Contour ProtégéAI

• K Number: K231765
• Device Name: Contour ProtégéAI
• Manufacturer: MIM Software Inc.
• Date Cleared: 2023-11-08 ( 145 days)
• Product Code: QKB CLA
• Regulation Number: 892.2050

Intended Use

Trained medical professionals use Contour ProtégéAI as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0. In addition, Contour ProtégéAI supports the following indications: · Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management. · Segmenting anatomical structures across a variety of CT anatomic locations. · And segmenting the prostate, the seminal vesicles, and the urethra within T2-weighted MR images. Appropriate image visualization software must be used to review and, if necessary, edit results automatically generated by Contour ProtégéAI.

Device Description

Contour ProtégéAI is an accessory to MIM software that automatically creates contours on medical images through the use of machine-learning algorithms. It is designed for use in the processing of medical images and operates on Windows, Mac, and Linux computer systems. Contour ProtégéAl is deployed on a remote server using the MIMcloud service for data management and transfer; or locally on the workstation or server running MIM software.

Predicate Devices

K071964

Reference Devices

K071964

AI/ML (Artificial Intelligence / Machine Learning)

Yes
The intended use and device description explicitly state the use of "machine-learning algorithms" for creating contours and segmenting anatomical structures.

Therapeutic

No
The device is an accessory to software that assists in image processing and contour creation for diagnostic and treatment planning purposes, but it does not directly apply therapy or treatment to a patient.

Diagnostic

No

The device is described as a tool to assist in the automated processing of digital medical images, creating contours and segmenting anatomical structures. It is used for applications like quantitative analysis, aiding adaptive therapy, and transferring contours to treatment planning systems. It does not provide a diagnosis, but rather processes data for other medical professionals to use in diagnosis or treatment planning.

SaMD (Software as a Medical Device)

Yes

The device description explicitly states that Contour ProtégéAI is an accessory to MIM software and operates on standard computer systems (Windows, Mac, Linux). It is deployed either remotely or locally on a workstation/server running MIM software. There is no mention of dedicated hardware components included with the device itself.

IVD (In Vitro Diagnostic)

Based on the provided information, this device is not an In Vitro Diagnostic (IVD). Here's why:

• IVDs are used to examine specimens derived from the human body. The intended use and device description clearly state that Contour ProtégéAI processes digital medical images (CT and MR) of patients. It does not analyze biological samples like blood, urine, or tissue.
• The purpose of an IVD is to provide information about a patient's health status, diagnosis, or condition based on the analysis of these specimens. Contour ProtégéAI's purpose is to assist trained medical professionals in the processing and analysis of medical images for tasks like contouring anatomical structures, which is a step in medical imaging workflows, particularly in radiation therapy planning.

The device is a medical image processing tool, not a diagnostic test performed on biological samples.

PCCP (Predetermined Change Control Plan) Authorized

No
The provided text does not contain any explicit statement that the FDA has reviewed, approved, or cleared a PCCP for this specific device.

Overview

Intended Use / Indications for Use

Trained medical professionals use Contour ProtégéAI as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0. In addition, Contour ProtégéAI supports the following indications:

· Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.

· Segmenting anatomical structures across a variety of CT anatomic locations.

· And segmenting the prostate, the seminal vesicles, and the urethra within T2-weighted MR images.

Appropriate image visualization software must be used to review and, if necessary, edit results automatically generated by Contour ProtégéAI.

Product codes

QKB

Device Description

Contour ProtégéAI is an accessory to MIM software that automatically creates contours on medical images through the use of machine-learning algorithms. It is designed for use in the processing of medical images and operates on Windows, Mac, and Linux computer systems. Contour ProtégéAl is deployed on a remote server using the MIMcloud service for data management and transfer; or locally on the workstation or server running MIM software.

Mentions image processing

Trained medical professionals use Contour ProtégéAI as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0.
Contour ProtégéAI is an accessory to MIM software that automatically creates contours on medical images through the use of machine-learning algorithms. It is designed for use in the processing of medical images and operates on Windows, Mac, and Linux computer systems.

Mentions AI, DNN, or ML

Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.
Contour ProtégéAI is an accessory to MIM software that automatically creates contours on medical images through the use of machine-learning algorithms.
A total of 550 CT images from 41 clinical sites across multiple continents was gathered for the training of the final 4.1.0 neural network models.

Input Imaging Modality

CT and MR

Anatomical Site

Segmenting anatomical structures across a variety of CT anatomical locations. And segmenting the prostate, the seminal vesicles, and the urethra within T2-weighted MR images.

Indicated Patient Age Range

Training data patient age ranges were 1-20: 0.2%, 20-40: 1.8%, 40-60: 8.9%, 60+: 24.2%, and not labeled 64.9%.
4.38% were between the ages of 20-40, 21.5% were between 40-60, and 47.9% were over the age of 60. 26.0% were unknown, and 0.3% were under the age 20.

Intended User / Care Setting

Trained medical professionals

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

A total of 550 CT images from 41 clinical sites across multiple continents was gathered for the training of the final 4.1.0 neural network models.
The CT images for this training set were obtained from clinical treatment plans for patients prescribed external beam or molecular radiotherapy, but the original segmentations were not used. Instead, the images were re-segmented by consultants (physicians and dosimetrists) specifically for this purpose, outside of clinical practice. Detailed instructions derived from relevant published clinical contouring quidelines were prepared for the dosimetrists. The initial segmentations were then reviewed and corrected by a radiation oncologist against the same standards and quidelines. Qualified staff at MIM Software (M.D. or licensed dosimetrists) then performed a final review and correction. All segmenters were instructed to spend additional time to ensure the highest quality training data. In particular, the consultants were asked to contour all specified OAR structures on all images according to referenced standards, whether or not they were proximal to the treatment field. All patients were imaged on an indexed couch in treatment position ("simulation CT"). Series that were non-axial, had slices thinner than 0.5mm, or had non-Fan Beam or MV acquisitions were excluded.

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

The proposed Contour ProtégéAl device were then evaluated on the test subjects from a pool of 754 independent images gathered from 27 institutions.
The verification data used for testing is from a set of institutions that are totally disjoint from the training datasets used to train each model in the Contour ProtégéAl device. The reference predicate MIM device was configured with Atlases built from the same training data used to train the models.
The images 48.4% were female, 35.2% were male, and 16.4% were unknown for sex. The manufacturer was GE for 37.9% of images, 25.1% was Siemens, 15.0% were Philips, 0.4% were from Toshiba, and the other 21.6% were unknown.

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

Noninferiority testing was used to compare the proposed Contour ProtégéAl device to Atlases created from the MIM Maestro (K071964) reference device.
The proposed Contour ProtégéAl device were then evaluated on the test subjects from a pool of 754 independent images gathered from 27 institutions.
We tested Contour ProtégéAl against the reference predicate device, and the goal of this testing is to show that it is equivalent or superior to the reference predicate. The performance of both segmentation devices was measured by calculating both the Dice score and MDA of the novel segmentations with the original ground-truth contours. User beta testing was also used to evaluate the performance of Contour ProtégéAl in the context of time savings compared to contouring from scratch. This user evaluation was made on a three-point scale for each contour, with one indicating negligible time savings, two indicating moderate, and three indicating significant time savings. Our acceptance criteria combine the statistical tests and the user evaluation - only structures that pass two or more of the following three tests could be included in the final models:

• Statistical non-inferiority of the Dice score compared with the reference predicate.
• Statistical non-inferiority of the MDA score compared with the reference predicate.
• Average user evaluation of 2 or higher, when measured on a three-point scale.

Further clinical validation was also conducted in-house to evaluate contours compared to detailed criteria based on established clinical quidelines.
In addition, each model as a whole was also evaluated. In order to be included in the released product, the cumulative Added Path Length of the contours in the model was evaluated relative to the ground-truth. Cumulative APL has been found to correlate well with the spent in editing and correcting auto-segmented contours (Vaassen et al, 2020)'. Each model was required to have statistically non-inferior cumulative APL compared to the reference predicate.
Finally, the localization accuracy of each structure in all models was measured. We did not impose a passing criterion on localization accuracy; the results however are included in this summary document and in our User Guide and Whitepaper to allow the user to better understand the performance of the device.

The mean and standard deviation Dice coefficients and MDA scores, along with the lower 95th percentile confidence bound, were calculated for both the proposed Contour ProtégéAl device and the MIM Maestro atlas seamentation reference device for each structure of each neural network model. Contour ProtégéAl results were equivalent or had better performance than the MIM Maestro atlas segmentation reference device. Equivalence is defined such that the lower 95th percentile confidence bound of the Contour ProtégéAl segmentation is greater than 0.1 Dice lower than the mean MIM atlas segmentation reference device performance.
Preliminary user evaluation conducted as part of testing demonstrated that Contour ProtégéAl yields comparable time-saving functionality when creating contours as other commercially available automatic segmentation products.
Contour ProtégéAl results were equivalent or had better performance than the MIM Maestro atlas segmentation reference device.

