Net
(Ronneberger 2015) and ResUNet
(Diakogiannisa 2020). We use the
Adam
optimization algorithm (Kingma 2014)
and the Sørensen-Dice coefficient
loss function (Sørensen 1948) to
train the network.
The models are trained with
examples of medical images and the
corresponding
human-generated contours for the
region of interest. During training, an
image is shown to the model, and the
model generates a contour. The
contour generated by the model is
compared to the ground truth
human-generated contour. The error
between the generated and ground
truth contours is used to adjust the
parameters (weights) of the model.
Backpropagation is the algorithm that
uses the error to adjust the model
parameters.
The scans used in training and
validating the models were collected
from a variety of anonymized and
pseudo-anonymized datasets that
included human-generated contours.
Datasets are from publicly available
clinical trials, and from the
company's clinical and research
partners.
Anatomical segmentation accuracy
investigation by Wong 2020
demonstrated that Limbus Contour
outputs are highly comparable to
expert contours (for the evaluated
Head and Neck and Prostate
treatment sites). Specifically, for most | Same |
| Item | Limbus Contour v1.8 | Limbus Contour v1.7 - K230575 | Similarities /
Differences |
| | most of the anatomical structures
examined, the Dice Scores (DSC)
between Limbus Contour and expert
contours were not significantly
different from the inter-observer
expert variability baselines. | of the anatomical structures
examined, the Dice Scores (DSC)
between Limbus Contour and expert
contours were not significantly
different from the inter-observer
expert variability baselines. | |
| | The investigation by Wong, Huang,
Wells et al., 2021 evaluated
implementation of Limbus Contour
prospectively at two Canadian
institutions. Limbus Contour was
used to generate OAR and CTVs for
all patients undergoing RT for a
central nervous system (CNS), head
and neck (H&N), or prostate cancer.
Automatic contours were generated
on approximately 551 eligible cases.
203 surveys were collected on 27
CNS, 54 H&N, and 93 prostate RT
plans, resulting in an overall survey
compliance rate of 32%. The
majority of OAR automatic contours
required minimal edits subjectively
(mean editing score ≤ 2) and
objectively (mean DSC and 95% HD
was ≥ 0.90 and ≤ 2.0 mm,
respectively).
Overall, thousands of total structures
were analyzed across all
investigations covering all supported
anatomical sites to provide a
substantial sample of device outputs
to draw conclusions that anatomical
structure contour outputs offer good
geometric alignment with expert
contours, require minimal editing,
and significantly reduce the time
needed for contouring. | The investigation by Wong, Huang,
Wells et al., 2021 evaluated
implementation of Limbus Contour
prospectively at two Canadian
institutions. Limbus Contour was
used to generate OAR and CTVs for
all patients undergoing RT for a
central nervous system (CNS), head
and neck (H&N), or prostate cancer.
Automatic contours were generated
on approximately 551 eligible cases.
203 surveys were collected on 27
CNS, 54 H&N, and 93 prostate RT
plans, resulting in an overall survey
compliance rate of 32%. The majority
of OAR automatic contours required
minimal edits subjectively (mean
editing score ≤ 2) and objectively
(mean DSC and 95% HD was ≥ 0.90
and ≤ 2.0 mm, respectively).
Overall, thousands of total structures
were analyzed across all
investigations covering all supported
anatomical sites to provide a
substantial sample of device outputs
to draw conclusions that anatomical
structure contour outputs offer good
geometric alignment with expert
contours, require minimal editing,
and significantly reduce the time
needed for contouring. | |
| Computer
Platform &
Operating
System | Operating System · Windows 10 /
Windows Server 2016 and Above

Hardware Requirements · 2 GHz or
faster multicore processor · 8 GB of
RAM · For GPU versions, a CUDA
capable NVIDIA GPU is required | Operating System · Windows 10 /
Windows Server 2016 and Above

6

7

8

9

10

11

12

Training and Validation Datasets

The scans used in training and validating the models were collected from a variety of anonymized and pseudo-anonymized datasets that included human-generated contours as ground truths.

Datasets are from publicly available clinical trials, and from the company's clinical and research partners. All data is acquired, stored, and managed in accordance with the company's dataset management procedure.

The fundamental criteria for including a dataset for training and validation is that the scans were generated for use in the course of radiotherapy treatment. Training dataset ground truth contours were reviewed and edited when necessary by our in-house clinicians and radiation oncologist to ensure consistency with established standards and guidelines for contouring.

To reduce bias in the training data, the following selection criteria are in place:

· Datasets from multiple clinical sites (institutions) are included
Datasets include clinical sites of multiple country origin. Including but not limited to: ●
- United States о
- Canada O
- United Kingdom O
- France O
- O Germany
- Italy O
- Netherlands O
- Switzerland O
- O Australia
- New Zealand O
- O Singapore
Ground truth annotations are consistent with recognized consensus contouring ● guidelines (RTOG 1106, RTOG 0848, EMBRACE II, DAHANCA, NRG, ESTRO, ACROP, EPTN)
Datasets include different makes and models of imaging devices (CT/MR scans) ●

13

O Sampling data from many different institutions and countries ensures a variety of imaging devices are captured
O lmaging devices included in training datasets include the following manufacturers: GE, Siemens, Phillips, Toshiba, Elekta
All imaqing data was captured as part of standard clinic protocols for image о acquisition in the course of radiation treatment planning. All imaging data is DICOM format and conforms to the DICOM standard (version 3.0 and newer)
The scans and ground truth contours in the datasets are from the general population of patients receiving radiotherapy treatment. The datasets are not restricted on the basis of age, ethnicity, race, gender, or disease states. A mix of healthy patients with disease were used as training data.
- O By sampling a large volume of data from a variety of institutions, countries, and including publically available datasets; a general population of patients receiving radiotherapy treatment is created. Multiple scan locations were used in the training data. For example, the scans used to train the parotid gland model were head and neck scans, while the scans used to train the bladder model were pelvic scans. The scans come from many different makes and models of imaging devices for both the CT and MRI scans.
- O The total number of unique scans included in training datasets exceeds 10,000 scans.

The following table shows the number of patients used in the training and validation data for each model. For each model, the training and validation sets are kept completely separate from the testing sets.

