Predicate Devices

Reference Devices

AI/ML (Artificial Intelligence / Machine Learning)

Yes
The document explicitly mentions "machine learning algorithm training and testing data" and "machine learning algorithm development".

Therapeutic

No.
The device is an image post-processing and measurement software tool that aids in identifying and measuring features from medical images, but it does not produce or recommend any medical diagnosis or treatment, nor does it provide therapy.

Diagnostic

No

The device explicitly states, "CoLumbo does not produce or recommend any type of medical diagnosis or treatment." It is an image post-processing and measurement tool that assists users in visualizing and quantifying features, with the user retaining responsibility for final medical judgment and diagnosis.

SaMD (Software as a Medical Device)

Yes

The device description explicitly states "CoLumbo is a medical device (software)". The functionality described is entirely software-based (image post-processing, measurement, segmentation, reporting) and there is no mention of accompanying hardware components that are part of the regulated device.

IVD (In Vitro Diagnostic)

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

Here's why:

IVD Definition: In Vitro Diagnostics are devices intended for use in the examination of specimens derived from the human body in order to provide information for diagnostic, monitoring, or compatibility purposes. This typically involves analyzing biological samples like blood, urine, tissue, etc.
CoLumbo's Function: CoLumbo processes and analyzes previously acquired medical images (MRI scans of the lumbar spine). It does not interact with or analyze biological specimens from the patient.
Intended Use: The intended use clearly states that CoLumbo is an "image post-processing and measurement software tool" that provides "quantitative spine measurements from previously-acquired DICOM lumbar spine Magnetic Resonance (MR) images." It explicitly states it "does not produce or recommend any type of medical diagnosis or treatment."
Device Description: The description reinforces that it's a "medical device (software) for viewing and interpreting magnetic resonance imaging (MRI)."

While CoLumbo is a medical device that provides information used in the diagnostic process, its function is based on analyzing medical images, not biological samples. Therefore, it falls outside the scope of an In Vitro Diagnostic device.

PCCP (Predetermined Change Control Plan) Authorized

No
The letter does not state that the FDA has reviewed and approved or cleared a PCCP for this specific device.

Overview

Intended Use / Indications for Use

CoLumbo is an image post-processing and measurement software tool that provides quantitative spine measurements from previously-acquired DICOM lumbar spine Magnetic Resonance (MR) images for users' review, and interpretation. It provides the following functionalizing, measuring and documenting and documenting out-of-range measurements:

Feature segmentation;
Feature measurement;
Threshold-based labeling of out-of-range measurement; and
Export of measurement results to a written report for user's review, revise and approval.

CoLumbo does not produce or recommend any type of medical diagnosis or treatment. Instead, it simply helps users to more easily identify and classify features in lumbar MR images and compile a report. The user is responsible for confirming/modifying settings, reviewing and verifying the software-generated measurements, inspecting out-of-range measurements, and approving draft report content using their medical judgment and discretion.

The device is intended to be used only by hospitals and other medical institutions.

Only DICOM images of MRI acquired from lumbar spine exams of patients aged 18 and above are considered to be valid input. CoLumbo does not support DICOM images of patients that are pregnant, undergo MRI scan with contrast media, or have post-operational complications, scoliosis, tumors, infections, fractures.

Product codes

QIH

Device Description

CoLumbo is a medical device (software) for viewing and interpreting magnetic resonance imaging (MRI) of the lumbar spine. The software is a quantitative imaging tool that assists radiologists and neuro- and spine surgeons ("users") to identify and measure lumbar spine features in medical images and record their observations in a report. The users then confirm whether the out-of-range measurements represent any true abnormality versus a spurious finding, such as an artifact or normal variation of the anatomy. The segmentation and measurements are classified using "modifiers" based on rule-based algorithms and thresholds set by each software user and stored in the user's individualized software settings. The user also identifies and classifies any other observations that the software may not annotate.

The purpose of CoLumbo is to provides information regarding common spine measurements confirmed by the user and the pre-determined thresholds confirmed or defined by the user. Every feature annotated by the software, based on the user-defined settings, must be reviewed and affirmed by the radiologist before the measurements of these features can be stored and reported. The software initiates adjustable measurements resulting from semi-automatic segmentation. If the user rejects a measurement the corresponding segmentation is rejected too. Segmentations are not intended to be a final output but serve the purpose of visualization and calculating measurements. The device outputs are intended to be a starting point for a clinical workflow and should not be interpreted or used as a diagnosis. The user is responsible for confirming segmentation and all measurement outputs. The output is an aid to the clinical workflow of measuring patient anatomy and should not be misused as a diagnosis tool.

User-confirmed defined settings control the sensitivity of the software for labelling measurements in an image. The user (not the software) controls the threshold for identifying out-of-range measurements, and, in every case once an out-of-range measurement is identified, the user must confirm or reject its presence. The software facilitates this process by annotating or drawing contours (segmentations) around features of the relevant anatomy and displaying measurements based on these contours. The user maintains control of the process by inspecting the segmentation, measurements and annotations upon which the measurements are based. The user may also examine other features of the imaging not annotated by the software to form a complete impression and diagnostic judgment of the overall state of disease, disorder, or trauma.