4.1.0 CT Model: Head and Neck	Structure: Bone_Mandible	Dice MIM Atlas: 0.81 ± 0.07	Dice Contour ProtégéAl: 0.86 ± 0.07 (0.83) *	MDA MIM Atlas: 1.00 ± 0.87	MDA Contour ProtégéAl: 0.64 ± 0.31 (0.93) *	External Evaluation Score: 2.86
	BrachialPlex_L	0.17 ± 0.08	0.22 ± 0.10 (0.14) *	3.24 ± 2.62	2.82 ± 2.61 (5.12) *	2.6
	BrachialPlex_R	0.15 ± 0.07	0.19 ± 0.09 (0.11) *	3.62 ± 2.79	2.73 ± 2.38 (5.20) *	2.6
	Brain	0.97 ± 0.01	0.98 ± 0.01 (0.97) *	0.64 ± 0.37	0.52 ± 0.36 (0.66) *	2.71
	Brainstem	0.78 ± 0.09	0.82 ± 0.08 (0.78) *	1.72 ± 0.81	1.20 ± 0.65 (1.47) *	2.71
	Cavity_Oral	0.75 ± 0.14	0.76 ± 0.13 (0.68) *	3.34 ± 2.13	3.20 ± 1.93 (4.35) *	2.71
	Cochlea_L	0.19 ± 0.15	0.30 ± 0.18 (0.22) *	1.26 ± 0.85	1.17 ± 0.87 (1.58) *	2.29
	Cochlea_R	0.17 ±0.15	0.32 ± 0.20 (0.23) *	1.12 ± 0.61	1.05 ± 0.68 (1.37) *	2.29
	Eye_L	0.80 ± 0.08	0.87 ± 0.06 (0.83) *	1.07 ± 0.60	0.65 ± 0.51 (0.91) *	2.57
	Eye_R	0.80 ± 0.10	0.86 ± 0.07 (0.82) *	1.11 ±0.64	0.66 ± 0.48 (0.92) *	2.57
	Glnd_Lacrimal_L	0.22 ± 0.17	0.40 ± 0.16 (0.27) *	0.86 ± 0.54	0.68 ± 0.35 (1.05) *	2.71
	Glnd_Lacrimal_R	0.23 ± 0.16	0.45 ± 0.14 (0.34) *	0.73 ± 0.38	0.71 ± 0.52 (1.09) *	2.71
	Glnd_Submand_L	0.57 ± 0.13	0.77 ± 0.11 (0.69) *	1.74 ± 0.72	0.76 ± 0.32 (1.12) *	3
	Glnd_Submand_R	0.56 ± 0.16	0.75 ± 0.11 (0.65) *	1.82 ± 0.87	0.81 ± 0.34 (1.28) *	3
	Glnd_Thyroid	0.47 ±0.18	0.71 ±0.19 (0.57) *	2.55 ± 2.14	1.38 ± 3.18 (3.38) *	2.71
	Lens_L	0.22 ± 0.23	0.61 ± 0.17 (0.52) *	0.77 ± 0.54	0.56 ± 0.47 (0.84) *	2.29
	Lens_R	0.22 ± 0.22	0.63 ± 0.16 (0.54) *	0.81 ± 0.47	0.55 ± 0.33 (0.78) *	2.29
	Lips	0.38 ± 0.14	0.37 ± 0.15 (0.28) *	4.14 ± 2.86	5.26 ± 3.41 (7.27)	2.83
	OpticChiasm	0.04 ± 0.07	0.13 ±0.13 (0.08) *	2.69 ± 1.87	2.46 ± 1.99 (3.55) *	2
	OpticNrv_L	0.45 ± 0.15	0.53 ± 0.14 (0.45) *	0.97 ± 0.81	0.77 ± 0.85 (1.19) *	2.57
	OpticNrv_R	0.44 ±0.14	0.52 ± 0.13 (0.44) *	1.11 ± 1.24	0.80 ± 0.83 (1.34) *	2.57
	Parotid_L	0.71 ±0.10	0.79 ± 0.09 (0.75) *	2.02 ± 0.95	1.30 ± 0.62 (1.68) *	3
	Parotid_R	0.71 ± 0.09	0.80 ± 0.06 (0.76) *	2.11 ±0.76	1.26 ± 0.43 (1.56) *	3
	Pituitary	0.38 ± 0.18	0.54 ± 0.15 (0.41) *	1.07 ± 0.59	0.87 ± 0.54 (1.29) *	3
	SpinalCord	0.66 ± 0.14	0.65 ± 0.16 (0.59) *	0.81 ±0.33	0.73 ± 0.41 (0.88) *	2.86
	LN_Neck_IA	0.48 ± 0.13	0.60 ± 0.14 (0.44) *	1.05 ± 1.51	0.47 ± 0.45 (1.81) *	2.75
	LN_Neck_IB_L	0.70 ±0.05	0.79 ± 0.04 (0.73) *	1.30 ± 0.47	0.73 ± 0.24 (1.16) *	2.75
	LN_Neck_IB_R	0.68 ±0.06	0.79 ±0.05 (0.72) *	1.41 ± 0.51	0.73 ± 0.20 (1.19) *	2.75
	LN_Neck_IIA_L	0.64 ± 0.04	0.75 ± 0.05 (0.70) *	2.02 ± 0.54	1.54 ± 0.44 (2.09) *	2.25
	LN_Neck_IIA_R	0.60 ± 0.08	0.76 ± 0.04 (0.69) *	2.24 ± 0.76	1.41 ±0.49 (2.15) *	2.25
	LN_Neck_IIB_L	0.68 ± 0.11	0.80 ± 0.06 (0.68) *	0.95 ± 0.49	0.63 ± 0.31 (1.12) *	3
	LN_Neck_IIB_R	0.68 ± 0.09	0.80 ± 0.06 (0.70) *	1.19 ± 0.64	0.58 ± 0.15 (1.14) *	3
	LN_Neck_III_L	0.54 ± 0.12	0.75 ± 0.07 (0.63) *	1.53 ± 0.60	0.92 ± 0.44 (1.51) *	3
	LN_Neck_III_R	0.57 ± 0.12	0.75 ± 0.07 (0.63) *	1.57 ± 0.61	0.99 ± 0.37 (1.57) *	3
	LN_Neck_IV_L	0.58 ± 0.08	0.69 ± 0.08 (0.59) *	1.62 ± 0.47	1.13 ± 0.37 (1.63) *	2.75
	LN_Neck_IV_R	0.55 ± 0.11	0.71 ± 0.09 (0.58) *	1.87 ± 0.75	1.06 ± 0.33 (1.77) *	2.75
	LN_Neck_V_L	0.53 ± 0.11	0.58 ± 0.09 (0.45) *	1.43 ± 0.82	1.13 ± 0.59 (2.02) *	3
	LN_Neck_V_R	0.50 ± 0.10	0.58 ± 0.11 (0.44) *	2.03 ± 1.66	1.43 ± 0.98 (3.18) *	3
	LN_Neck_VIA	0.34 ± 0.09	0.36 ± 0.11 (0.24) *	1.20 ± 0.55	0.80 ± 0.35 (1.33) *	2.75
	LN_Retropharynx_L	0.22 ± 0.07	0.28 ±0.10 (0.19) *	1.44 ± 0.87	1.54 ± 1.47 (2.93) *	2.75
	LN_Retropharynx_R	0.21 ± 0.07	0.28 ± 0.09 (0.18) *	1.64 ± 0.93	1.49 ± 1.36 (2.82) *	2.75
	LN_Retrostyloid_L	0.57 ± 0.10	0.64 ± 0.09 (0.53) *	1.58 ± 0.75	1.23 ± 0.48 (1.96) *	2.75
	LN_Retrostyloid_R	0.56 ± 0.11	0.66 ±0.11 (0.53) *	1.59 ± 0.74	1.13 ± 0.66 (1.95) *	2.75
4.1.0 CT Model: Thorax	BrachialPlex_L	0.28 ± 0.14	0.38 ± 0.14 (0.26) *	2.96 ± 1.79	2.94 ± 1.84 (4.49) *	3
	BrachialPlex_R	0.28 ± 0.11	0.37 ± 0.16 (0.24) *	3.37 ± 2.27	3.28 ± 2.60 (5.51)	3
	Breast_L	0.74 ±0.11	0.76 ± 0.15 (0.66) *	4.86 ± 3.45	3.64 ± 1.85 (5.65) *	2.8
	Breast_R	0.76 ± 0.12	0.78 ± 0.15 (0.67) *	4.02 ± 1.70	3.45 ± 1.51 (4.66) *	2.6
	Bronchus	0.60 ± 0.17	0.66 ± 0.13 (0.55) *	2.18 ± 1.45	1.90 ± 1.22 (2.78) *	2.6
	Carina	0.36 ± 0.18	0.49 ± 0.12 (0.40) *	1.27 ± 0.55	1.03 ± 0.47 (1.33) *	2.4
	Cricoid	0.02 ± 0.04	0.04 ± 0.05 (-0.01) *	4.64 ± 1.54	5.71 ± 1.24 (7.05)	2.4
	Esophagus	0.48 ± 0.16	0.70 ± 0.15 (0.64) *	2.49 ± 2.48	0.97 ± 0.54 (1.56) *	2.8
	Glnd_Thyroid	0.46 ± 0.18	0.65 ± 0.17 (0.54) *	2.92 ± 2.60	1.75 ± 1.78 (3.19) *	2.75
	GreatVes	0.70 ± 0.08	0.74 ±0.10 (0.65) *	3.37 ± 1.42	2.67 ± 1.73 (4.14) *	3
	Heart	0.88 ± 0.08	0.90 ± 0.09 (0.87) *	3.39 ± 1.91	2.55 ± 1.40 (3.06) *	3
	Humerus_Head_L	0.95 ± 0.02	0.95 ± 0.02 (0.94) *†	0.40 ± 0.23	0.40 ± 0.31 (0.63) *	2.8
	Humerus_Head_R	0.93 ± 0.09	0.96 ± 0.02 (0.90) *†	0.82 ± 2.37	0.31 ± 0.26 (1.81) *	2.8
	Kidney_L	0.74 ±0.18	0.92 ± 0.05 (0.85) *	4.25 ± 3.38	0.94 ± 0.80 (2.27) *	3
	Kidney_R	0.74 ± 0.18	0.91 ± 0.07 (0.84) *	3.77 ± 2.28	0.98 ± 0.66 (1.91) *	3
	Larynx	0.45 ± 0.21	0.52 ± 0.14 (0.44) *	3.13 ± 1.18	3.06 ± 1.08 (3.59) *	2.4
	Liver	0.85 ± 0.11	0.93 ± 0.06 (0.89) *	5.30 ± 4.79	2.00 ± 1.79 (3.54) *	3
	Lung_L	0.95 ± 0.02	0.96 ± 0.02 (0.95) *	0.90 ± 0.39	0.77 ± 0.37 (0.90) *	3
	Lung_R	0.95 ± 0.03	0.96 ± 0.02 (0.95) *	1.11 ±0.58	0.89 ± 0.43 (1.07) *	3
	Musc_Constrict	0.41 ±0.16	0.45 ± 0.16 (0.33) *	2.22 ± 1.86	2.07 ± 1.52 (3.11) *	2
	Pancreas	0.17 ±0.19	0.41 ±0.24 (0.24) *	15.18 ± 18.92	8.68 ± 11.12 (22.23)	2.25
	SpinalCord	0.66 ± 0.16	0.66 ± 0.16 (0.62) *	1.47 ± 0.88	1.41 ±0.63 (1.62) *	2.75
	Stomach	0.46 ±0.21	0.79 ± 0.19 (0.70) *	10.47 ± 6.72	2.33 ± 2.12 (4.62)	3
	Trachea	0.67 ± 0.17	0.74 ±0.16 (0.65) *	1.22 ± 0.61	0.88 ± 0.54 (1.17) *	3
	LN_Ax_L1_L	0.59 ± 0.09	0.65 ± 0.11 (0.53) *	2.73 ± 0.81	2.58 ± 1.17 (3.90) *	3
	LN_Ax_L1_R	0.52 ± 0.12	0.58 ± 0.14 (0.46) *	3.57 ± 1.31	3.19 ± 1.61 (4.80) *	3
	LN_Ax_L2_L	0.58 ± 0.16	0.63 ± 0.16 (0.50) *	2.56 ± 1.37	2.10 ± 0.96 (3.12) *	2.67
	LN_Ax_L2_R	0.55 ± 0.20	0.60 ± 0.20 (0.46) *	2.88 ± 1.45	2.70 ± 1.77 (3.87) *	2.67
	LN_Ax_L3_L	0.55 ± 0.17	0.61 ± 0.19 (0.46) *	2.33 ± 1.38	2.15 ± 1.52 (3.36) *	3
	LN_Ax_L3_R	0.46 ± 0.16	0.50 ± 0.18 (0.36) *	2.82 ± 1.49	2.53 ± 1.41 (3.90) *	3
	LN_IMN_L	0.17 ± 0.11	0.37 ± 0.16 (0.20) *	2.66 ± 1.05	1.63 ± 0.96 (2.93) *	2.67
	LN_IMN_R	0.24 ± 0.16	0.48 ± 0.19 (0.33) *	2.55 ± 1.84	1.55 ± 1.60 (2.99) *	2.67
	LN_Sclav_L	0.60 ± 0.15	0.67 ± 0.15 (0.53) *	2.18 ± 0.72	1.87 ± 0.89 (2.64) *	2.67
	LN_Sclav_R	0.48 ± 0.09	0.55 ± 0.10 (0.44) *	2.89 ± 0.88	2.60 ± 0.85 (3.54) *	2.67
4.1.0 CT Model: Whole Body Physiological Uptake Organs	Bone	0.76 ± 0.08	0.87 ± 0.05 (0.75) *	3.80 ± 2.88	0.91 ± 1.02 (1.87) *	3
	Bowel	0.46 ± 0.14	0.74 ± 0.10 (0.62) *	1.32 ± 1.01	0.48 ± 0.13 (3.32)	3
	BowelBag	0.59 ± 0.14	0.75 ± 0.15 (0.68) *	10.57 ± 5.08	7.69 ± 5.07 (12.61)	3
	Brain	0.97 ± 0.01	0.97 ± 0.03 (0.96) *	18.17 ± 5.89	17.18 ± 5.24 (20.30)	3
	Cavity_Oral	0.77 ± 0.13	0.78 ± 0.12 (0.68) *	0.69 ± 0.37	0.91 ± 1.17 (1.32) *	3
	Glnd_Lacrimal_L	0.23 ± 0.17	0.37 ± 0.16 (0.24) *	3.18 ± 1.96	3.19 ± 1.66 (4.56) *	3
	Glnd_Lacrimal_R	0.24 ± 0.16	0.46 ± 0.14 (0.34) *	0.88 ± 0.51	0.72 ± 0.43 (1.13) *	3
	Glnd_Submand_L	0.58 ± 0.11	0.79 ± 0.06 (0.72) *	0.74 ± 0.39	0.81 ± 0.79 (1.36) *	3
	Glnd_Submand_R	0.55 ± 0.16	0.77 ± 0.06 (0.68) *	1.65 ± 0.53	0.81 ± 0.35 (1.14) *	3
	Heart	0.89 ± 0.04	0.91 ± 0.04 (0.87) *	1.83 ± 0.90	0.81 ± 0.38 (1.38) *	3
	Kidney_L	0.72 ± 0.17	0.93 ± 0.04 (0.84) *	3.10 ± 0.84	2.65 ± 0.93 (3.42) *	3
	Kidney_R	0.73 ± 0.19	0.92 ± 0.07 (0.82) *	3.76 ± 2.43	0.90 ± 0.51 (2.14) *	3
	Liver	0.84 ± 0.12	0.93 ± 0.07 (0.87) *	3.74 ± 2.91	0.85 ± 0.44 (2.34) *	3
	LN_Iliac	0.63 ± 0.04	0.71 ± 0.03 (0.68) *	5.28 ± 3.59	2.26 ± 3.82 (4.41) *	3
	Parotid_L	0.71 ± 0.09	0.81 ± 0.05 (0.77) *	2.59 ± 0.56	1.59 ± 0.33 (1.97) *	3
	Parotid_R	0.71 ± 0.09	0.82 ± 0.05 (0.78) *	1.97 ± 0.65	1.30 ± 0.47 (1.62) *	3
	Prostate	0.71 ± 0.14	0.85 ± 0.05 (0.76) *	2.08 ± 0.71	1.34 ± 0.56 (1.70) *	3
	Spleen	0.72 ± 0.10	0.95 ± 0.01 (0.87) *	3.21 ± 1.54	1.55 ± 0.61 (2.53) *	3