Structure	Modality	Body site	Number of training scans	Number of validation scans
Canal_Anal	CT	Pelvis	1084	50
Canal_Anal (Female)	CT	Pelvis	383	43
Canal_Anal_HDR	CT	Pelvis	466	49
Canal_Anal_HDR	MRI	Pelvis	304	34
A_Aorta	CT	Thorax	715	50
A_Aorta_Base	CT	Thorax	331	37
A_Aorta_I	CT	Abdomen	352	40
Applicator_Cylinder (beta)	CT	Pelvis	217	24
Applicator_Ring (beta)	CT	Pelvis	163	19
Bladder	CT	Pelvis	1105	50
Bladder_CBCT	CT	Pelvis	356	39
Bladder (Female)	CT	Pelvis	378	43
Bladder_HDR	CT	Pelvis	712	49
Bladder_HDR	MRI	Pelvis	303	34

Bowel	CT	Pelvis	697	50
Bowel_Extend	CT	Pelvis	697	50
Bowel_Extend (Female)	CT	Pelvis	697	50
Bowel (Female)	CT	Pelvis	697	50
Bowel_Full	CT	Pelvis	697	50
Bowel_Full (Female)	CT	Pelvis	697	50
Bowel_HDR	CT	Pelvis	682	50
Bowel_HDR	MRI	Pelvis	305	36
Bowel_Superior	CT	Abdomen	453	50
Bladder	MRI	Pelvis	579	50
Bowel_Bag	CT	Pelvis	697	50
Bowel_Bag_Extend	CT	Pelvis	697	50
Bowel_Bag_Extend (Female)	CT	Pelvis	697	50
Bowel_Bag (Female)	CT	Pelvis	697	50
Bowel_Bag_Full	CT	Pelvis	697	50
Bowel_Bag_Full (Female)	CT	Pelvis	697	50
Bowel_Bag_Superior	CT	Abdomen	453	50
Brain	CT	Head Neck	452	47
Brainstem	CT	Head Neck	614	50
Brainstem	MRI	Head Neck	241	27
Breasts	CT	Thorax	660	50
Breast_L	CT	Thorax	660	50
Wire_Breast_L (beta)	CT	Thorax	660	50
Breast_R	CT	Thorax	660	50
Wire_Breast_R (beta)	CT	Thorax	660	50
LN_Ax_Sclav_L	CT	Thorax	343	39
LN_Ax_Sclav_R	CT	Thorax	343	39
LN_Ax_L1_L	CT	Thorax	343	39
LN_Ax_L1_R	CT	Thorax	343	39
LN_Ax_L2_L	CT	Thorax	343	39
LN_Ax_L2_R	CT	Thorax	343	39
LN_Ax_L3_L	CT	Thorax	343	39
LN_Ax_L3_R	CT	Thorax	343	39
LN_Sclav_L	CT	Thorax	343	39
LN_Sclav_R	CT	Thorax	343	39
A_Celiac	CT	Thorax	435	44
Carina	CT	Thorax	865	50
CaudaEquina	CT	Pelvis	663	50
Cerebellum	CT	Head Neck	135	16

Chestwall	CT	Thorax	223	25
Chestwall_L	CT	Thorax	223	25
CW2cm_L	CT	Thorax	223	25
Chestwall_R	CT	Thorax	223	25
CW2cm_R	CT	Thorax	223	25
OpticChiasm	CT	Head Neck	1254	140
Clavicle_L	CT	Head Neck	104	12
Clavicle_R	CT	Head Neck	104	12
Cochlea_L	CT	Head Neck	248	29
Cochlea_R	CT	Head Neck	248	29
Musc_Constrict	CT	Head Neck	186	21
Cornea_L	CT	Head Neck	729	50
Cornea_L	MRI	Head Neck	199	23
Cornea_R	CT	Head Neck	729	50
Cornea_R	MRI	Head Neck	199	23
Bone_llium	CT	Pelvis	214	24
Bone_Illium_L	CT	Pelvis	214	24
Bone_Ilium_R	CT	Pelvis	214	24
LN_Neck_L	CT	Head Neck	374	42
LN_Neck_R	CT	Head Neck	374	42
LN_Neck_IA	CT	Head Neck	112	13
LN_Neck_IA6	CT	Head Neck	112	13
LN_Neck_VI	CT	Head Neck	112	13
LN_Neck_234_L	CT	Head Neck	672	50
LN_Neck_234_R	CT	Head Neck	672	50
LN_Neck_2347AB_L	CT	Head Neck	672	50
LN_Neck_2347AB_R	CT	Head Neck	672	50
LN_Neck_IB_L	CT	Head Neck	672	50
LN_Neck_IB_R	CT	Head Neck	672	50
LN_Neck_II_L	CT	Head Neck	672	50
LN_Neck_II_R	CT	Head Neck	672	50
LN_Neck_III_L	CT	Head Neck	672	50
LN_Neck_III_R	CT	Head Neck	672	50
LN_Neck_IV_L	CT	Head Neck	672	50
LN_Neck_IV_R	CT	Head Neck	672	50
LN_Neck_V_L	CT	Head Neck	672	50
LN_Neck_V_R	CT	Head Neck	672	50
LN_Neck_VIIA_L	CT	Head Neck	672	50
LN_Neck_VIIA_R	CT	Head Neck	672	50
LN_Neck_VIIAB_L	CT	Head Neck	672	50

LN_Neck_VIIAB_R	CT	Head Neck	672	50
LN_Neck_VIIB_L	CT	Head Neck	672	50
LN_Neck_VIIB_R	CT	Head Neck	672	50
Mesorectum	CT	Pelvis	302	33
Duodenum	CT	Abdomen	382	37
Esophagus	CT	Thorax	1643	50
ESTRO_LN_IMN_L	CT	Thorax	409	45
ESTRO_LN_IMN_R	CT	Thorax	409	45
ESTRO_LN_IMN_L_Expand	CT	Thorax	409	45
ESTRO_LN_IMN_R_Expand	CT	Thorax	409	45
ESTRO_LN_Ax_IP_L	CT	Thorax	395	44
ESTRO_LN_Ax_IP_R	CT	Thorax	395	44
ESTRO_LN_Ax_L1_L	CT	Thorax	395	44
ESTRO_LN_Ax_L1_R	CT	Thorax	395	44
ESTRO_LN_Ax_L2+IP_Fill_L	CT	Thorax	395	44
ESTRO_LN_Ax_L2+IP_Fill_R	CT	Thorax	395	44
ESTRO_LN_Ax_L2_L	CT	Thorax	404	45
ESTRO_LN_Ax_L2_R	CT	Thorax	404	45
ESTRO_LN_Ax_L3_L	CT	Thorax	404	45
ESTRO_LN_Ax_L3_R	CT	Thorax	404	45
ESTRO_LN_Sclav_L	CT	Thorax	409	45
ESTRO_LN_Sclav_R	CT	Thorax	409	45
Eyes	CT	Head Neck	729	50
Femur_Heads	CT	Pelvis	511	50
Femur_Head_L	CT	Pelvis	511	50
Femur_Head_L_CBCT	CT	Pelvis	139	17
Femur_Head_L	MRI	Pelvis	67	8
Femur_Head_R	CT	Pelvis	511	50
Femur_Head_R_CBCT	CT	Pelvis	139	17
Femur_Head_R	MRI	Pelvis	67	8
Gallbladder	CT	Abdomen	539	50
Eye_L	CT	Head Neck	729	50
Eye_L	MRI	Head Neck	199	23
Eye_R	CT	Head Neck	729	50
Eye_R	MRI	Head Neck	199	23
GreatVes	CT	Thorax	715	50
Heart	CT	Thorax	1000	50
Heart+A_Pulm	CT	Thorax	33	4