Mentions image processing

Yes

Mentions AI, DNN, or ML

Yes

Input Imaging Modality

MR

Anatomical Site

Lumbar Spine

Indicated Patient Age Range

18 and above

Intended User / Care Setting

Used only by hospitals and other medical institutions.
Intended users: Radiologist and neuro- and spine-surgeons.

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

The CoLumbo software machine learning algorithm training and testing data used during the algorithm development, as well as validation data used in the U.S. standalone software performance assessment study were all independent data sets.

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

The CoLumbo software machine learning algorithm training and testing data used during the algorithm development, as well as validation data used in the U.S. standalone software performance assessment study were all independent data sets.

Summary of Performance Studies

Standalone Software Performance Validation

To validate the CoLumbo software from a clinical perspective, a clinical data based standalone software performance assessment study was conducted in the U.S. The standalone software performance assessment study of CoLumbo included 101 MR image studies for 101 patients of different ages and racial groups, collected from seven (7) sites across the U.S. The standalone software performance assessment study compared the CoLumbo software outputs without any editing by a radiologist to the ground truth defined by 3 radiologists on segmentations and measurements.

Study Subjects:
Total Number of Subjects: 101
Gender – Male: 53 (52.5%)
Gender – Female: 48 (47.5%)
Age – 18 through 21: 3 (3.0%)
Age – 22 through 50: 74 (73.3%)
Age – 51 and above: 24 (23.8%)
Racial – Caucasian: 83 (82.2%)
Racial – Black/African American: 9 (8.9%)
Racial – Hispanic: 3 (3.0%)
Racial – American Indian: 3 (3.0%)
Racial – Others: 3 (3.0%)

Imaging Systems:
The 101 study images were acquired on MRI imaging systems made by five (5) manufacturers. All scans were conducted using the protocols standard for the investigational center, containing at least one axial and sagittal T2 series.
Toshiba (1.5T & 3.0T): 65 (64.4%)
Siemens (1.5T): 17 (16.8%)
Philips (1.5T): 1 (1.0%)
Hitachi (1.5T): 1 (1.0%)
GE (1.5T): 17 (16.8%)
Total: 101 (100%)

Ground Truth:
The ground truths for segmentations and measurements were independently established by three (3) U.S. radiologists without using the CoLumbo software. Each radiologist used a specialized pixel labeling tool to independently label the pixels of the tissues at the predetermined levels of the preselected axial and sagittal slices. The per-pixel majority opinion of the three (3) radiologists established the ground truth for each segmented tissue. Similarly, each radiologist used a commercial software tool to produce a standard set of areal, angular and linear measurements. The ground truth measurements were established by taking the median of three radiologists' measurements.

Acceptance Criteria and Results:
Primary endpoint (measurement accuracy):

the maximum Mean Absolute Error as defined as the upper limit of the 95% confidence interval for MAE is below a predetermined allowable error limit (MAEijmir) for each measurement listed.
Primary end point results: all primary endpoints were met.
Dural Sac Area (Axial): MAE 14.8 mm², 95% CI 12.4 - 17.3 mm², MAELimit 20 mm²
Lordotic Angle (Sagittal): MAE 2.6°, 95% CI 1.9 - 3.3°, MAELimit 6°
Listhesis/AP Slip (Sagittal): MAE 0.9 mm, 95% CI 0.8 - 1.1 mm, MAELimit 2 mm

Secondary endpoint (measurement and segmentation accuracy):

the maximum Mean Absolute Error, defined as the upper limit of the 95% confidence interval for -MAE, is below a predetermined allowable error limit (MAELimit) for each measurement listed.
the minimum Mean Dice Coefficient, defined as the lower limit of the 95% confidence interval for MDC, is above a predetermined allowable limit (MDC jmit) for each segmentation listed.
Secondary endpoint results: all secondary endpoints on measurement and segmentation were met.
Disc Material Outside IV Space (Axial): MAE 1.4 mm, 95% CI 1.1 - 1.6 mm, MAELimit 2 mm
Disc Material Migration (Sagittal): MAE 1.2 mm, 95% CI 1.0 - 1.4 mm, MAELimit 2 mm
Disc Material Bulge (Axial): MAE 1.0 mm, 95% CI 0.8 - 1.2 mm, MAELimit 2 mm
Dural Sac AP Diameter (Axial): MAE 1.0 mm, 95% CI 0.8 - 1.1 mm, MAELimit 2 mm
Intervertebral Angle (Sagittal): MAE 2.2°, 95% CI 1.9 - 2.5°, MAELimit 6°
Anterior VB Height (Sagittal): MAE 0.8 mm, 95% CI 0.7 - 0.9 mm, MAELimit 2 mm
Middle VB Height (Sagittal): MAE 0.8 mm, 95% CI 0.7 - 0.9 mm, MAELimit 2 mm
Posterior VB Height (Sagittal): MAE 1.0 mm, 95% CI 0.9 – 1.2 mm, MAELimit 2 mm
Anterior Disc Height (Sagittal): MAE 1.0 mm, 95% CI 0.7 – 1.0 mm, MAELimit 2 mm
Middle Disc Height (Sagittal): MAE 0.8 mm, 95% CI 0.7 – 0.9 mm, MAELimit 2 mm
Posterior Disc Height (Sagittal): MAE 1.1 mm, 95% CI 1.0 – 1.2 mm, MAELimit 2 mm