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

Dice score, MDA (Mean Distance to Agreement), Cumulative APL (Added Path Length), External Evaluation Score / User evaluation

Predicate Device(s)

K223774 Contour ProtégéAI

Reference Device(s)

K071964 MIM 4.1 SEASTAR (tradename MIM Maestro)

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

November 8, 2023

Image /page/0/Picture/1 description: The image shows the logo of the U.S. Food and Drug Administration (FDA). The logo consists of two parts: the Department of Health & Human Services logo on the left and the FDA text logo on the right. The FDA text logo is in blue and reads "FDA U.S. FOOD & DRUG ADMINISTRATION".

MIM Software Inc. % Lynn Hanigan Principal Regulatory and Quality Specialist 25800 Science Park Drive, Suite 180 CLEVELAND. OH 44122

Re: K231765

Trade/Device Name: Contour ProtégéAI Regulation Number: 21 CFR 892.2050 Regulation Name: Medical Image Management And Processing System Regulatory Class: Class II Product Code: QKB Dated: October 9, 2023 Received: October 10, 2023

Dear Lynn Hanigan:

We have reviewed your section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (the Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. Although this letter refers to your product as a device, please be aware that some cleared products may instead be combination products. The 510(k) Premarket Notification Database available at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm identifies combination product submissions. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading.

If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.

Additional information about changes that may require a new premarket notification are provided in the FDA guidance documents entitled "Deciding When to Submit a 510(k) for a Change to an Existing Device" (https://www.fda.gov/media/99812/download) and "Deciding When to Submit a 510(k) for a Software Change to an Existing Device" (https://www.fda.gov/media/99785/download).

Your device is also subject to, among other requirements, the Quality System (QS) regulation (21 CFR Part 820), which includes, but is not limited to, 21 CFR 820.30, Design controls; 21 CFR 820.90, Nonconforming

1

product; and 21 CFR 820.100, Corrective and preventive action. Please note that regardless of whether a change requires premarket review, the QS regulation requires device manufacturers to review and approve changes to device design and production (21 CFR 820.30 and 21 CFR 820.70) and document changes and approvals in the device master record (21 CFR 820.181).

Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting of medical device-related adverse events) (21 CFR Part 803) for devices or postmarketing safety reporting (21 CFR Part 4, Subpart B) for combination products (see https://www.fda.gov/combination-products/guidance-regulatory-information/postmarketing-safety-reportingcombination-products); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820) for devices or current good manufacturing practices (21 CFR Part 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR Parts 1000-1050.

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-regulatoryassistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email ([e-mail]) or phone [phone] or 301-796-7100).

Sincerely,

Loca Weidner

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

2

Indications for Use

510(k) Number (if known) K231765

Device Name Contour ProtégéAI

Indications for Use (Describe)

Trained medical professionals use Contour ProtégéAI as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0. In addition, Contour ProtégéAI supports the following indications:

· Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.

· Segmenting anatomical structures across a variety of CT anatomic locations.

· And segmenting the prostate, the seminal vesicles, and the urethra within T2-weighted MR images.