Pericardium+A_Pulm	CT	Thorax	33	4
Hippocampus_L	CT	Head Neck	104	12
Hippocampus_L	MRI	Head Neck	144	17
Hippocampus_R	CT	Head Neck	104	12
Hippocampus_R	MRI	Head Neck	144	17
Humerus_L	CT	Thorax	114	13
Humerus_R	CT	Thorax	114	13
InternalAuditoryCanal_L	CT	Head Neck	253	27
InternalAuditoryCanal_R	CT	Head Neck	253	27
LN_IMN_L	CT	Thorax	574	50
LN_IMN_R	CT	Thorax	574	50
LN_IMN_L_Expand	CT	Thorax	574	50
LN_IMN_R_Expand	CT	Thorax	574	50
Breast_Implant_L	CT	Thorax	75	9
Breast_Implant_R	CT	Thorax	75	9
LN_Inguinal_L	CT	Pelvis	282	32
LN_Inguinal_R	CT	Pelvis	282	32
Bone_Ischium_L	CT	Pelvis	234	25
Bone_Ischium_R	CT	Pelvis	234	25
V_Venacava_I	CT	Thorax	1000	50
Kidneys	CT	Abdomen	1018	50
Kidney_L	CT	Abdomen	1018	50
Kidney_R	CT	Abdomen	999	50
Atrium_L	CT	Thorax	331	37
A_LAD	CT	Thorax	588	50
Glnd_Lacrimal_L	CT	Head Neck	281	32
Glnd_Lacrimal_R	CT	Head Neck	281	32
Larynx	CT	Head Neck	365	30
Lens_L	CT	Head Neck	670	50
Lens_R	CT	Head Neck	670	50
Liver	CT	Abdomen	1000	50
Lips	CT	Head Neck	265	30
Bone_Hyoid	CT	Head Neck	326	35
Musc_PecMinor_L	CT	Thorax	398	42
Musc_PecMinor_R	CT	Thorax	398	42
Sternum	CT	Thorax	261	30
PubicSymphys	CT	Pelvis	540	50
PubicSymphys	MRI	Pelvis	617	49
Lungs	CT	Thorax	572	50
Lung_L	CT	Thorax	572	50
Lung_R	CT	Thorax	567	50
Ventricle_L	CT	Thorax	331	37
Bone_Mandible	CT	Head Neck	521	50
OpticNrv_L	CT	Head Neck	1254	140
OpticNrv_R	CT	Head Neck	1254	140
Optics	CT	Head Neck	1254	140
Optics	MRI	Head Neck	214	24
Cavity_Oral	CT	Head Neck	243	27
Pancreas	CT	Abdomen	765	50
Parotid_L	CT	Head Neck	688	50
Parotid_R	CT	Head Neck	688	50
Bone_Pelvic	CT	Pelvis	35	4
LN_Pelvics	CT	Pelvis	921	50
LN_Pelvics_CBCT	CT	Pelvis	189	24
BoneMarrow_Pelvic	CT	Pelvis	35	4
PelvisVessels	CT	Pelvis	921	50
PenileBulb	CT	Pelvis	657	50
PenileBulb	MRI	Pelvis	293	34
Pericardium	CT	Thorax	225	24
Pituitary	CT	Head Neck	295	33
BrachialPlexs	CT	Thorax	774	50
BrachialPlex_L	CT	Thorax	774	50
BrachialPlex_R	CT	Thorax	774	50
Prostate	CT	Pelvis	888	50
Prostate_CBCT	CT	Pelvis	267	29
ProstateBed	CT	Pelvis	150	16
ProstateFiducials (beta)	CT	Pelvis	888	50
Prostate	MRI	Pelvis	1698	50
Prostate+SeminalVes	CT	Pelvis	902	50
Bronchus	CT	Thorax	865	50
A_Pulmonary	CT	Thorax	670	75
Atrium_R	CT	Thorax	331	37
Rectum	CT	Pelvis	1084	50
Rectum_CBCT	CT	Pelvis	369	42
Rectum (Female)	CT	Pelvis	215	24
Rectum_HDR	CT	Pelvis	334	35
Rectum_HDR	MRI	Pelvis	304	34
Rectum	MRI	Pelvis	571	50
Retina_L	CT	Head Neck	729	50
Retina_L	MRI	Head Neck	199	23

Retina_R	CT	Head Neck	729	50
Retina_R	MRI	Head Neck	199	23
Ribs	CT	Thorax	84	11
Ribs_L	CT	Thorax	84	11
Ribs_R	CT	Thorax	84	11
Ventricle_R	CT	Thorax	331	37
Sacrum	CT	Pelvis	195	22
Sacrum	MRI	Pelvis	333	35
Musc_Sclmast_L	CT	Head Neck	86	10
Musc_Sclmast_R	CT	Head Neck	86	10
Colon_Sigmoid	CT	Pelvis	497	50
Colon_Sigmoid (Female)	CT	Pelvis	497	50
Colon_Sigmoid_HDR	CT	Pelvis	494	48
Colon_Sigmoid_HDR	MRI	Pelvis	240	24
A_Mesenteric_S	CT	Thorax	428	43
GInd_Submand_L	CT	Head Neck	573	50
GInd_Submand_R	CT	Head Neck	573	50
Spleen	CT	Abdomen	137	16
SeminalVes	CT	Pelvis	902	50
SeminalVes_CBCT	CT	Pelvis	235	26
SeminalVes	MRI	Pelvis	589	50
V_Venacava_S	CT	Thorax	384	43
SpaceOARVue (beta)	CT	Pelvis	36	5
SpinalCord	CT	Head
Neck/Thorax	1000	49
SpinalCanal	CT	Head
Neck/Thorax	1000	49
Stomach	CT	Abdomen	1414	50
Lobe_Temporal_L	CT	Head Neck	211	23
Lobe_Temporal_R	CT	Head Neck	211	23
GInd_Thyroid	CT	Thorax	685	50
Trachea	CT	Thorax	865	50
Urethra_HDR	CT	Pelvis	244	25
Urethra_HDR	MRI	Pelvis	236	24
Uterus+Cervix	CT	Pelvis	360	40
Vagina	CT	Pelvis	95	11
VB_C1	CT	Head Neck	132	13
VB_C2	CT	Head Neck	132	13
VB_C3	CT	Head Neck	132	13
VB_C4	CT	Head Neck	132	13
VB_C5	CT	Head Neck	132	13
VB_C6	CT	Head Neck	132	13
VB_C7	CT	Head Neck	132	13
VB_L1	CT	Pelvis	219	24
VB_L2	CT	Pelvis	219	24
VB_L3	CT	Pelvis	219	24
VB_L4	CT	Pelvis	219	24
VB_L5	CT	Pelvis	219	24
VB_T01	CT	Thorax	230	26
VB_T02	CT	Thorax	230	26
VB_T03	CT	Thorax	230	26
VB_T04	CT	Thorax	230	26
VB_T05	CT	Thorax	230	26
VB_T06	CT	Thorax	230	26
VB_T07	CT	Thorax	230	26
VB_T08	CT	Thorax	230	26
VB_T09	CT	Thorax	230	26
VB_T10	CT	Thorax	230	26
VB_T11	CT	Thorax	230	26
VB_T12	CT	Thorax	230	26
VBs	CT	All	260	28

Models underwent additional final performance testing on a separate set of scans (see VII. PERFORMANCE DATA). Validation scans are used only to determine at what epoch to stop model training.

14

15

16

17

18

19

20

VII. PERFORMANCE DATA

The following performance data were provided in support of the substantial equivalence determination.

Sterilization & Shelf-life Testing

The subject device is a software-only device. Therefore sterilization and shelf-life are not applicable.