Tissue Segmentation:
Disc/Vertebral Body (Axial): MDC 0.97, 95% CI 0.96 - 0.97, MDCLimit 0.8
Vertebral Arch and Adjacent Ligaments (Axial): MDC 0.87, 95% CI 0.86 - 0.88, MDCLimit 0.8
Dural Sac (Axial): MDC 0.92, 95% CI 0.92 - 0.93, MDCLimit 0.8
Nerve Roots (Axial): MDC 0.75, 95% CI 0.72 - 0.78, MDCLimit 0.6
Disc Material Outside Intervertebral Space (Axial): MDC 0.76, 95% CI 0.72 - 0.80, MDCLimit 0.6
Disc (Sagittal): MDC 0.93, 95% CI 0.93 - 0.94, MDCLimit 0.8
Vertebral Body (Sagittal): MDC 0.95, 95% CI 0.94 - 0.95, MDCLimit 0.8
Sacrum S1 (Sagittal): MDC 0.93, 95% CI 0.92 - 0.94, MDCLimit 0.8
Disc Mat. Outside IV Space and/or Bulging Part (Sagittal): MDC 0.69, 95% CI 0.66 - 0.72, MDCLimit 0.6

CoLumbo was shown to produce measurements and segmentations accurate to within a prospectively-defined margin of error around the Ground Truth. This accuracy was preserved for all critical subgroups, including MRI scanner manufacturer, race, sex, and patient age.

MAE for software measurements and MDC for software segmentations by MRI scanner manufacturer:
Angular Measurements:
Toshiba: MAE 2.0°, 95% CI 1.8 - 2.3°, MAELimit 6°
Siemens: MAE 2.3°, 95% CI 1.6° - 3.0°, MAELimit 6°
GE: MAE 3.6°, 95% CI 1.7° - 5.6°, MAELimit 6°
Linear Measurements:
Toshiba: MAE 0.9 mm, 95% CI 0.9 - 1.0 mm, MAELimit 2 mm
Siemens: MAE 0.9 mm, 95% CI 0.8 - 1.0 mm, MAELimit 2 mm
GE: MAE 1.0 mm, 95% CI 0.9 - 1.1 mm, MAELimit 2 mm
Compression-related Tissue Segmentations:
Toshiba: MDC 0.73, 95% CI 0.71 - 0.76, MDCLimit 0.6
Siemens: MDC 0.69, 95% CI 0.63 - 0.75, MDCLimit 0.6
GE: MDC 0.76, 95% CI 0.71 - 0.81, MDCLimit 0.6
Other Tissue Segmentations:
Toshiba: MDC 0.93, 95% CI 0.92 - 0.93, MDCLimit 0.8
Siemens: MDC 0.93, 95% CI 0.92 - 0.93, MDCLimit 0.8
GE: MDC 0.94, 95% CI 0.93 - 0.95, MDCLimit 0.8

MAE for software measurements and MDC for software segmentations by race:
Angular Measurements:
White: MAE 2.4°, 95% CI 1.9 - 2.9°, MAELimit 6°
Non-White: MAE 2.3°, 95% CI 1.7 - 2.9°, MAELimit 6°
Linear Measurements:
White: MAE 0.9 mm, 95% CI 0.9 - 1.0 mm, MAELimit 2 mm
Non-White: MAE 1.1 mm, 95% CI 1.0 - 1.2 mm, MAELimit 2 mm
Compression-related Tissue Segmentations:
White: MDC 0.73, 95% CI 0.71 - 0.75, MDCLimit 0.6
Non-White: MDC 0.72, 95% CI 0.67 - 0.78, MDCLimit 0.6
Other Tissue Segmentations:
White: MDC 0.93, 95% CI 0.925 - 0.933, MDCLimit 0.8
Non-White: MDC 0.93, 95% CI 0.92 - 0.94, MDCLimit 0.8

MAE for software measurements and MDC for software segmentations by gender:
Angular Measurements:
Male: MAE 2.3°, 95% CI 1.7 - 3.1°, MAELimit 6°
Female: MAE 2.4°, 95% CI 2.1 - 2.7°, MAELimit 6°
Linear Measurements:
Male: MAE 0.8 mm, 95% CI 0.9 - 1.0 mm, MAELimit 2 mm
Female: MAE 0.9 mm, 95% CI 0.9 - 1.0 mm, MAELimit 2 mm
Compression-related Tissue Segmentations:
Male: MDC 0.75, 95% CI 0.72 - 0.78, MDCLimit 0.6
Female: MDC 0.71, 95% CI 0.68 - 0.74, MDCLimit 0.6