Appropriate image visualization software must be used to review and, if necessary, edit results automatically generated by Contour ProtégéAI.

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

K231765

Image /page/3/Picture/1 description: The image shows the logo for MIM Software. The logo consists of a red square with rounded corners, partially covered by a gray square with rounded corners and a white circle in the middle. To the right of the squares is the text "mim" in a bold, sans-serif font, with the word "SOFTWARE" in a smaller font below it.

25800 Science Park Drive - Suite 180 Cleveland, OH 44122 866-421-2536 www.mimsoftware.com

510(k) Summary of Safety and Effectiveness (The following information is in conformance with 21 CFR 807.92)

Submitter:

MIM Software Inc. 25800 Science Park Drive - Suite 180 Cleveland, OH 44122

216-455-0600 Phone: [phone] Fax:

Contact Person:

Lynn Hanigan

October 5, 2023

Date Summary Prepared:

Device Name

Trade Name: Common Name: Regulation Number / Product Code: Classification Name:

Contour ProtégéAl Medical Imaqing Software 21 CFR 892.2050 Product Code QKB Radiological Image Processing Software For Radiation Therapy

Predicate Devices

Primary -	K223774	Contour ProtégéAI	MIM Software Inc.
Reference -	K071964	MIM 4.1 SEASTAR (tradename MIM Maestro)	MIMvista Corp.

Intended Use

Contour ProtégéAI is an accessory to MIM software. It includes processing components to automatically contour imaging data using machine-learning based algorithms.

Contour ProtégéAl must be used in conjunction with MIM software to review and, if necessary, edit results automatically generated by Contour ProtégéAI.

Contour ProtégéAl is not intended to automatically detect lesions.

Indications for Use

Trained medical professionals use Contour ProtégéAl as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0. In addition, Contour ProtégéAl supports the following indications:

• Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation

4

Image /page/4/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping rounded squares, one gray and one red, with a white circle where they overlap. To the right of the squares is the text "mim" in a sans-serif font, with the word "SOFTWARE" in a smaller font below it. The logo is simple and modern, with a focus on the company's name.

therapy treatment planning systems, and archiving contours for patient follow-up and management.

• Segmenting anatomical structures across a variety of CT anatomical locations.
• And segmenting the prostate, the seminal vesicles, and the urethra within T2-weighted MR images.

Appropriate image visualization software must be used to review and, if necessary, edit results automatically generated by Contour ProtégéAl.

Device Description

Contour ProtégéAI is an accessory to MIM software that automatically creates contours on medical images through the use of machine-learning algorithms. It is designed for use in the processing of medical images and operates on Windows, Mac, and Linux computer systems. Contour ProtégéAl is deployed on a remote server using the MIMcloud service for data management and transfer; or locally on the workstation or server running MIM software.

A total of 550 CT images from 41 clinical sites across multiple continents was gathered for the training of the final 4.1.0 neural network models. The following table lists the data used for the training of the 4.1.0 final production models. Across the training data, 56.4% were male, 25.8% were female, and 17.8% were not labeled. Training data patient age ranges were 1-20: 0.2%, 20-40: 1.8%, 40-60: 8.9%, 60+: 24.2%, and not labeled 64.9%. Images were acquired by scanners made by: Phillips, 38.5%, GE, 17.6%; Siemens, 16.2%; Toshiba, 2.7% and not labeled 25.0%. Over the dataset, 38.7% of the images were acquired with IV or oral contrast, and 61.3% were acquired with no contrast.

Institution	Country	# of images
Institution 1	USA	2
Institution 2	USA	5
Institution 3	USA	26
Institution 4	USA	8
Institution 5	USA	3
Institution 6	USA	1
Institution 7	USA	34
Institution 8	USA	17
Institution 9	USA	6
Institution 10	Hong Kong	30
Institution 11	USA	13
Institution 12	USA	6
Institution 13	USA	10
Institution 14	USA	6
Institution 15	USA	24
Institution 16	USA	2
Institution 17	USA	14

Table 1: CT data used to train the final production of the 4.1.0 CT models.

5

Image /page/5/Picture/0 description: The image shows the MIM Software logo. The logo consists of two overlapping rounded squares, one gray and one red, with a white circle where they overlap. To the right of the squares is the text "mim" in a bold, sans-serif font, with the word "SOFTWARE" in a smaller font below it. The logo is simple and modern, with a focus on geometric shapes and clean lines.

Institution	Country	# of images
Institution 18	USA	7
Institution 19	USA	27
Institution 20	USA	27
Institution 21	USA	6
Institution 22	USA	41
Institution 23	Australia	19
Institution 24	USA	10
Institution 25	USA	11
Institution 26	USA	10
Institution 27	USA	2
Institution 28	USA	1
Institution 29	USA	6
Institution 30	USA	9
Institution 31	USA	4
Institution 32	USA	2
Institution 33	USA	6
Institution 34	USA	2
Institution 35	USA	6
Institution 36	USA	14
Institution 37	USA	4
Institution 38	USA	5
Institution 39	USA	75
Institution 40	Australia	48
Institution 41	USA	1

Substantial Equivalence

| ITEM | Subject Device:
Contour ProtégéAl
(K231765) | Predicate Device:
Contour ProtégéAl
(K223774) | Reference
Predicate:
MIM 4.1 SEASTAR
[i.e., MIM Maestro]
(K071964) |
|--------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Clearance Dates | TBD | 04/06/2023 | 9/26/2007 |
| ITEM | Subject Device:
Contour ProtégéAl
(K231765) | Predicate Device:
Contour ProtégéAl
(K223774) | Reference
Predicate:
MIM 4.1 SEASTAR
[i.e., MIM Maestro]
(K071964) |
| Intended Use | Contour ProtégéAI is an
accessory to MIM software
used for the contouring of
anatomical structures in
imaging data using machine-
learning-based algorithms
automatically.
Appropriate image visualization
software must be used to
review and, if necessary, edit
results automatically generated
by Contour ProtégéAI.
Contour ProtégéAl is not
intended to detect or contour
lesions. | Contour ProtégéAI is an
accessory to MIM software used
for the contouring of anatomical
structures in imaging data using
machine-learning-based
algorithms automatically.
Appropriate image visualization
software must be used to review
and, if necessary, edit results
automatically generated by
Contour ProtégéAI.
Contour ProtégéAl is not
intended to detect or contour
lesions. | MIM 4.1
(SEASTAR)
software is
intended for trained
medical
professionals
including, but not
limited to,
radiologists,
oncologists.
physicians, medical
technologists,
dosimetrists, and
physicists.
MIM 4.1
(SEASTAR) is a
medical image and
information
management
system that is
intended to receive,
transmit, store,
retrieve, display,
print and process
digital medical
images, as well as
create, display and
print reports from
those images. The
medical modalities
of these medical
imaging systems
include, but are not
limited to, CT, MRI,
CR, DX, MG, US,
SPECT, PET and
XA as supported by
ACR/NEMA
DICOM 3.0.
MIM 4.1
(SEASTAR)
provides tools to
quickly create,
transform, and
modify contours for |
| ITEM | Subject Device:
Contour ProtégéAl
(K231765) | Predicate Device:
Contour ProtégéAl
(K223774) | Reference
Predicate:
MIM 4.1 SEASTAR
[i.e., MIM Maestro]
(K071964) |
| Indications for Use | Trained medical professionals
use Contour ProtégéAl as a
tool to assist in the automated
processing of digital medical
images of modalities CT and
MR, as supported by
ACR/NEMA DICOM 3.0. In
addition, Contour ProtégéAl
supports the following
indications: | Trained medical professionals
use Contour ProtégéAl as a tool
to assist in the automated
processing of digital medical
images of modalities CT and
MR, as supported by
ACR/NEMA DICOM 3.0. In
addition, Contour ProtégéAl
supports the following
indications: | MIM 4.1
(SEASTAR)
software is used by
trained medical
professionals as a
tool to aid in
evaluation and
information
management of
digital medical |
| | Creation of contours using
machine-learning
algorithms for applications
including, but not limited to,
quantitative analysis, aiding
adaptive therapy,
transferring contours to
radiation therapy treatment
planning systems, and
archiving contours for
patient follow-up and
management. | Creation of contours using
machine-learning algorithms
for applications including,
but not limited to,
quantitative analysis, aiding
adaptive therapy,
transferring contours to
radiation therapy treatment
planning systems, and
archiving contours for
patient follow-up and
management. | images. The
medical image
modalities include,
but are not limited
to, CT, MRI, CR,
DX, MG, US,
SPECT, PET and
XA as supported by
ACR/NEMA
DICOM 3.0. MIM
4.1 (SEASTAR)
assists in the
following |
| | Segmenting anatomical
structures across a variety
of CT anatomical locations. And segmenting the
prostate, the seminal
vesicles, and the urethra | Segmenting anatomical
structures across a variety of
CT anatomical locations. And segmenting the
prostate, the seminal
vesicles, and the urethra | indications:
Receive,
transmit, store, retrieve, display,
print, and process
medical images |
| ITEM | Subject Device:
Contour ProtégéAl
(K231765) | Predicate Device:
Contour ProtégéAl
(K223774) | Reference
Predicate:
MIM 4.1 SEASTAR
[i.e., MIM Maestro]
(K071964) |
| | within T2-weighted MR
images.
Appropriate image visualization
software must be used to
review and, if necessary, edit
results automatically generated
by Contour ProtégéAI. | within T2-weighted MR
images.
Appropriate image visualization
software must be used to review
and, if necessary, edit results
automatically generated by
Contour ProtégéAI. | and DICOM
objects.
· Create, display
and print reports
from medical
images.
· Registration,
fusion display, and
review of medical
images for
diagnosis,
treatment
evaluation, and
treatment planning.
• Localization and
definition of objects
such as tumors and
normal tissues in
medical images.
· Creation,
transformation, and
modification of
contours for
applications
including, but not
limited to,
quantitative
analysis, aiding
adaptive therapy,
transferring
contours to
radiation therapy
treatment planning
systems, and
archiving contours
for patient follow-up
and management. |
| ITEM | Subject Device:
Contour ProtégéAl
(K231765) | Predicate Device:
Contour ProtégéAl
(K223774) | Reference
Predicate:
MIM 4.1 SEASTAR
[i.e., MIM Maestro]
(K071964) |
| Modalities | CT and MR | CT and MR | CT, MR, CR, DX,
MG, US, SPECT,
PET and XA |
| Atlas-Based
Segmentation | No | No | Yes |
| Automatically
Contour Imaging
Data Using
Machine-Learning | Yes | Yes | No |
| Operating
Platform | Server-based application
supporting
Linux-based OS