Biocompatibility

The subject device is a software-only device. There are no direct patient-contacting components of the subject device. Therefore, patient contact information is not needed for this device

Electrical safety and electromagnetic compatibility (EMC)

The subject device is a software-only device. It contains no electric components, generates no electrical emissions, and uses no electrical energy of any type; therefore, this section is not applicable.

Software Verification and Validation Testing

Software verification and validation testing were conducted as recommended by FDA's Guidance for Industry and FDA Staff, "Content of Premarket Submissions for Device Software Functions". The enhanced level of documentation was provided.

21

Two different types of verification testing were conducted to verify the software requirements: Manual and Automated. All tests passed, demonstrating that the software performance is in accordance with the stated software requirements.

Validation testing of the following functions of the Limbus Contour application demonstrated that the software meets user needs and intended uses and to support substantial equivalence:

· Automatic Contouring Validation Test (Benchtop performance testing)

Automatic Contouring - Validation Test

Each structure undergoes performance testing prior to release. The objective of the testing is to verify that the software outputs meet the input requirements and that the software can consistently and accurately fulfill its intended use by contouring images to a level deemed sufficient based on published data, and expert opinion. This is achieved through testing the system functionality with controlled test datasets of medical scans with manual segmentations of structures that have been previously treated clinically and reviewed by a medical expert.

For each structure, a set of at least 10 patient scans is used for testing. A single scan may be used to test multiple structures. These test scans are randomly selected from our total pool of patient scans that contain the structure and are not included in training or validation of the models. Our total pool of patient scans that contain the structures were selected to reflect the general population of patients receiving radiation treatments.

Each of the 10 scans selected randomly for performance testing is run through the Limbus Contour software: this constitutes a single trial. Note that the same scan may appear in the test set of multiple structures, but each scan need only be run through the software once. The output of each trial is a DICOM RT-Structure file generated by the software that may contain multiple structure contours. Structure contours will be generated for scans of the appropriate body site only. For example, for pelvic scans only pelvic structures (like BLADDER) will be generated (not thorax or head and neck structures).

Test inputs:

A patient test scan on the computer file system -
-User interaction to select, import, choose structures, and start-contouring the patient test scans listed above.

Test outputs:

A DICOM RT-Structure file written to the test computer file system (may contain multiple structure contours)
Supplemental data required:
Each patient test scan has a corresponding DICOM RT-Structure file which contains ground truth contours for some structures. The ground truth contours are from multiple experts at multiple institutions. They have all been reviewed by a board certified radiation oncologist to ensure structures are labeled properly and contoured according to clinical trial guidelines.
Contours generated by Limbus Contour are compared to ground truth expert contours using the Sørensen-Dice Similarity Coefficient (Dice Similarity Coefficient, or DSC, for short) (Sørensen 1948). DSC is a measurement of the volumetric overlap of two contours. It is widely used in medical imaging, and specifically for evaluating the performance of autoseqmentation approaches (Balagopal 2018, Cardenas 2018, Dong 2019, van Harten 2019, Nikolov 2018, Trullo 2017, Liang 2019, Ren 2018, Wong 2019, Wong 2020).

22

Mean DSC values and standard deviations from published machine learning autosegmention models as references for the pass/fail criteria for benchtop testing. For each structure, we compare the lower edge of the 95% confidence interval to the mean minus the standard deviation of the reference model's DSC. If the lower edge of the confidence interval is less than the reference mean minus the standard deviation, the test fails.

The table below includes the Test DSC threshold from the reference studies. Note that there is a wide range of mean DSC scores across structures: large, easily identifiable structures (e.g. lungs) will have higher DSC scores than small, hard-to-identify structures (e.g. lacrimal glands).

Limbus Mean DSC and DSC Std Dev and DSC lower 95% conf edge are compared with the Test DSC Threshold. Full testing results for all structures are provided in the table below:

| Structure | Mean DSC | Limbus
DSC Std
Dev | Number
of Scans | Limbus DSC
lower 95% conf
edge | Test DSC
Threshold | Result |
|----------------------------|--------- | A_Aorta | 0.909095 | A_Aorta_Base | 0.979588 | A_Aorta_I | 0.938016 | A_Celiac | 0.781502 | A_LAD | 0.692766 | A_Mesenteric_S | 0.857257 | A_Pulmonary | 0.901867 | Applicator_Cylinder (beta) | 0.80111733 | Applicator_Ring (beta) | 0.963863 | Atrium_L | 0.977044 | Atrium_R | 0.978451 | Bladder | 0.96601238 | Bladder | 0.963518 | Bladder_CBCT | 0.959173 | Bladder_HDR | 0.931167 | Bladder_HDR | 0.896883 | Bone Marrow_Pelvic | 0.995414 | Bone_Hyoid | 0.85411417 | Bone_Illium_L | 0.9888075 | Bone_Illium_R | 0.99058833 | Bone_Ischium_L | 0.938985 | Bone_Ischium_R | 0.93923 | Bone_Mandible | 0.94024769 | Bone_Pelvic | 0.98383 | Bowel | 0.90743217 | Structure | Mean DSC | Limbus
DSC Std
Dev | Number
of Scans | Limbus DSC
lower 95% conf
edge | Test DSC
Threshold | Result |
| Bowel_Bag | 0.93979478 | Bowel_Bag_Extend | 0.971576 | Bowel_Bag_Full | 0.9679985 | Bowel_Bag_Superior | 0.93686455 | Bowel_Extend | 0.941682 | Bowel_Full | 0.92351381 | Bowel_HDR | 0.841368 | Bowel_HDR | 0.55831818 | Bowel_Superior | 0.90214273 | BrachialPlex_L | 0.691605 | BrachialPlex_R | 0.693809 | Brain | 0.992205 | Brainstem | 0.90334688 | Brainstem | 0.925526 | Breast_Implant_L | 0.992884 | Breast_Implant_R | 0.973663 | Breast_L | 0.954514 | Breast_R | 0.93952091 | Bronchus | 0.839515 | CW2cm_L | 0.998955 | CW2cm_R | 0.999376 | Canal_Anal | 0.87596095 | Canal_Anal_HDR | 0.942891 | Canal_Anal_HDR | 0.610295 | Carina | 1 | CaudaEquina | 0.882098 | Cavity_Oral | 0.913113 | Cerebellum | 0.983219 | Chestwall_L | 0.95907091 | Chestwall_R | 0.95957182 | Clavicle_L | 0.98014375 | Clavicle_R | 0.981565 | Cochlea_L | 0.702311 | Cochlea_R | 0.686758 | Colon_Sigmoid | 0.81625381 | Colon_Sigmoid_HDR | 0.865505 | Structure | Limbus Mean DSC | Colon_Sigmoid_HDR | 0.753036 | Cornea_L | 0.96183182 | Cornea_L | 0.913718 | Cornea_R | 0.96934727 | Cornea_R | 0.927223 | Duodenum | 0.828433 | ESTRO_LN_Ax_IP_L | 0.984552 | ESTRO_LN_Ax_IP_R | 0.988801 | ESTRO_LN_Ax_L1_L | 0.997122 | ESTRO_LN_Ax_L1_R | 0.967503 | ESTRO_LN_Ax_L2+IP_Fill_L | 0.992986 | ESTRO_LN_Ax_L2+IP_Fill_R | 0.994206 | ESTRO_LN_Ax_L2_L | 0.995192 | ESTRO_LN_Ax_L2_R | 0.997352 | ESTRO_LN_Ax_L3_L | 0.993382 | ESTRO_LN_Ax_L3_R | 0.992149 | ESTRO_LN_IMN_L | 0.980597 | ESTRO_LN_IMN_L_Expand | 0.982079 | ESTRO_LN_IMN_R | 0.974402 | ESTRO_LN_IMN_R_Expand | 0.977852 | ESTRO_LN_Sclav_L | 0.97586 | ESTRO_LN_Sclav_R | 0.98735 | Esophagus | 0.83741083 | Eye_L | 0.93511824 | Eye_L | 0.950337 | Eye_R | 0.94191706 | Eye_R | 0.939666 | Femur_Head_L | 0.961299 | Femur_Head_L | 0.938162 | Femur_Head_L_CBCT | 0.977939 | Femur_Head_R | 0.961381 | Femur_Head_R | 0.948586 | Femur_Head_R_CBCT | 0.989667 | Gallbladder | 0.946422 | GInd Lacrimal L | 0.76574538 | Structure | Limbus Mean DSC | Glnd_Lacrimal_R | 0.73474077 | Glnd_Submand_L | 0.838183 | Glnd_Submand_R | 0.882672 | Glnd_Thyroid | 0.840575 | GreatVes | 0.956281 | Heart | 0.95488833 | Heart+A_Pulm | 0.995663 | Hippocampus_L | 0.897474 | Hippocampus_L | 0.801092 | Hippocampus_R | 0.841933 | Hippocampus_R | 0.804229 | Humerus_L | 0.981592 | Humerus_R | 0.983804 | InternalAuditoryCanal_L | 0.719663 | InternalAuditoryCanal_R | 0.778302 | Kidney_L | 0.97211 | Kidney_R | 0.971235 | LN_Ax_L1_L | 0.93347 | LN_Ax_L1_R | 0.957366 | LN_Ax_L2_L | 0.797847 | LN_Ax_L2_R | 0.836689 | LN_Ax_L3_L | 0.841469 | LN_Ax_L3_R | 0.833202 | LN_Ax_Sclav_L | 0.854859 | LN_Ax_Sclav_R | 0.839354 | LN_IMN_L | 0.681072 | LN_IMN_L_Expand | 0.974158 | LN_IMN_R | 0.754624 | LN_IMN_R_Expand | 0.969235 | LN_Inguinal_L | 0.987752 | LN_Inguinal_R | 0.975856 | LN_Neck_IA | 0.88038818 | LN_Neck_IA6 | 0.94597364 | LN_Neck_IB_L | 0.918553 | LN_Neck_IB_R | 0.916248 | LN_Neck_III_L | 0.924377 | Structure | Mean DSC | Limbus
DSC Std
Dev | Number
of Scans | Limbus DSC
lower 95% conf
edge | Test DSC
Threshold | Result |
| LN_Neck_III_R | 0.903805 | LN_Neck_II_L | 0.921425 | LN_Neck_II_R | 0.919918 | LN_Neck_IV_L | 0.837067 | LN_Neck_IV_R | 0.813474 | LN_Neck_L | 0.86875 | LN_Neck_R | 0.86855 | LN_Neck_VI | 0.93822083 | LN_Neck_VIIAB_L | 0.704562 | LN_Neck_VIIAB_R | 0.684087 | LN_Neck_VIIA_L | 0.973697 | LN_Neck_VIIA_R | 0.963045 | LN_Neck_VIIB_L | 0.979443 | LN_Neck_VIIB_R | 0.9727 | LN_Neck_V_L | 0.899668 | LN_Neck_V_R | 0.855186 | LN_Pelvics | 0.90169318 | LN_Pelvics_CBCT | 0.974742 | LN_Sclav_L | 0.96093 | LN_Sclav_R | 0.958498 | Larynx | 0.898777 | Lens_L | 0.78292471 | Lens_R | 0.76047471 | Lips | 0.824696 | Liver | 0.97773385 | Lobe_Temporal_L | 0.944744 | Lobe_Temporal_R | 0.948456 | Lung_L | 0.983115 | Lung_R | 0.983649 | Mesorectum | 0.827965 | Musc_Constrict | 0.869097 | Musc_PecMinor_L | 0.869259 | Musc_PecMinor_R | 0.863584 | Musc_Sclmast_L | 0.946117 | Musc_Sclmast_R | 0.945291 | OpticChiasm | 0.65929882 | Structure | Mean DSC | Limbus
DSC Std
Dev | Number
of Scans | Limbus DSC
lower 95% conf
edge | Test DSC
Threshold | Result |