Performance by age group:
18-21 years old:
Angle-based Measurements: 1.14°, Std Dev 1.41°, Num Samples 6, CI (0.48 – 1.81°), Acceptance Criteria 6°
Linear Measurements: 0.83 mm, Std Dev 1.06mm, Num Samples 36, CI (0.60 – 1.05mm), Acceptance Criteria 2 mm
Compression-related Tissue Segmentations: 0.76, Std Dev 0.087, Num Samples 7, CI (0.70 – 0.83), Acceptance Criteria 0.6
Other Tissue Segmentations: 0.92, Std Dev 0.039, Num Samples 18, CI (0.91 – 0.94), Acceptance Criteria 0.8
22-50 years old:
Angle-based Measurements: 2.43°, CI (1.94–2.93°), Acceptance Criteria 6°
Linear Measurements: 0.95 mm, CI (0.90–1.00mm), Acceptance Criteria 2 mm
Compression-related Tissue Segmentations: 0.73, CI (0.71 – 0.76), Acceptance Criteria 0.6
Other Tissue Segmentations: 0.93, CI (0.926–0.934), Acceptance Criteria 0.8
51 years old and above:
Angle-based Measurements: 2.09°, CI (1.63 – 2.56°), Acceptance Criteria 6°
Linear Measurements: 0.97 mm, CI (0.87 – 1.06mm), Acceptance Criteria 2 mm
Compression-related Tissue Segmentations: 0.72, CI (0.68 – 0.76), Acceptance Criteria 0.6
Other Tissue Segmentations: 0.93, CI (0.92 – 0.93), Acceptance Criteria 0.8

Key Metrics

Mean Absolute Error (MAE), Mean Dice Coefficient (MDC)

Predicate Device(s)

K193290

Reference Device(s)

K183268, K193267

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

Image /page/0/Picture/0 description: The image shows the logo of the U.S. Food and Drug Administration (FDA). On the left is the Department of Health & Human Services logo. To the right of that is the FDA logo in blue, with the words "U.S. FOOD & DRUG ADMINISTRATION" in blue as well. The FDA is a federal agency responsible for regulating and supervising the safety of food, drugs, and other products.

Smart Soft Healthcare AD % Yu Zhao Strategy Advisor LightSource Research LLC 2108 N St., Suite N SACRAMENTO CA 95816

Re: K220497

June 23, 2022

Trade/Device Name: CoLumbo Regulation Number: 21 CFR 892.2050 Regulation Name: Medical image management and processing system Regulatory Class: Class II Product Code: QIH Dated: May 31, 2022 Received: June 1, 2022

Dear Yu Zhao:

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

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

Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801 and Part 809); medical device reporting of medical device-related adverse events) (21 CFR

1

for devices or postmarketing safety reporting (21 CFR 4, Subpart B) for combination products (see https://www.fda.gov/combination-products/guidance-regulatory-information/postmarketing-safety-reportingcombination-products); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820) for devices or current good manufacturing practices (21 CFR 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.

Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR Part 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to https://www.fda.gov/medical-device-safety/medical-device-reportingmdr-how-report-medical-device-problems.

For comprehensive regulatory information about medical devices and radiation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/medicaldevices/device-advice-comprehensive-regulatory-assistance) and CDRH Learn (https://www.fda.gov/training-and-continuing-education/cdrh-learn). Additionally, you may contact the Division of Industry and Consumer Education (DICE) to ask a question about a specific regulatory topic. See the DICE website (https://www.fda.gov/medical-device-advice-comprehensive-regulatoryassistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).

Sincerely,

Jessica Lamb. Ph.D. Assistant Director Imaging Software Team DHT8B: Division of Radiological Imaging Devices and Electronic Products OHT8: Office of Radiological Health Office of Product Evaluation and Quality Center for Devices and Radiological Health

Enclosure

2

Indications for Use

510(k) Number (if known) K220497

Device Name CoLumbo

Indications for Use (Describe)

CoLumbo is an image post-processing and measurement software tool that provides quantitative spine measurements from previously-acquired DICOM lumbar spine Magnetic Resonance (MR) images for users' review, and interpretation. It provides the following functionalizing, measuring and documenting and documenting out-ofrange measurements:

Feature segmentation;
Feature measurement;
Threshold-based labeling of out-of-range measurement; and
Export of measurement results to a written report for user's review, revise and approval.

CoLumbo does not produce or recommend any type of medical diagnosis or treatment. Instead, it simply helps users to more easily identify and classify features in lumbar MR images and compile a report. The user is responsible for confirming/modifying settings, reviewing and verifying the software-generated measurements, inspecting out-of-range measurements, and approving draft report content using their medical judgment and discretion.

The device is intended to be used only by hospitals and other medical institutions.

Only DICOM images of MRI acquired from lumbar spine exams of patients aged 18 and above are considered to be valid input. CoLumbo does not support DICOM images of patients that are pregnant, undergo MRI scan with contrast media, or have post-operational complications, scoliosis, tumors, infections, fractures.

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 comblete 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

510(k) Summary

1. Submitter

Smart Soft Healthcare AD 113 General Kolev Str., Primorski Distri., Office 7.2 Varna 9002, Bulgaria Phone: +35952919513 Fax: None Contact Person: Nedelcho Georgiev Date Prepared: February 5, 2022

2. Device

Name of Device: CoLumbo Common or Usual Name: CoLumbo Classification Name: Medical image management and processing system (21 CFR 892.2050) Product Code: OIH Regulatory Class: II

3. Predicate Devices

Predicate Device:

Device Name: AI-Rad Companion Brain MR Manufacturer: Siemens Healthcare GmbH Classification Name: Medical image management and processing system (21 CFR 892.2050) Secondary Classification Name: Magnetic resonance diagnostic device Classification Product Code: LLZ Subsequent Product Code: LNH Classification Panel: Radiology Device Class: Class II 510(k) Number: K193290 cleared July 5, 2019

Reference Device 1:

Device Name: AI-Rad Companion (Cardiovascular) Manufacturer: Siemens Medical Solutions USA, Inc. Classification Name: Computed tomography x-ray system Regulation Number: 21 CFR 892.1750 Classification Product Code: JAK Subsequent Product Code: LLZ Classification Panel: Radiology Device Class: Class II

4

510(k) Number: K183268 cleared September 10, 2019

Reference Device 2:

Device Name: AI-Rad Companion (Musculoskeletal) Manufacturer: Siemens Medical Solutions USA, Inc. Classification Name: Computed tomography x-ray system Regulation Number: 21 CFR 892.1750 Classification Product Code: JAK Classification Panel: Radiology Device Class: Class II 510(k) Number: K193267 cleared March 16, 2020

4. Device Description

CoLumbo is a medical device (software) for viewing and interpreting magnetic resonance imaging (MRI) of the lumbar spine. The software is a quantitative imaging tool that assists radiologists and neuro- and spine surgeons ("users") to identify and measure lumbar spine features in medical images and record their observations in a report. The users then confirm whether the out-of-range measurements represent any true abnormality versus a spurious finding, such as an artifact or normal variation of the anatomy. The segmentation and measurements are classified using "modifiers" based on rule-based algorithms and thresholds set by each software user and stored in the user's individualized software settings. The user also identifies and classifies any other observations that the software may not annotate.

The purpose of CoLumbo is to provides information regarding common spine measurements confirmed by the user and the pre-determined thresholds confirmed or defined by the user. Every feature annotated by the software, based on the user-defined settings, must be reviewed and affirmed by the radiologist before the measurements of these features can be stored and reported. The software initiates adjustable measurements resulting from semi-automatic segmentation. If the user rejects a measurement the corresponding segmentation is rejected too. Segmentations are not intended to be a final output but serve the purpose of visualization and calculating measurements. The device outputs are intended to be a starting point for a clinical workflow and should not be interpreted or used as a diagnosis. The user is responsible for confirming segmentation and all measurement outputs. The output is an aid to the clinical workflow of measuring patient anatomy and should not be misused as a diagnosis tool.

User-confirmed defined settings control the sensitivity of the software for labelling measurements in an image. The user (not the software) controls the threshold for identifying out-of-range measurements, and, in every case once an out-of-range measurement is identified, the user must confirm or reject its presence. The software facilitates this process by annotating or drawing contours (segmentations) around features of the relevant anatomy and displaying measurements based on these contours. The user maintains control of the process by inspecting the segmentation, measurements and annotations upon which the measurements are based. The user may also examine other features of the imaging not annotated by the software to form a complete impression and diagnostic judgment of the overall state of disease, disorder, or trauma.

5. Indications for Use

5

CoLumbo is an image post-processing and measurement software tool that provides quantitative spine measurements from previously-acquired DICOM lumbar spine Magnetic Resonance (MR) images for users' review, analysis, and interpretation. It provides the following functionality to assist users in visualizing, measuring and documenting out-of-range measurements:

. Feature segmentation;
. Feature measurement;
. Threshold-based labeling of out-of-range measurement; and
. Export of measurement results to a written report for user's revise and approval.

CoLumbo does not produce or recommend any type of medical diagnosis or treatment. Instead, it simply helps users to more easily identify and classify features in lumbar MR images and compile a report. The user is responsible for confirming/modifying settings. reviewing and verifying the software-generated measurements, inspecting out-of-range measurements, and approving draft report content using their medical judgment and discretion.

The device is intended to be used only by hospitals and other medical institutions.

Only DICOM images of MRI acquired from lumbar spine exams of patients aged 18 and above are considered to be valid input. CoLumbo does not support DICOM images of patients that are prognant, undergo MRI scan with contrast media, or have post-operational complications, scoliosis, tumors, infections, fractures.

6. Comparison of the Technological Characteristics with the Predicate Devices

In comparison to the Predicate Device and the Reference Devices, the Subject Device provides comparable outputs in terms of segmentation. measurement and labeling. A tabular high-level comparison of the Subject Device, the Predicate Device and the Reference Devices is provided as Table 8.1 below.