• and -
  Local deployment on Windows
  or Mac | Server-based application
  supporting
  Linux-based OS
• and -
  Local deployment on Windows
  or Mac | Windows, Mac |
  | Cloud-based
  deployment | Yes | Yes | No |
  | Locally deployed
  (or installed) | Yes | Yes | No |
  | ITEM | Subject Device:
  Contour Protégé Al
  (K231765) | Predicate Device:
  Contour ProtégéAl
  (K223774) | Reference
  Predicate:
  MIM 4.1 SEASTAR
  [i.e., MIM Maestro]
  (K071964) |
  | Neural Network
  Models included | (1.0.0 models)
  Head and Neck CT
  Prostate CT
  Thorax CT
  Liver CT
  Prostate MR

(1.1.0 model)
Prostate MR

(2.0.0 models)
Head and Neck CT
Prostate CT
Thorax CT
Abdomen CT
Lungs and Liver CT

(3.1.0 models)*
Head and Neck CT
Prostate CT
Thorax CT
Abdomen CT
Lungs and Liver CT
MRT Additional Structures CT

(4.0.0 models)
Head and Neck CT
Thorax CT
Abdomen CT
Pelvis CT | (1.0.0 models)
Head and Neck CT
Prostate CT
Thorax CT
Liver CT
Prostate MR

(1.1.0 model)
Prostate MR

(2.0.0 models)
Head and Neck CT
Prostate CT
Thorax CT
Abdomen CT
Lungs and Liver CT

(3.0.0 models)
Head and Neck CT
Prostate CT
Thorax CT
Abdomen CT
Lungs and Liver CT
MRT Additional Structures CT

(4.0.0 models)
Head and Neck CT
Thorax CT
Abdomen CT
Pelvis CT | None |
| ITEM | Subject Device:
Contour ProtégéAl
(K231765) | Predicate Device:
Contour ProtégéAl
(K223774) | Reference
Predicate:
MIM 4.1 SEASTAR
[i.e., MIM Maestro]
(K071964) |
| | SurePlan MRT CT | SurePlan MRT CT | |
| | (4.1.0 models) | | |
| | Head and Neck CT | | |
| | Thorax CT | | |
| | Whole Body - Physiological
Uptake Organs CT | | |

6

Image /page/6/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping squares, one gray and one red, with a white circle cut out of the red square. To the right of the squares is the text "mim" in a sans-serif font, with the word "SOFTWARE" in a smaller font below it. The logo is simple and modern, and the colors are eye-catching.

7

Image /page/7/Picture/0 description: The image shows the MIM Software logo. The logo consists of a red square with a white circle cut out of the top left corner, and a gray square overlapping the top left corner of the red square. To the right of the squares is the word "mim" in black, with the word "SOFTWARE" in smaller black letters below it.

8

Image /page/8/Picture/0 description: The image is a logo for MIM Software. The logo consists of two overlapping rounded squares, one gray and one red, with a white circle where they overlap. To the right of the squares is the text "mim" in a bold, sans-serif font, with the word "SOFTWARE" in a smaller font below it.

9

Image /page/9/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping rounded squares, one gray and one red, with a white circle where they overlap. To the right of the squares is the text "mim" in a stylized font, with the word "SOFTWARE" in smaller letters below it.

10

Image /page/10/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping rounded squares, one gray and one red, with a white circle cut out of the red square. To the right of the squares is the text "mim" in a bold, sans-serif font, with the word "SOFTWARE" in a smaller font below it.

11

Image /page/11/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping rounded squares, one gray and one red, with a white circle cut out where they overlap. To the right of the squares is the text "mim" in a bold, sans-serif font, with the word "SOFTWARE" in a smaller font below.

Discussion

Changes within this submission include new 4.1.0 neural network models with additional contours. These changes differ when comparing to Contour ProtégéAl 510(K)223774. Noninferiority testing was used to compare the proposed Contour ProtégéAl device to Atlases created from the MIM Maestro (K071964) reference device.

Testing and Performance Data

The proposed Contour ProtégéAl device were then evaluated on the test subjects from a pool of 754 independent images gathered from 27 institutions.

The CT images for this training set were obtained from clinical treatment plans for patients prescribed external beam or molecular radiotherapy, but the original segmentations were not used. Instead, the images were re-segmented by consultants (physicians and dosimetrists) specifically for this purpose, outside of clinical practice. Detailed instructions derived from relevant published clinical contouring quidelines were prepared for the dosimetrists. The initial segmentations were then reviewed and corrected by a radiation oncologist against the same standards and quidelines. Qualified staff at MIM Software (M.D. or licensed dosimetrists) then performed a final review and correction. All segmenters were instructed to spend additional time to ensure the highest quality training data. In particular, the consultants were asked to contour all specified OAR structures on all images according to referenced standards, whether or not they were proximal to the treatment field. All patients were imaged on an indexed couch in treatment position ("simulation CT"). Series that were non-axial, had slices thinner than 0.5mm, or had non-Fan Beam or MV acquisitions were excluded.

The verification data used for testing is from a set of institutions that are totally disioint from the training datasets used to train each model in the Contour ProtégéAl device. The reference predicate MIM device was configured with Atlases built from the same training data used to train

12

Image /page/12/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping rounded squares, one gray and one red, with a white circle where they intersect. To the right of the squares is the text "mim" in a bold, sans-serif font, with the word "SOFTWARE" in a smaller font below it.

the models. We tested Contour ProtégéAl against the reference predicate device, and the goal of this testing is to show that it is equivalent or superior to the reference predicate. The performance of both segmentation devices was measured by calculating both the Dice score and MDA of the novel segmentations with the original ground-truth contours. User beta testing was also used to evaluate the performance of Contour ProtégéAl in the context of time savings compared to contouring from scratch. This user evaluation was made on a three-point scale for each contour, with one indicating negligible time savings, two indicating moderate, and three indicating significant time savings. Our acceptance criteria combine the statistical tests and the user evaluation - only structures that pass two or more of the following three tests could be included in the final models:

• Statistical non-inferiority of the Dice score compared with the reference predicate. ●
• . Statistical non-inferiority of the MDA score compared with the reference predicate.
• Average user evaluation of 2 or higher, when measured on a three-point scale. .

Further clinical validation was also conducted in-house to evaluate contours compared to detailed criteria based on established clinical quidelines.

In addition, each model as a whole was also evaluated. In order to be included in the released product, the cumulative Added Path Length of the contours in the model was evaluated relative to the ground-truth. Cumulative APL has been found to correlate well with the spent in editing and correcting auto-segmented contours (Vaassen et al, 2020)'. Each model was required to have statistically non-inferior cumulative APL compared to the reference predicate.

Finally, the localization accuracy of each structure in all models was measured. We did not impose a passing criterion on localization accuracy; the results however are included in this summary document and in our User Guide and Whitepaper to allow the user to better understand the performance of the device.

Across the testing data the images 48.4% were female, 35.2% were male, and 16.4% were unknown for sex. The manufacturer was GE for 37.9% of images, 25.1% was Siemens, 15.0% were Philips, 0.4% were from Toshiba, and the other 21.6% were unknown. 4.38% were between the ages of 20-40, 21.5% were between 40-60, and 47.9% were over the age of 60. 26.0% were unknown, and 0.3% were under the age 20.

The mean and standard deviation Dice coefficients and MDA scores, along with the lower 95th percentile confidence bound, were calculated for both the proposed Contour ProtégéAl device and the MIM Maestro atlas seamentation reference device for each structure of each neural network model. Contour ProtégéAl results were equivalent or had better performance than the MIM Maestro atlas segmentation reference device. Equivalence is defined such that the lower 95th percentile confidence bound of the Contour ProtégéAl segmentation is greater than 0.1 Dice lower than the mean MIM atlas segmentation reference device performance.