| OpticNrv_L | 0.82576941 | OpticNrv_R | 0.82894294 | Optics | 0.764846 | Pancreas | 0.884343 | Parotid_L | 0.88352083 | Parotid_R | 0.88281667 | PelvisVessels | 0.914998 | PenileBulb | 0.84850818 | PenileBulb | 0.73231 | Pericardium | 0.984828 | Pericardium+A_Pulm | 0.994973 | Pituitary | 0.75041867 | Prostate | 0.934093 | Prostate | 0.915164 | ProstateBed | 0.74691333 | ProstateFiducials (beta) | 0.61422 | Prostate_CBCT | 0.961269 | PubicSymphys | 0.943743 | PubicSymphys | 0.779585 | Rectum | 0.88681762 | Rectum | 0.934619 | Rectum_CBCT | 0.963103 | Rectum_HDR | 0.918553 | Rectum_HDR | 0.781 | Retina_L | 0.90761364 | Retina_L | 0.953271 | Retina_R | 0.91191636 | Retina_R | 0.92854 | Ribs_L | 0.94473545 | Ribs_R | 0.94621636 | Sacrum | 0.97012438 | Sacrum | 0.966632 | SeminalVes | 0.82148 | SeminalVes | 0.833995 | SeminalVes_CBCT | 0.904653 | SpaceOARVue (beta) | 0.866934 | Structure | Mean DSC | Limbus
DSC Std
Dev | Number
of Scans | Limbus DSC
lower 95% conf
edge | Test DSC
Threshold | Result |
| SpinalCanal | 0.8971765 | SpinalCord | 0.87788679 | Spleen | 0.98238429 | Sternum | 0.968506 | Stomach | 0.92353818 | Trachea | 0.900195 | Urethra_HDR | 0.68898 | Urethra_HDR | 0.558433 | Uterus+Cervix | 0.923876 | VB_C1 | 0.871119 | VB_C2 | 0.890465 | VB_C3 | 0.882984 | VB_C4 | 0.847607 | VB_C5 | 0.735888 | VB_C6 | 0.662313 | VB_C7 | 0.686772 | VB_L1 | 0.7397 | VB_L2 | 0.85353818 | VB_L3 | 0.89139273 | VB_L4 | 0.88930818 | VB_L5 | 0.971761 | VB_T01 | 0.749499 | VB_T02 | 0.900861 | VB_T03 | 0.846845 | VB_T04 | 0.871065 | VB_T05 | 0.868184 | VB_T06 | 0.856586 | VB_T07 | 0.895207 | VB_T08 | 0.90946273 | VB_T09 | 0.895233 | VB_T10 | 0.85180692 | VB_T11 | 0.90543538 | VB_T12 | 0.73466077 | VBs | 0.984448 | V_Venacava_l | 0.95366786 | V_Venacava_S | 0.851219 | Structure | Mean DSC | Limbus
DSC Std Dev | Number
of Scans | Limbus DSC
lower 95% conf edge | Test DSC
Threshold | Result |
| Vagina | 0.897341 | Ventricle_L | 0.951144 | Ventricle_R | 0.980784 | Wire_Breast_L (beta) | 0.750245 | Wire_Breast_R (beta) | 0.896053 Limbus
-----------|--------------------------|--------------------|--------------------------------------|-----------------------|--------|
| 0.0455771 | 10 | 0.87649337 | 0.81 | Passed |
| 0.0286193 | 10 | 0.95911641 | 0.81 | Passed |
| 0.10304303 | 10 | 0.86430858 | 0.81 | Passed |
| 0.27272084 | 10 | 0.58642282 | 0.26 | Passed |
| 0.06590144 | 10 | 0.64562622 | 0.26 | Passed |
| 0.14185425 | 10 | 0.75578763 | 0.26 | Passed |
| 0.03499015 | 10 | 0.87683829 | 0.85 | Passed |
| 0.33037573 | 15 | 0.60816274 | 0.374 | Passed |
| 0.07306595 | 10 | 0.9115984 | 0.374 | Passed |
| 0.0180652 | 10 | 0.96412183 | 0.79 | Passed |
| 0.01852677 | 10 | 0.96519867 | 0.78 | Passed |
| 0.05220935 | 21 | 0.94024138 | 0.935 | Passed |
| 0.01177413 | 10 | 0.95509588 | 0.88 | Passed |
| 0.04229406 | 10 | 0.92891975 | 0.91 | Passed |
| 0.06094679 | 10 | 0.88757132 | 0.56674243 | Passed |
| 0.13171833 | 10 | 0.80266393 | 0.79 | Passed |
| 0.00407252 | 10 | 0.9925009 | 0.805 | Passed |
| 0.04163051 | 12 | 0.82693015 | 0.77 | Passed |
| 0.01103973 | 12 | 0.98159874 | 0.76 | Passed |
| 0.00575056 | 12 | 0.98683332 | 0.76 | Passed |
| 0.01573502 | 10 | 0.92772963 | 0.76 | Passed |
| 0.01613541 | 10 | 0.92768822 | 0.76 | Passed |
| 0.01266685 | 13 | 0.93230094 | 0.922 | Passed |
| 0.00511637 | 10 | 0.98017022 | 0.929 | Passed |
| 0.06592406 | 23 | 0.87633846 | 0.74 | Passed |
Limbus
| 0.03659061 | 23 | 0.92253647 | 0.752 | Passed |
| 0.01582803 | 20 | 0.9635702 | 0.752 | Passed |
| 0.01354157 | 20 | 0.9611492 | 0.752 | Passed |
| 0.09357214 | 11 | 0.8730466 | 0.752 | Passed |
| 0.03040818 | 20 | 0.92630159 | 0.74 | Passed |
| 0.03444493 | 21 | 0.90651148 | 0.74 | Passed |
| 0.05558462 | 10 | 0.80160792 | 0.2008343 | Passed |
| 0.24969816 | 11 | 0.38801937 | 0.31 | Passed |
| 0.03756911 | 11 | 0.87651989 | 0.74 | Passed |
| 0.10786794 | 10 | 0.61444628 | 0.39 | Passed |
| 0.11005989 | 10 | 0.61508237 | 0.39 | Passed |
| 0.00251205 | 16 | 0.99078444 | 0.988 | Passed |
| 0.03859191 | 16 | 0.88152315 | 0.695 | Passed |
| 0.02877815 | 10 | 0.90494078 | 0.725 | Passed |
| 0.00662727 | 10 | 0.98814346 | 0.865 | Passed |
| 0.03491225 | 10 | 0.94869001 | 0.865 | Passed |
| 0.02763163 | 10 | 0.9347489 | 0.726 | Passed |
| 0.04383671 | 11 | 0.90962345 | 0.7345 | Passed |
| 0.06515951 | 10 | 0.79290593 | 0.76 | Passed |
| 0.00118886 | 10 | 0.9981046 | 0.72 | Passed |
| 0.00101477 | 10 | 0.99865013 | 0.72 | Passed |
| 0.13633659 | 21 | 0.808664 | 0.803 | Passed |
| 0.04773688 | 10 | 0.90874446 | 0.56167132 | Passed |
| 0.35031087 | 10 | 0.35971511 | 0.31 | Passed |
| 0 | 10 | 1 | 0.77 | Passed |
| 0.06633305 | 10 | 0.83464949 | 0.722 | Passed |
| 0.0386665 | 10 | 0.88545458 | 0.8 | Passed |
| 0.01399611 | 10 | 0.97320748 | 0.83 | Passed |
| 0.00299448 | 11 | 0.95702862 | 0.72 | Passed |
| 0.00327572 | 11 | 0.95733772 | 0.72 | Passed |
| 0.01256694 | 16 | 0.97303715 | 0.93 | Passed |
| 0.01013648 | 16 | 0.97583282 | 0.93 | Passed |
| 0.10183115 | 10 | 0.62947045 | 0.533 | Passed |
| 0.14712802 | 10 | 0.58151627 | 0.545 | Passed |
| 0.15924956 | 21 | 0.73764681 | 0.704 | Passed |
| 0.12156688 | 10 | 0.77854734 | 0.30928644 | Passed |
| Limbus DSC Std Dev | Number of Scans | Limbus DSC lower 95% conf edge | Test DSC Threshold | Result |