6

	Predicate Device -	Reference Device 1 -	Reference Device 2 -	Subject Device	Remark/Discussion
	AI-Rad Companion	AI-Rad Companion	AI-Rad Companion	CoLumbo
	Brain MR	(Cardiovascular)	(Musculoskeletal)
	(K193290)	(K183268)	(K193267)
Intended User	Radiologist	Radiologists &
Physicians from
emergency medicine,
specialty care, urgent
care, and general
practice	Radiologists &
Physicians from
emergency medicine,
specialty care, urgent
care, and general
practice	Radiologist and neuro-
and spine-surgeons	Highly similar
Intended Patient
Population	The intended patient
target group consists
of patients of age 2
years or higher. In
this age range the
brain segmentation
algorithm
works properly.	The intended patient
population is not
subject to any
restrictions.
Automation support
requires images of
patients of 22 years
and older.	The intended patient
population is not
subject to any
restrictions.
Automation support
requires images of
patients of 22 years
and older.	The intended patient
population is not subject
to any restrictions.
Automation support
requires images of
patients of 18 years and
older, not pregnant,
without post-operational
complications, scoliosis,
tumors, infections,
fractures.	Similar
Supported Body
Part	Brain	Thorax	Thorax (including
thoracic spine)	Lumbar Spine	Similar to Reference
Devices:
Different from Primary
Device
Segmentation	Yes
Segmentation and
quantitative analysis	Yes
Segmentation and
quantitative analysis	Yes
Segmentation of
vertebrae	Yes
Segmentation and
quantitative analysis	Same
Measurement	Yes
Quantitative
comparison of
structure with
normative data or
user-set thresholds	Yes
Volume
measurement of the
heart, total calcium
volume in the
coronary arteries.	Yes
Measure Hounsfield
values within the
vertebras	Yes
Quantitative comparison
of structure with
normative data or user-set
thresholds	Same
	Predicate Device -
AI-Rad Companion
Brain MR
(K193290)	Reference Device 1 -
AI-Rad Companion
(Cardiovascular)
(K183268)	Reference Device 2 -
AI-Rad Companion
(Musculoskeletal)
(K193267)	Subject Device
CoLumbo	Remark/Discussion
Threshold-Based
Out-of-Range
Measurements	Yes
Quantitative
comparison of
structure with
normative data or
user-set thresholds	Yes
Threshold-based
highlighting of
findings.
Classify each finding
(e.g., enlarged
diameters) by
comparing
measurements
against user-set
thresholds	Yes
Labeling of vertebras
based on the
individual heights of
the vertebras and
whether they
critically differ from
their direct neighbors	Yes
Quantitative comparison
of structure with
normative data or user-set
thresholds	Similar
Reporting	Yes
Exportation of results
with the findings for
further reporting	Yes
Exportation of results
with the findings for
further reporting	Yes
Exportation of results
with the findings for
further reporting	Yes
Exportation of results with
the measurements for
further reporting	Same
None of the reports are
to be used as final
reports. Trained
radiologist or neuro-
and spine-surgeons
need to review, edit and
approve the final report
SaMD	Yes	Yes	Yes	Yes	Same
Algorithm	(no information)	3D Deep Image-to-
Image Network	3D Deep Image-to-
Image Network	Deep Convolutional
Image-to-Image Neural
Network	Similar
Supported
Modality	MR	CT	CT	MR	Same as Primary
Predicate Device 1;
Similar to Reference
Devices

Table 8.1 – Comparison of Technological Characteristics with Predicate/Reference Devices

7

8

The Subject Device is substantially equivalent in comparison to the Predicate/Reference Devices. The information regarding the Subject Device do not raise new questions about safety and effectiveness and demonstrate that CoLumbo is at least as safe and effective as the legally marketed devices.

7. Performance Data

7.1.Biocompatibility Testing

Not applicable.

7.2.Electrical Safety and Electromagnetic Compatibility (EMC)

Not applicable.

7.3.Animal Study

Not applicable.

7.4.Voluntary Conformance Standards

CoLumbo has been tested to meet the requirements of conformity to multiple industry standards. Non-clinical performance testing demonstrated that CoLumbo complies with the following voluntary FDA recognized Consensus Standards listed in Table 8.2 below.

| Recognition

| Standard |

|------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------|
| 13-79 | IEC 62304:2006/AMD 1:2015 Medical device software — Software life cycle
processes — Amendment 1 |
| 5-125 | ISO 14971:2019 Medical devices — Application of risk management to medical
devices |
| 5-129 | IEC 62366-1:2015+AMD1:2020 Medical devices — Part 1: Application of usability
engineering to medical devices |
| 5-117 | ISO 15223-1:2016 Medical devices — Symbols to be used with medical device
labels, labelling and information to be supplied — Part 1: General requirements |
| 12-300 | NEMA PS 3.1 - 3.20 (2016) Digital Imaging and Communications in Medicine
(DICOM) Set |

Table 8.2 - Voluntary Conformance Standards

7.5.Nonclinical Tests

Smart Soft Healthcare has performed software design verification testing and has sponsored external standalone performance assessment study. The performance data demonstrates continued conformance with special controls for medical devices containing software.

Software documentation for a Moderate Level of Concern software, per FDA Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices, issued on May 11, 2005, were provided. Smart Soft Healthcare has conducted software verification and validation, in accordance with the FDA guidance. General Principles of Software Validation: Final Guidance for

9

Industry and FDA Staff, issued on January 11, 2002. All software reguirements and risk analysis have been successfully verified and traced.

Smart Soft Healthcare conforms to the cybersecurity requirements by implementing a process of preventing unauthorized access, modifications, misuse or the unauthorized use of information that is stored, accessed or transferred from a medical device to an external recipient. The vulnerability assessment and penetration testing demonstrates satisfactory security performance.

In addition to the human factors validation, Smart Soft Healthcare conducted a standalone software performance study in the U.S. to validate clinical performance of the CoLumbo software.

The nonclinical test data demonstrated conformance with special controls and substantial equivalence to predicate devices' performance.

Standalone Software Performance Validation

To validate the CoLumbo software from a clinical perspective, a clinical data based standalone software performance assessment study was conducted in the U.S. The standalone software performance assessment study of CoLumbo included 101 MR image studies for 101 patients of different ages and racial groups, collected from seven (7) sites across the U.S. The standalone software performance assessment study compared the CoLumbo software outputs without any editing by a radiologist to the ground truth defined by 3 radiologists on segmentations and measurements.