Results over the validation set compared to the reference device are presented here:

| 4.1.0 CT
Model: | Structure: | Dice
MIM Atlas | Dice
Contour
ProtégéAl | MDA
MIM Atlas | MDA
Contour
ProtégéAl | External
Evaluation
Score |
|--------------------------------------------------|-------------------|-------------------|------------------------------|------------------|-----------------------------|---------------------------------|
| Head and
Neck | Bone_Mandible | 0.81 ± 0.07 | 0.86 ± 0.07
(0.83) * | 1.00 ± 0.87 | 0.64 ± 0.31
(0.93) * | 2.86 |
| | BrachialPlex_L | 0.17 ± 0.08 | 0.22 ± 0.10
(0.14) * | 3.24 ± 2.62 | 2.82 ± 2.61
(5.12) * | 2.6 |
| 4.1.0 CT
Model: | Structure: | Dice
MIM Atlas | Dice
Contour
ProtégéAl | MDA
MIM Atlas | MDA
Contour
ProtégéAl | External
Evaluation
Score |
| | BrachialPlex_R | 0.15 ± 0.07 | 0.19 ± 0.09
(0.11) * | 3.62 ± 2.79 | 2.73 ± 2.38
(5.20) * | 2.6 |
| | Brain | 0.97 ± 0.01 | 0.98 ± 0.01
(0.97) * | 0.64 ± 0.37 | 0.52 ± 0.36
(0.66) * | 2.71 |
| | Brainstem | 0.78 ± 0.09 | 0.82 ± 0.08
(0.78) * | 1.72 ± 0.81 | 1.20 ± 0.65
(1.47) * | 2.71 |
| | Cavity_Oral | 0.75 ± 0.14 | 0.76 ± 0.13
(0.68) * | 3.34 ± 2.13 | 3.20 ± 1.93
(4.35) * | 2.71 |
| | Cochlea_L | 0.19 ± 0.15 | 0.30 ± 0.18
(0.22) * | 1.26 ± 0.85 | 1.17 ± 0.87
(1.58) * | 2.29 |
| | Cochlea_R | 0.17 ±0.15 | 0.32 ± 0.20
(0.23) * | 1.12 ± 0.61 | 1.05 ± 0.68
(1.37) * | 2.29 |
| | Eye_L | 0.80 ± 0.08 | 0.87 ± 0.06
(0.83) * | 1.07 ± 0.60 | 0.65 ± 0.51
(0.91) * | 2.57 |
| | Eye_R | 0.80 ± 0.10 | 0.86 ± 0.07
(0.82) * | 1.11 ±0.64 | 0.66 ± 0.48
(0.92) * | 2.57 |
| | Glnd_Lacrimal_L | 0.22 ± 0.17 | 0.40 ± 0.16
(0.27) * | 0.86 ± 0.54 | 0.68 ± 0.35
(1.05) * | 2.71 |
| | Glnd_Lacrimal_R | 0.23 ± 0.16 | 0.45 ± 0.14
(0.34) * | 0.73 ± 0.38 | 0.71 ± 0.52
(1.09) * | 2.71 |
| | Glnd_Submand_L | 0.57 ± 0.13 | 0.77 ± 0.11
(0.69) * | 1.74 ± 0.72 | 0.76 ± 0.32
(1.12) * | 3 |
| | Glnd_Submand_R | 0.56 ± 0.16 | 0.75 ± 0.11
(0.65) * | 1.82 ± 0.87 | 0.81 ± 0.34
(1.28) * | 3 |
| | Glnd_Thyroid | 0.47 ±0.18 | 0.71 ±0.19
(0.57) * | 2.55 ± 2.14 | 1.38 ± 3.18
(3.38) * | 2.71 |
| | Lens_L | 0.22 ± 0.23 | 0.61 ± 0.17
(0.52) * | 0.77 ± 0.54 | 0.56 ± 0.47
(0.84) * | 2.29 |
| | Lens_R | 0.22 ± 0.22 | 0.63 ± 0.16
(0.54) * | 0.81 ± 0.47 | 0.55 ± 0.33
(0.78) * | 2.29 |
| | Lips | 0.38 ± 0.14 | 0.37 ± 0.15
(0.28) * | 4.14 ± 2.86 | 5.26 ± 3.41
(7.27) | 2.83 |
| | OpticChiasm | 0.04 ± 0.07 | 0.13 ±0.13
(0.08) * | 2.69 ± 1.87 | 2.46 ± 1.99
(3.55) * | 2 |
| | OpticNrv_L | 0.45 ± 0.15 | 0.53 ± 0.14
(0.45) * | 0.97 ± 0.81 | 0.77 ± 0.85
(1.19) * | 2.57 |
| | OpticNrv_R | 0.44 ±0.14 | 0.52 ± 0.13
(0.44) * | 1.11 ± 1.24 | 0.80 ± 0.83
(1.34) * | 2.57 |
| | Parotid_L | 0.71 ±0.10 | 0.79 ± 0.09
(0.75) * | 2.02 ± 0.95 | 1.30 ± 0.62
(1.68) * | 3 |
| | Parotid_R | 0.71 ± 0.09 | 0.80 ± 0.06
(0.76) * | 2.11 ±0.76 | 1.26 ± 0.43
(1.56) * | 3 |
| | Pituitary | 0.38 ± 0.18 | 0.54 ± 0.15 | 1.07 ± 0.59 | 0.87 ± 0.54 | 3 |
| 4.1.0 CT
Model: | Structure: | Dice
MIM Atlas | Dice
Contour
ProtégéAl | MDA
MIM Atlas | MDA
Contour
ProtégéAl | External
Evaluation
Score |
| | | | (0.41) * | | (1.29) * | |
| | SpinalCord | 0.66 ± 0.14 | 0.65 ± 0.16
(0.59) * | 0.81 ±0.33 | 0.73 ± 0.41
(0.88) * | 2.86 |
| | LN_Neck_IA | 0.48 ± 0.13 | 0.60 ± 0.14
(0.44) * | 1.05 ± 1.51 | 0.47 ± 0.45
(1.81) * | 2.75 |
| | LN_Neck_IB_L | 0.70 ±0.05 | 0.79 ± 0.04
(0.73) * | 1.30 ± 0.47 | 0.73 ± 0.24
(1.16) * | 2.75 |
| | LN_Neck_IB_R | 0.68 ±0.06 | 0.79 ±0.05
(0.72) * | 1.41 ± 0.51 | 0.73 ± 0.20
(1.19) * | 2.75 |
| | LN_Neck_IIA_L | 0.64 ± 0.04 | 0.75 ± 0.05
(0.70) * | 2.02 ± 0.54 | 1.54 ± 0.44
(2.09) * | 2.25 |
| | LN_Neck_IIA_R | 0.60 ± 0.08 | 0.76 ± 0.04
(0.69) * | 2.24 ± 0.76 | 1.41 ±0.49
(2.15) * | 2.25 |
| | LN_Neck_IIB_L | 0.68 ± 0.11 | 0.80 ± 0.06
(0.68) * | 0.95 ± 0.49 | 0.63 ± 0.31
(1.12) * | 3 |
| | LN_Neck_IIB_R | 0.68 ± 0.09 | 0.80 ± 0.06
(0.70) * | 1.19 ± 0.64 | 0.58 ± 0.15
(1.14) * | 3 |
| | LN_Neck_III_L | 0.54 ± 0.12 | 0.75 ± 0.07
(0.63) * | 1.53 ± 0.60 | 0.92 ± 0.44
(1.51) * | 3 |
| | LN_Neck_III_R | 0.57 ± 0.12 | 0.75 ± 0.07
(0.63) * | 1.57 ± 0.61 | 0.99 ± 0.37
(1.57) * | 3 |
| | LN_Neck_IV_L | 0.58 ± 0.08 | 0.69 ± 0.08
(0.59) * | 1.62 ± 0.47 | 1.13 ± 0.37
(1.63) * | 2.75 |
| | LN_Neck_IV_R | 0.55 ± 0.11 | 0.71 ± 0.09
(0.58) * | 1.87 ± 0.75 | 1.06 ± 0.33
(1.77) * | 2.75 |
| | LN_Neck_V_L | 0.53 ± 0.11 | 0.58 ± 0.09
(0.45) * | 1.43 ± 0.82 | 1.13 ± 0.59
(2.02) * | 3 |
| | LN_Neck_V_R | 0.50 ± 0.10 | 0.58 ± 0.11
(0.44) * | 2.03 ± 1.66 | 1.43 ± 0.98
(3.18) * | 3 |
| | LN_Neck_VIA | 0.34 ± 0.09 | 0.36 ± 0.11
(0.24) * | 1.20 ± 0.55 | 0.80 ± 0.35
(1.33) * | 2.75 |
| | LN_Retropharynx_L | 0.22 ± 0.07 | 0.28 ±0.10
(0.19) * | 1.44 ± 0.87 | 1.54 ± 1.47
(2.93) * | 2.75 |
| | LN_Retropharynx_R | 0.21 ± 0.07 | 0.28 ± 0.09
(0.18) * | 1.64 ± 0.93 | 1.49 ± 1.36
(2.82) * | 2.75 |
| | LN_Retrostyloid_L | 0.57 ± 0.10 | 0.64 ± 0.09
(0.53) * | 1.58 ± 0.75 | 1.23 ± 0.48
(1.96) * | 2.75 |
| | LN_Retrostyloid_R | 0.56 ± 0.11 | 0.66 ±0.11
(0.53) * | 1.59 ± 0.74 | 1.13 ± 0.66
(1.95) * | 2.75 |
| Thorax | BrachialPlex_L | 0.28 ± 0.14 | 0.38 ± 0.14
(0.26) * | 2.96 ± 1.79 | 2.94 ± 1.84
(4.49) * | 3 |
| | BrachialPlex_R | 0.28 ± 0.11 | 0.37 ± 0.16
(0.24) * | 3.37 ± 2.27 | 3.28 ± 2.60
(5.51) | 3 |
| 4.1.0 CT
Model: | Structure: | Dice