| 0.15966944 | 10 | 0.6388233 | 0.47 | Passed |
| 0.06990272 | 11 | 0.91415686 | 0.489 | Passed |
| 0.03513108 | 10 | 0.88858848 | 0.489 | Passed |
| 0.05299966 | 11 | 0.93320051 | 0.498 | Passed |
| 0.02302511 | 10 | 0.91075297 | 0.498 | Passed |
| 0.18461132 | 10 | 0.69637919 | 0.649 | Passed |
| 0.0225043 | 10 | 0.96845451 | 0.79 | Passed |
| 0.01830441 | 10 | 0.97570773 | 0.796 | Passed |
| 0.00681545 | 10 | 0.99224686 | 0.66 | Passed |
| 0.01693458 | 10 | 0.95538957 | 0.66 | Passed |
| 0.01314147 | 10 | 0.98358582 | 0.73 | Passed |
| 0.01093726 | 10 | 0.9863825 | 0.73 | Passed |
| 0.01077199 | 10 | 0.98748672 | 0.73 | Passed |
| 0.00448639 | 10 | 0.99414286 | 0.73 | Passed |
| 0.00884358 | 10 | 0.98705612 | 0.51 | Passed |
| 0.01468854 | 10 | 0.98164218 | 0.51 | Passed |
| 0.02745317 | 10 | 0.96095955 | 0.39 | Passed |
| 0.05662552 | 10 | 0.94157436 | 0.39 | Passed |
| 0.04214952 | 10 | 0.94425214 | 0.39 | Passed |
| 0.06968747 | 10 | 0.92800405 | 0.39 | Passed |
| 0.03136266 | 10 | 0.95342606 | 0.7 | Passed |
| 0.02268151 | 10 | 0.97112575 | 0.7 | Passed |
| 0.02651585 | 12 | 0.82009643 | 0.67 | Passed |
| 0.03324782 | 17 | 0.91687796 | 0.894 | Passed |
| 0.01463563 | 10 | 0.93986803 | 0.847 | Passed |
| 0.03257919 | 17 | 0.92404361 | 0.902 | Passed |
| 0.03581356 | 10 | 0.9140483 | 0.849 | Passed |
| 0.00888921 | 10 | 0.95494049 | 0.93 | Passed |
| 0.04781144 | 10 | 0.90396214 | 0.77 | Passed |
| 0.01378171 | 10 | 0.96808085 | 0.88 | Passed |
| 0.01105991 | 10 | 0.95346976 | 0.937 | Passed |
| 0.02852155 | 10 | 0.92818433 | 0.77 | Passed |
| 0.01081208 | 10 | 0.98193304 | 0.88 | Passed |
| 0.05882969 | 10 | 0.9043407 | 0.809 | Passed |
| 0.0785035 | 13 | 0.71649497 | 0.489 | Passed |
| Limbus DSC Std Dev | Number of Scans | Limbus DSC lower 95% conf edge | Test DSC Threshold | Result |
| 0.09508335 | 13 | 0.67508872 | 0.498 | Passed |
| 0.08845188 | 10 | 0.77491273 | 0.725 | Passed |
| 0.0245712 | 10 | 0.86509605 | 0.595 | Passed |
| 0.03434333 | 10 | 0.81600897 | 0.716 | Passed |
| 0.01660489 | 10 | 0.9444034 | 0.81 | Passed |
| 0.02805647 | 12 | 0.93656793 | 0.89 | Passed |
| 0.01079707 | 10 | 0.98793977 | 0.89 | Passed |
| 0.14363431 | 10 | 0.79473135 | 0.45 | Passed |
| 0.07687695 | 10 | 0.74610136 | 0.618 | Passed |
| 0.23470004 | 10 | 0.67405037 | 0.45 | Passed |
| 0.07348396 | 10 | 0.75166539 | 0.618 | Passed |
| 0.03366976 | 10 | 0.95750778 | 0.93 | Passed |
| 0.02794829 | 10 | 0.96381239 | 0.93 | Passed |
| 0.27119782 | 10 | 0.52567325 | 0.41 | Passed |
| 0.29907667 | 10 | 0.5643703 | 0.41 | Passed |
| 0.0055787 | 10 | 0.96811951 | 0.83 | Passed |
| 0.00508737 | 10 | 0.96759597 | 0.85 | Passed |
| 0.03827463 | 10 | 0.90609188 | 0.66 | Passed |
| 0.01855924 | 10 | 0.94409044 | 0.66 | Passed |
| 0.03448156 | 10 | 0.77318209 | 0.73 | Passed |
| 0.03793359 | 10 | 0.80955483 | 0.73 | Passed |
| 0.02407574 | 10 | 0.82424745 | 0.51 | Passed |
| 0.05413932 | 10 | 0.79447576 | 0.51 | Passed |
| 0.07708553 | 10 | 0.79971917 | 0.66 | Passed |
| 0.0636715 | 10 | 0.79380932 | 0.66 | Passed |
| 0.05716488 | 10 | 0.64018155 | 0.39 | Passed |
| 0.08171958 | 10 | 0.9157034 | 0.39 | Passed |
| 0.0588019 | 10 | 0.71256258 | 0.39 | Passed |
| 0.09728747 | 10 | 0.89964457 | 0.39 | Passed |
| 0.01196273 | 10 | 0.97919497 | 0.779 | Passed |
| 0.01828094 | 10 | 0.96277951 | 0.779 | Passed |
| 0.10469436 | 11 | 0.80898467 | 0.41 | Passed |
| 0.03537206 | 11 | 0.92184923 | 0.896 | Passed |
| 0.02691603 | 10 | 0.89929977 | 0.896 | Passed |
| 0.01954066 | 10 | 0.90227042 | 0.896 | Passed |
| 0.02716647 | 10 | 0.90494463 | 0.752 | Passed |
Limbus
| 0.03651978 | 10 | 0.87768214 | 0.775 | Passed |
| 0.02096226 | 10 | 0.90643054 | 0.894 | Passed |
| 0.02031001 | 10 | 0.9053901 | 0.894 | Passed |
| 0.10669372 | 10 | 0.76074821 | 0.655 | Passed |
| 0.07643769 | 10 | 0.75879757 | 0.655 | Passed |
| 0.04264226 | 12 | 0.84090532 | 0.779 | Passed |
| 0.04499896 | 12 | 0.83916643 | 0.779 | Passed |
| 0.07273804 | 12 | 0.89072412 | 0.722 | Passed |
| 0.14161814 | 10 | 0.60326153 | 0.55 | Passed |
| 0.15673354 | 10 | 0.57197437 | 0.55 | Passed |
| 0.03639132 | 10 | 0.94766603 | 0.54 | Passed |
| 0.0518865 | 10 | 0.92593022 | 0.54 | Passed |
| 0.02540021 | 10 | 0.96127405 | 0.69 | Passed |
| 0.02326651 | 10 | 0.9560573 | 0.71 | Passed |
| 0.05719485 | 10 | 0.85875611 | 0.785 | Passed |
| 0.05671539 | 10 | 0.81461707 | 0.775 | Passed |
| 0.05091482 | 22 | 0.877139 | 0.779 | Passed |
| 0.04202674 | 10 | 0.94467997 | 0.58 | Passed |
| 0.05712461 | 10 | 0.92006835 | 0.7 | Passed |
| 0.02948648 | 10 | 0.93740611 | 0.7 | Passed |
| 0.05827018 | 10 | 0.85709592 | 0.77 | Passed |
| 0.08119035 | 17 | 0.73838241 | 0.616 | Passed |
| 0.07902615 | 17 | 0.71711973 | 0.449 | Passed |
| 0.14948194 | 10 | 0.71777049 | 0.68 | Passed |
| 0.01147248 | 13 | 0.9705364 | 0.92 | Passed |
| 0.07569022 | 10 | 0.89060224 | 0.83 | Passed |
| 0.06837365 | 10 | 0.89954784 | 0.83 | Passed |
| 0.00654768 | 10 | 0.9784314 | 0.96 | Passed |
| 0.00652109 | 10 | 0.97898441 | 0.96 | Passed |
| 0.05209883 | 10 | 0.79069833 | 0.779 | Passed |
| 0.05737849 | 10 | 0.82805376 | 0.61 | Passed |