	Number of Subjects	Percent of Total
Total Number of Subjects	101	100%
Gender – Male	53	52.5%
Gender - Female	48	47.5%
Age – 18 through 21	3	3.0%
Age - 22 through 50	74	73.3%
Age - 51 and above	24	23.8%
Racial - Caucasian	83	82.2%
Racial - Black/African American	9	8.9%
Racial – Hispanic	3	3.0%
Racial - American Indian	3	3.0%
Racial - Others	3	3.0%

Study Subjects:

Imaging Systems:

The 101 study images were acquired on MRI imaging systems made by five (5) manufacturers. All scans were conducted using the protocols standard for the investigational center, containing at least one axial and sagittal T2 series.

Manufacturer	Number of MRI Exams Collected	Percent of Total
Toshiba (1.5T & 3.0T)	65	64.4%
Siemens (1.5T)	17	16.8%
Philips (1.5T)	1	1.0%
Hitachi (1.5T)	1	1.0%
GE (1.5T)	17	16.8%

10

Total	101	100%

Ground Truth:

The ground truths for segmentations and measurements were independently established by three (3) U.S. radiologists without using the CoLumbo software. Each radiologist used a specialized pixel labeling tool to independently label the pixels of the tissues at the predetermined levels of the preselected axial and sagittal slices. The per-pixel majority opinion of the three (3) radiologists established the ground truth for each segmented tissue. Similarly, each radiologist used a commercial software tool to produce a standard set of areal, angular and linear measurements. The ground truth measurements were established by taking the median of three radiologists' measurements.

Acceptance Criteria and Results:

Primary endpoint (measurement accuracy):

the maximum Mean Absolute Error as defined as the upper limit of the 95% confidence interval for MAE is below a predetermined allowable error limit (MAEijmir) for each measurement listed.
Primary end point results: all primary endpoints were met.

Measurement	Mean Absolute	95% Confidence Interval (CI)	MAELimit
	Error (MAE)
Dural Sac Area (Axial)	14.8 mm²	12.4 - 17.3 mm²	20 mm²
Lordotic Angle (Sagittal)	2.6°	1.9 - 3.3°	6°
Listhesis/AP Slip (Sagittal)	0.9 mm	0.8 - 1.1 mm	2 mm

Secondary endpoint (measurement and segmentation accuracy):

the maximum Mean Absolute Error, defined as the upper limit of the 95% confidence interval for -MAE, is below a predetermined allowable error limit (MAELimit) for each measurement listed.
the minimum Mean Dice Coefficient, defined as the lower limit of the 95% confidence interval for MDC, is above a predetermined allowable limit (MDC jmit) for each segmentation listed.

Secondary endpoint results: all secondary endpoints on measurement and segmentation were met.

| Measurement | Mean Absolute
Error (MAE) | 95% Confidence Interval (CI) | MAELimit |
|-------------------------------------------|------------------------------|------------------------------|----------|
| Disc Material Outside
IV Space (Axial) | 1.4 mm | 1.1 - 1.6 mm | 2 mm |
| Disc Material
Migration (Sagittal) | 1.2 mm | 1.0 - 1.4 mm | 2 mm |
| Disc Material Bulge
(Axial) | 1.0 mm | 0.8 - 1.2 mm | 2 mm |
| Dural Sac AP
Diameter (Axial) | 1.0 mm | 0.8 - 1.1 mm | 2 mm |
| Intervertebral Angle
(Sagittal) | 2.2° | 1.9 - 2.5° | 6° |
| Anterior VB Height
(Sagittal) | 0.8 mm | 0.7 - 0.9 mm | 2 mm |
| Middle VB Height | 0.8 mm | 0.7 - 0.9 mm | 2 mm |

11

| (Sagittal)

Posterior VB Height	1.0 mm	0.9 – 1.2 mm	2 mm
(Sagittal)
Anterior Disc Height	1.0 mm	0.7 – 1.0 mm	2 mm
(Sagittal)
Middle Disc Height	0.8 mm	0.7 – 0.9 mm	2 mm
(Sagittal)
Posterior Disc Height	1.1 mm	1.0 – 1.2 mm	2 mm

| Tissue Segmentation | Mean Dice
Coefficient (MDC) | 95% Confidence
Interval (CI) | MDCLimit |
|--------------------------------------------------------------|--------------------------------|---------------------------------|----------|
| Disc/Vertebral Body (Axial) | 0.97 | 0.96 - 0.97 | 0.8 |
| Vertebral Arch and
Adjacent Ligaments (Axial) | 0.87 | 0.86 - 0.88 | 0.8 |
| Dural Sac (Axial) | 0.92 | 0.92 - 0.93 | 0.8 |
| Nerve Roots (Axial) | 0.75 | 0.72 - 0.78 | 0.6 |
| Disc Material Outside
Intervertebral Space (Axial) | 0.76 | 0.72 - 0.80 | 0.6 |
| Disc (Sagittal) | 0.93 | 0.93 - 0.94 | 0.8 |
| Vertebral Body (Sagittal) | 0.95 | 0.94 - 0.95 | 0.8 |
| Sacrum S1 (Sagittal) | 0.93 | 0.92 - 0.94 | 0.8 |
| Disc Mat. Outside IV Space
and/or Bulging Part (Sagittal) | 0.69 | 0.66 - 0.72 | 0.6 |

CoLumbo was shown to produce measurements and segmentations accurate to within a prospectively-defined margin of error around the Ground Truth. This accuracy was preserved for all critical subgroups, including MRI scanner manufacturer, race, sex, and patient age.