MIM Atlas | Dice
Contour
ProtégéAl | MDA
MIM Atlas | MDA
Contour
ProtégéAl | External
Evaluation
Score |
| | Breast_L | 0.74 ±0.11 | 0.76 ± 0.15
(0.66) * | 4.86 ± 3.45 | 3.64 ± 1.85
(5.65) * | 2.8 |
| | Breast_R | 0.76 ± 0.12 | 0.78 ± 0.15
(0.67) * | 4.02 ± 1.70 | 3.45 ± 1.51
(4.66) * | 2.6 |
| | Bronchus | 0.60 ± 0.17 | 0.66 ± 0.13
(0.55) * | 2.18 ± 1.45 | 1.90 ± 1.22
(2.78) * | 2.6 |
| | Carina | 0.36 ± 0.18 | 0.49 ± 0.12
(0.40) * | 1.27 ± 0.55 | 1.03 ± 0.47
(1.33) * | 2.4 |
| | Cricoid | 0.02 ± 0.04 | 0.04 ± 0.05 (-
0.01) * | 4.64 ± 1.54 | 5.71 ± 1.24
(7.05) | 2.4 |
| | Esophagus | 0.48 ± 0.16 | 0.70 ± 0.15
(0.64) * | 2.49 ± 2.48 | 0.97 ± 0.54
(1.56) * | 2.8 |
| | Glnd_Thyroid | 0.46 ± 0.18 | 0.65 ± 0.17
(0.54) * | 2.92 ± 2.60 | 1.75 ± 1.78
(3.19) * | 2.75 |
| | GreatVes | 0.70 ± 0.08 | 0.74 ±0.10
(0.65) * | 3.37 ± 1.42 | 2.67 ± 1.73
(4.14) * | 3 |
| | Heart | 0.88 ± 0.08 | 0.90 ± 0.09
(0.87) * | 3.39 ± 1.91 | 2.55 ± 1.40
(3.06) * | 3 |
| | Humerus_Head_L | 0.95 ± 0.02 | 0.95 ± 0.02
(0.94) *† | 0.40 ± 0.23 | 0.40 ± 0.31
(0.63) * | 2.8 |
| | Humerus_Head_R | 0.93 ± 0.09 | 0.96 ± 0.02
(0.90) *† | 0.82 ± 2.37 | 0.31 ± 0.26
(1.81) * | 2.8 |
| | Kidney_L | 0.74 ±0.18 | 0.92 ± 0.05
(0.85) * | 4.25 ± 3.38 | 0.94 ± 0.80
(2.27) * | 3 |
| | Kidney_R | 0.74 ± 0.18 | 0.91 ± 0.07
(0.84) * | 3.77 ± 2.28 | 0.98 ± 0.66
(1.91) * | 3 |
| | Larynx | 0.45 ± 0.21 | 0.52 ± 0.14
(0.44) * | 3.13 ± 1.18 | 3.06 ± 1.08
(3.59) * | 2.4 |
| | Liver | 0.85 ± 0.11 | 0.93 ± 0.06
(0.89) * | 5.30 ± 4.79 | 2.00 ± 1.79
(3.54) * | 3 |
| | Lung_L | 0.95 ± 0.02 | 0.96 ± 0.02
(0.95) * | 0.90 ± 0.39 | 0.77 ± 0.37
(0.90) * | 3 |
| | Lung_R | 0.95 ± 0.03 | 0.96 ± 0.02
(0.95) * | 1.11 ±0.58 | 0.89 ± 0.43
(1.07) * | 3 |
| | Musc_Constrict | 0.41 ±0.16 | 0.45 ± 0.16
(0.33) * | 2.22 ± 1.86 | 2.07 ± 1.52
(3.11) * | 2 |
| | Pancreas | 0.17 ±0.19 | 0.41 ±0.24
(0.24) * | 15.18 ±
18.92 | 8.68 ± 11.12
(22.23) | 2.25 |
| | SpinalCord | 0.66 ± 0.16 | 0.66 ± 0.16
(0.62) * | 1.47 ± 0.88 | 1.41 ±0.63
(1.62) * | 2.75 |
| | Stomach | 0.46 ±0.21 | 0.79 ± 0.19
(0.70) * | 10.47 ± 6.72 | 2.33 ± 2.12
(4.62) | 3 |
| | Trachea | 0.67 ± 0.17 | 0.74 ±0.16 | 1.22 ± 0.61 | 0.88 ± 0.54 | 3 |
| 4.1.0 CT
Model: | Structure: | Dice
MIM Atlas | Dice
Contour
ProtégéAl | MDA
MIM Atlas | MDA
Contour
ProtégéAl | External
Evaluation
Score |
| | | | (0.65) * | | (1.17) * | |
| | LN_Ax_L1_L | 0.59 ± 0.09 | 0.65 ± 0.11
(0.53) * | 2.73 ± 0.81 | 2.58 ± 1.17
(3.90) * | 3 |
| | LN_Ax_L1_R | 0.52 ± 0.12 | 0.58 ± 0.14
(0.46) * | 3.57 ± 1.31 | 3.19 ± 1.61
(4.80) * | 3 |
| | LN_Ax_L2_L | 0.58 ± 0.16 | 0.63 ± 0.16
(0.50) * | 2.56 ± 1.37 | 2.10 ± 0.96
(3.12) * | 2.67 |
| | LN_Ax_L2_R | 0.55 ± 0.20 | 0.60 ± 0.20
(0.46) * | 2.88 ± 1.45 | 2.70 ± 1.77
(3.87) * | 2.67 |
| | LN_Ax_L3_L | 0.55 ± 0.17 | 0.61 ± 0.19
(0.46) * | 2.33 ± 1.38 | 2.15 ± 1.52
(3.36) * | 3 |
| | LN_Ax_L3_R | 0.46 ± 0.16 | 0.50 ± 0.18
(0.36) * | 2.82 ± 1.49 | 2.53 ± 1.41
(3.90) * | 3 |
| | LN_IMN_L | 0.17 ± 0.11 | 0.37 ± 0.16
(0.20) * | 2.66 ± 1.05 | 1.63 ± 0.96
(2.93) * | 2.67 |
| | LN_IMN_R | 0.24 ± 0.16 | 0.48 ± 0.19
(0.33) * | 2.55 ± 1.84 | 1.55 ± 1.60
(2.99) * | 2.67 |
| | LN_Sclav_L | 0.60 ± 0.15 | 0.67 ± 0.15
(0.53) * | 2.18 ± 0.72 | 1.87 ± 0.89
(2.64) * | 2.67 |
| | LN_Sclav_R | 0.48 ± 0.09 | 0.55 ± 0.10
(0.44) * | 2.89 ± 0.88 | 2.60 ± 0.85
(3.54) * | 2.67 |
| Whole Body
Physiologic
al Uptake
Organs | Bone | 0.76 ± 0.08 | 0.87 ± 0.05
(0.75) * | 3.80 ± 2.88 | 0.91 ± 1.02
(1.87) * | 3 |
| | Bowel | 0.46 ± 0.14 | 0.74 ± 0.10
(0.62) * | 1.32 ± 1.01 | 0.48 ± 0.13
(3.32) | 3 |
| | BowelBag | 0.59 ± 0.14 | 0.75 ± 0.15
(0.68) * | 10.57 ± 5.08 | 7.69 ± 5.07
(12.61) | 3 |
| | Brain | 0.97 ± 0.01 | 0.97 ± 0.03
(0.96) * | 18.17 ± 5.89 | 17.18 ± 5.24
(20.30) | 3 |
| | Cavity_Oral | 0.77 ± 0.13 | 0.78 ± 0.12
(0.68) * | 0.69 ± 0.37 | 0.91 ± 1.17
(1.32) * | 3 |
| | Glnd_Lacrimal_L | 0.23 ± 0.17 | 0.37 ± 0.16
(0.24) * | 3.18 ± 1.96 | 3.19 ± 1.66
(4.56) * | 3 |
| | Glnd_Lacrimal_R | 0.24 ± 0.16 | 0.46 ± 0.14
(0.34) * | 0.88 ± 0.51 | 0.72 ± 0.43
(1.13) * | 3 |
| | Glnd_Submand_L | 0.58 ± 0.11 | 0.79 ± 0.06
(0.72) * | 0.74 ± 0.39 | 0.81 ± 0.79
(1.36) * | 3 |
| | Glnd_Submand_R | 0.55 ± 0.16 | 0.77 ± 0.06
(0.68) * | 1.65 ± 0.53 | 0.81 ± 0.35
(1.14) * | 3 |
| | Heart | 0.89 ± 0.04 | 0.91 ± 0.04 | 1.83 ± 0.90 | 0.81 ± 0.38 | 3 |
| 4.1.0 CT
Model: | Structure: | Dice
MIM Atlas | Dice
Contour
ProtégéAl | MDA
MIM Atlas | MDA
Contour
ProtégéAl | External
Evaluation
Score |
| | | | (0.87) * | | (1.38) * | |
| | Kidney_L | 0.72 ± 0.17 | 0.93 ± 0.04
(0.84) * | 3.10 ± 0.84 | 2.65 ± 0.93
(3.42) * | 3 |
| | Kidney_R | 0.73 ± 0.19 | 0.92 ± 0.07
(0.82) * | 3.76 ± 2.43 | 0.90 ± 0.51
(2.14) * | 3 |
| | Liver | 0.84 ± 0.12 | 0.93 ± 0.07
(0.87) * | 3.74 ± 2.91 | 0.85 ± 0.44
(2.34) * | 3 |
| | LN_Iliac | 0.63 ± 0.04 | 0.71 ± 0.03
(0.68) * | 5.28 ± 3.59 | 2.26 ± 3.82
(4.41) * | 3 |
| | Parotid_L | 0.71 ± 0.09 | 0.81 ± 0.05
(0.77) * | 2.59 ± 0.56 | 1.59 ± 0.33
(1.97) * | 3 |
| | Parotid_R | 0.71 ± 0.09 | 0.82 ± 0.05
(0.78) * | 1.97 ± 0.65 | 1.30 ± 0.47
(1.62) * | 3 |
| | Prostate | 0.71 ± 0.14 | 0.85 ± 0.05
(0.76) * | 2.08 ± 0.71 | 1.34 ± 0.56
(1.70) * | 3 |
| | Spleen | 0.72 ± 0.10 | 0.95 ± 0.01
(0.87) * | 3.21 ± 1.54 | 1.55 ± 0.61
(2.53) * | 3 |