| 0.04744788 | 10 | 0.83531919 | 0.79 | Passed |
| 0.06177418 | 10 | 0.81939649 | 0.796 | Passed |
| 0.02773018 | 10 | 0.9262814 | 0.803 | Passed |
| 0.03302699 | 10 | 0.92166656 | 0.803 | Passed |
| 0.1679447 | 17 | 0.56716174 | 0.41 | Passed |
Limbus
| 0.06203798 | 17 | 0.79173441 | 0.73 | Passed |
| 0.06130553 | 17 | 0.79530977 | 0.72 | Passed |
| 0.05410538 | 10 | 0.72614403 | 0.504 | Passed |
| 0.09900704 | 10 | 0.81352255 | 0.769 | Passed |
| 0.06794505 | 12 | 0.83915386 | 0.778 | Passed |
| 0.05035732 | 12 | 0.84993419 | 0.803 | Passed |
| 0.02637213 | 10 | 0.89613382 | 0.26 | Passed |
| 0.04605243 | 11 | 0.81709956 | 0.705 | Passed |
| 0.27283179 | 10 | 0.53715145 | 0.46 | Passed |
| 0.0185493 | 10 | 0.97155955 | 0.8688 | Passed |
| 0.01235765 | 10 | 0.98613348 | 0.89 | Passed |
| 0.15158537 | 15 | 0.66188586 | 0.41 | Passed |
| 0.02193268 | 10 | 0.91840439 | 0.88 | Passed |
| 0.03096645 | 10 | 0.89301348 | 0.8 | Passed |
| 0.1454049 | 15 | 0.6619902 | 0.5 | Passed |
| 0.24931989 | 10 | 0.43587968 | 0.41 | Passed |
| 0.04241179 | 10 | 0.93093154 | 0.79 | Passed |
| 0.02100908 | 10 | 0.92871505 | 0.76 | Passed |
| 0.11210947 | 10 | 0.69939229 | 0.54 | Passed |
| 0.08654191 | 21 | 0.84409976 | 0.803 | Passed |
| 0.02030278 | 10 | 0.92009628 | 0.77 | Passed |
| 0.02896341 | 10 | 0.94238526 | 0.87 | Passed |
| 0.09535355 | 10 | 0.85034592 | 0.56167132 | Passed |
| 0.15188698 | 10 | 0.67235415 | 0.58 | Passed |
| 0.1929304 | 11 | 0.7760315 | 0.489 | Passed |
| 0.04079533 | 10 | 0.92408981 | 0.489 | Passed |
| 0.1905022 | 11 | 0.78199031 | 0.498 | Passed |
| 0.05520214 | 10 | 0.88905351 | 0.498 | Passed |
| 0.00563264 | 11 | 0.94089389 | 0.81 | Passed |
| 0.00495204 | 11 | 0.94283898 | 0.81 | Passed |
| 0.01642714 | 16 | 0.96083483 | 0.82 | Passed |
| 0.04786265 | 10 | 0.9323955 | 0.77 | Passed |
| 0.16089309 | 10 | 0.70639202 | 0.5 | Passed |
| 0.05384245 | 10 | 0.79548111 | 0.39 | Passed |
| 0.05295257 | 10 | 0.86677565 | 0.621 | Passed |
| 0.0421535 | 10 | 0.8367813 | 0.5 | Passed |
Limbus
| 0.06232767 | 20 | 0.86565125 | 0.722 | Passed |
| 0.06353613 | 28 | 0.8507265 | 0.722 | Passed |
| 0.00724712 | 14 | 0.97800307 | 0.958 | Passed |
| 0.00927831 | 10 | 0.96186916 | 0.8 | Passed |
| 0.04262548 | 11 | 0.89446681 | 0.64 | Passed |
| 0.04891354 | 10 | 0.86520679 | 0.77 | Passed |
| 0.26984128 | 10 | 0.49596059 | 0.26 | Passed |
| 0.30664371 | 10 | 0.33908855 | 0.26 | Passed |
| 0.07561527 | 10 | 0.86978785 | 0.8525 | Passed |
| 0.09654266 | 10 | 0.80206134 | 0.389 | Passed |
| 0.08375338 | 10 | 0.83055561 | 0.389 | Passed |
| 0.07768781 | 10 | 0.82741335 | 0.389 | Passed |
| 0.13488571 | 10 | 0.75112228 | 0.389 | Passed |
| 0.25082081 | 10 | 0.55647407 | 0.389 | Passed |
| 0.36427326 | 10 | 0.40174571 | 0.389 | Passed |
| 0.36426675 | 10 | 0.42620937 | 0.389 | Passed |
| 0.36842487 | 12 | 0.49912477 | 0.389 | Passed |
| 0.3119903 | 11 | 0.64075498 | 0.389 | Passed |
| 0.29651247 | 11 | 0.68916569 | 0.389 | Passed |
| 0.29491517 | 11 | 0.68817053 | 0.389 | Passed |
| 0.01978041 | 10 | 0.95761193 | 0.389 | Passed |
| 0.20357809 | 10 | 0.60387813 | 0.389 | Passed |
| 0.10665309 | 10 | 0.82457127 | 0.389 | Passed |
| 0.23384828 | 10 | 0.67957164 | 0.389 | Passed |
| 0.16033768 | 10 | 0.7563743 | 0.389 | Passed |
| 0.10527773 | 10 | 0.79287808 | 0.389 | Passed |
| 0.1735606 | 10 | 0.73243685 | 0.389 | Passed |
| 0.09111821 | 10 | 0.83002949 | 0.389 | Passed |
| 0.09927246 | 11 | 0.84175706 | 0.389 | Passed |
| 0.19379417 | 10 | 0.75661063 | 0.389 | Passed |
| 0.25673522 | 13 | 0.69073999 | 0.389 | Passed |
| 0.17253913 | 13 | 0.79719022 | 0.389 | Passed |
| 0.38432697 | 13 | 0.49354712 | 0.389 | Passed |
| 0.01042268 | 10 | 0.97699258 | 0.579 | Passed |
| 0.05427303 | 14 | 0.92085737 | 0.72 | Passed |
| 0.0503676 | 10 | 0.81519069 | 0.8 | Passed |
Limbus
| 0.06081143 | 10 | 0.85384215 | 0.665 | Passed |
| 0.00689877 | 10 | 0.94620926 | 0.9 | Passed |
| 0.01685223 | 10 | 0.96872948 | 0.8 | Passed |
| 0.28619472 | 10 | 0.54552785 | 0.39 | Passed |
| 0.15011288 | 10 | 0.78867618 | 0.39 | Passed |

23

24

25

26

27

28

29

Mechanical and Acoustic Testing

Not Applicable (Standalone Software)

Animal Study

Animal performance testing was not required to demonstrate safety and effectiveness of the device.

Clinical Studies

Clinical testing was not required to demonstrate the safety and effectiveness of Limbus Contour. Instead, substantial equivalence is based upon benchtop performance testing.

VIII. CONCLUSIONS

The minor differences between the subject Limbus Contour software and the predicate Limbus Contour software do not constitute a different intended use. The technological characteristics of the Limbus Contour software are similar to those of the predicate Limbus Contour software.

Results of software verification and validation testing demonstrate that the Limbus Contour software performs in accordance with specifications and that the performance is comparable to that of the predicate device. Therefore, the Limbus Contour software can be found to be substantially equivalent to the predicate Limbus Contour software device.