Table 2: DICE, MDA and External Evaluation Score Comparison Per Structure

13

Image /page/13/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping rounded squares, one gray and one red, with a white circle where they overlap. To the right of the squares is the text "mim SOFTWARE" in black, with "mim" in a larger font than "SOFTWARE".

14

Image /page/14/Picture/0 description: The image is a logo for MIM Software. The logo consists of a gray square overlapping a red square with a white circle in the middle. To the right of the squares is the text "mim SOFTWARE" in black font. The word "mim" is in a larger font than the word "SOFTWARE".

15

Image /page/15/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping squares, one gray and one red, with a white circle where they overlap. To the right of the squares is the text "mim" in a stylized black font, with the word "SOFTWARE" in smaller black letters below it. The logo is clean and modern, with a focus on simplicity and visual appeal.

16

Image /page/16/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping squares, one gray and one red, with a white circle where they overlap. To the right of the squares is the text "mim" in a sans-serif font, with the word "SOFTWARE" below it in a smaller font.

17

Image /page/17/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping squares, one gray and one red, with a white circle cut out of the red square. To the right of the squares is the text "mim" in a bold, sans-serif font, with the word "SOFTWARE" in a smaller font below it. The logo is simple and modern, and the colors are eye-catching.

Mean ± Std Dice coefficient (lower 95th percentile confidence bound based on normal distribution in parentheses) * Equivalence demonstrated at p=0.05 significance level between Contour ProtégéAl and MIM Atlas † Comparisons for both atlas and Contour ProtégéAI calculated only on axial slices that contained the ground truth.

Additionally, preliminary user evaluation conducted as part of testing demonstrated that Contour ProtégéAl yields comparable time-saving functionality when creating contours as other commercially available automatic segmentation products.

4.1.0 CT Models:	MIM Atlas	Contour ProtégéAl
Head and Neck	38.69 ± 33.36	28.61 ± 29.59 (31.61) *
Thorax	89.24 ± 82.73	65.44 ± 68.85 (72.54) *
Whole Body -
Physiological Uptake
Organs	138.06 ± 142.42	98.20 ± 127.11 (118.75) *

Table 3: 4.1.0 CT models cumulative APL in mm

Mean ± Std APL mm (upper 95th percentile confidence bound based on normal distribution in parentheses) * Equivalence demonstrated at p=0.05 significance level between Contour ProtégéAl and MIM Atlas

4.1.0 CT Models:	Structure:	Relevant FOV	Whole Body CT
Head and Neck	Bone_Mandible	100	100
	BrachialPlex_L	100	91
	BrachialPlex_R	100	95
4.1.0 CT Models:	Structure:	Relevant FOV	Whole Body CT
	Brain	100	100
	Brainstem	100	100
	Cavity_Oral	100	100
	Cochlea_L	95	95
	Cochlea_R	96	95
	Eye_L	100	100
	Eye_R	100	100
	Glnd_Lacrimal_L	98	100
	Glnd_Lacrimal_R	100	100
	Glnd_Submand_L	100	100
	Glnd_Submand_R	100	100
	Glnd_Thyroid	98	95
	Lens_L	97	100
	Lens_R	97	95
	Lips	100	95
	OpticChiasm	80	91
	OpticNrv_L	99	100
	OpticNrv_R	99	100
	Parotid_L	99	100
	Parotid_R	100	100
	Pituitary	100	86
	SpinalCord	99	100
	LN_Neck_IA	100	100
	LN_Neck_IB_L	100	100
	LN_Neck_IB_R	100	100
	LN_Neck_IIA_L	100	100
	LN_Neck_IIA_R	100	95
	LN_Neck_IIB_L	100	100
	LN_Neck_IIB_R	100	95
	LN_Neck_III_L	100	100
4.1.0 CT Models:	Structure:	Relevant FOV	Whole Body CT
	LN_Neck_III_R	100	100
	LN_Neck_IV_L	100	100
	LN_Neck_IV_R	100	95
	LN_Neck_V_L	100	100
	LN_Neck_V_R	100	91
	LN_Neck_VIA	100	100
	LN_Retropharynx_L	100	100
	LN_Retropharynx_R	95	100
	LN_Retrostyloid_L	100	100
	LN_Retrostyloid_R	100	100
Thorax	BrachialPlex_L	100	100
	BrachialPlex_R	100	100
	Breast_L	100	100
	Breast_R	96	91
	Bronchus	100	100
	Carina	98	100
	Cricoid	67	100
	Esophagus	100	100
	Glnd_Thyroid	100	100
	GreatVes	100	100
	Heart	100	100
	Humerus_Head_L	100	95
	Humerus_Head_R	100	100
	Kidney_L	97	91
	Kidney_R	97	95
	Larynx	99	95
	Liver	100	100
	Lung_L	100	100
	Lung_R	100	100
	Musc_Constrict	100	91
4.1.0 CT Models:	Structure:	Relevant FOV	Whole Body CT
	Pancreas	54	91
	SpinalCord	100	100
	Stomach	100	100
	Trachea	99	100
	LN_Ax_L1_L	100	100
	LN_Ax_L1_R	100	100
	LN_Ax_L2_L	100	100
	LN_Ax_L2_R	100	100
	LN_Ax_L3_L	100	100
	LN_Ax_L3_R	100	100
	LN_IMN_L	100	100
	LN_IMN_R	100	100
	LN_Sclav_L	100	100
	LN_Sclav_R	100	100
Whole Body -
Physiological Uptake
Organs	Bladder	N/A	95
	Bone	N/A	100
	Bowel	N/A	100
	BowelBag	N/A	100
	Brain	N/A	100
	Cavity_Oral	N/A	100
	Glnd_Lacrimal_L	N/A	100
	Glnd_Lacrimal_R	N/A	100
	Glnd_Submand_L	N/A	100
	Glnd_Submand_R	N/A	100
	Heart	N/A	100
	Kidney_L	N/A	95
	Kidney_R	N/A	100
	Liver	N/A	100
	LN_Iliac	N/A	64
4.1.0 CT Models:	Structure:	Relevant FOV	Whole Body CT
	Parotid_L	N/A	100
	Parotid_R	N/A	100
	Prostate	N/A	100
	Spleen	N/A	100

18

Image /page/18/Picture/0 description: The image shows the logo for MIM Software. The logo consists of two overlapping squares, one gray and one red, with a white circle where they overlap. To the right of the squares is the text "mim" in a bold, sans-serif font. Below the text "mim" is the text "SOFTWARE" in a smaller, sans-serif font.

F

19

Image /page/19/Picture/0 description: The image is a logo for MIM Software. The logo consists of two overlapping rounded squares, one gray and one red, with a white circle where they overlap. To the right of the squares is the text "mim" in a bold, sans-serif font, with the word "SOFTWARE" in a smaller font below it. The logo is clean and modern, with a simple color palette.

20

Image /page/20/Picture/0 description: The image is a logo for MIM Software. The logo consists of two overlapping squares, one gray and one red, with a white circle in the center where they overlap. To the right of the squares is the text "mim" in a bold, sans-serif font. Below the text "mim" is the word "SOFTWARE" in a smaller, sans-serif font.

T

21

Image /page/21/Picture/0 description: The image shows the logo for MIM Software. The logo consists of a red square with rounded corners, partially covered by a gray square with rounded corners. A white circle is visible where the two squares overlap. To the right of the squares, the text "mim" is written in a bold, sans-serif font, with the word "SOFTWARE" written in a smaller font below it.

Percentage of images that were successfully localized by Contour ProtégéAl. N/A (not applicable) refers to models which the relevant FOV is a Whole Body CT.

Known Limitations and Biases:

Contour ProtégéAl can produce incorrect or implausible segmentations when the underlying anatomy is in the field of view of the scan but not clearly discernable due to image quality, noise, or the generally low contrast of some structures. This can occur in any situation where the organ at risk is not clearly discernable in the image, but has been observed for submandibular glands, optic nerves, and optic chiasm, in particular.

While both images with and without IV contrast are represented in the training set for Contour ProtégéAl, the effect of iodinated contrast can vary due to patient weight, contrast dosage, and time since injection. In cases where there is unusually intense brain enhancement, the posterior boundary of the brain can be incorrectly segmented.

The age breakdown of the training data, where patient age was available, was predominantly in the 40-60 and 60+ age ranges. While this is appropriate to the intended patient population, only a small proportion of testing and training data was for patients under 40, so more intensive review may be warranted for younger patients.

Where patient sex was available, the ratio of male to female patients in the training set was roughly 2:1. The testing data was chosen with a more even ratio (about 0.82:1) to establish applicability of the model to both.

Conclusion

Based on the Discussion and Testing and Performance Data above, the proposed device is determined to be as safe and effective as its predicate device. Contour ProtégéAl 510 K223774. In addition, the proposed device performs as well as the reference predicate device. MIM 4.1 SEASTAR (tradename MIM Maestro) K071964.

References

1 Vaassen, F. et al. Evaluation of measures for assessing of automatic organ-at-risk segmentation in radiotherapy. Physics and Imaging in Radiation Oncology 13, 1-6 (2020).