MAE for software measurements and MDC for software segmentations by MRI scanner manufacturer:

| Measurement

Type	MAE			95% Confidence Intervals			MAELimit
	Toshiba	Siemens	GE	Toshiba	Siemens	GE
Angular
Measurements	2.0°	2.3°	3.6°	1.8 - 2.3°	1.6° - 3.0°	1.7° - 5.6°	6°
Linear
Measurements	0.9 mm	0.9 mm	1.0 mm	0.9 - 1.0 mm	0.8 - 1.0 mm	0.9 - 1.1 mm	2 mm
Segmentation
Type	MDC			95% Confidence Intervals			MDCLimit
	Toshiba	Siemens	GE	Toshiba	Siemens	GE
Compression-
related Tissue
Segmentations	0.73	0.69	0.76	0.71 - 0.76	0.63 - 0.75	0.71 - 0.81	0.6
Other Tissue
Segmentations	0.93	0.93	0.94	0.92 - 0.93	0.92 - 0.93	0.93 - 0.95	0.8

MAE for software measurements and MDC for software segmentations by race:

Measurement Type	MAE	MAE	95% Confidence Intervals	MAE_{Limit}
------------------	-----	-----	--------------------------	-------------

12

	White	Non-White	White	Non-White
Angular
Measurements	2.4°	2.3°	1.9 - 2.9°	1.7 - 2.9°	6°
Linear Measurements	0.9 mm	1.1 mm	0.9 - 1.0 mm	1.0 - 1.2 mm	2 mm
Segmentation Type	MDC
White	MDC
Non-White	95% Confidence Intervals
White	95% Confidence Intervals
Non-White	MDCLimit
Compression-related
Tissue Segmentations	0.73	0.72	0.71 - 0.75	0.67 - 0.78	0.6
Other Tissue
Segmentations	0.93	0.93	0.925 - 0.933	0.92 - 0.94	0.8

MAE for software measurements and MDC for software segmentations by gender:

| Measurement Type | MAE
Male | MAE
Female | 95% Confidence
Intervals | | MAELimit |
|---------------------------------------------|-------------|---------------|-----------------------------|--------------|----------|
| Angular
Measurements | 2.3° | 2.4° | 1.7 - 3.1° | 2.1 - 2.7° | 6° |
| Linear Measurements | 0.8 mm | 0.9 mm | 0.9 - 1.0 mm | 0.9 - 1.0 mm | 2 mm |
| Segmentation Type | MDC
Male | MDC
Female | 95% Confidence
Intervals | | MDCLimit |
| Compression-related
Tissue Segmentations | 0.75 | 0.71 | 0.72 - 0.78 | 0.68 - 0.74 | 0.6 |

The following tables represent MAE and DICE statistics for each of the age groups in the aforementioned table.

| Tissue Segmentation Type
or Measurement | Acceptance
Criteria | Between
18 and
21 years
old | Standard
Deviation | Number
of
Samples | Confidence
Interval |
|---------------------------------------------|------------------------|--------------------------------------|-----------------------|-------------------------|------------------------|
| Angle-based Measurements | 6° | 1.14° | 1.41° | 6 | (0.48 – 1.81°) |
| Linear Measurements | 2 mm | 0.83 mm | 1.06mm | 36 | (0.60 –
1.05mm) |
| Compression-related Tissue
Segmentations | 0.6 | 0.76 | 0.087 | 7 | (0.70 – 0.83) |
| Other Tissue Segmentations | 0.8 | 0.92 | 0.039 | 18 | (0.91 – 0.94) |
| Tissue Segmentation Type
or Measurement | Acceptance
Criteria | Between 22 and
50 years old | | Confidence Interval | |
| Angle-based Measurements | 6° | 2.43° | | (1.94–2.93°) | |
| Linear Measurements | 2 mm | 0.95 mm | | (0.90–1.00mm) | |
| Compression-related Tissue
Segmentations | 0.6 | 0.73 | | (0.71 – 0.76) | |
| Other Tissue Segmentations | 0.8 | 0.93 | | (0.926–0.934) | |

13

| Tissue Segmentation Type
or Measurement | Acceptance
Criteria | 51 years old and
above | Confidence Interval |
|---------------------------------------------|------------------------|---------------------------|---------------------|
| Angle-based Measurements | 6° | 2.09° | (1.63 – 2.56°) |
| Linear Measurements | 2 mm | 0.97 mm | (0.87 – 1.06mm) |
| Compression-related Tissue
Segmentations | 0.6 | 0.72 | (0.68 – 0.76) |
| Other Tissue Segmentations | 0.8 | 0.93 | (0.92 – 0.93) |

Training, Testing and Validation Data Independence:

The CoLumbo software machine learning algorithm training and testing data used during the algorithm development, as well as validation data used in the U.S. standalone software performance assessment study were all independent data sets.

7.6.Clinical Validation Study

No human clinical study was conducted to support the pre-market clearance.

8. Conclusions

The CoLumbo software is as safe and effective as the predicate device. The subject device has the same intended uses and similar indications, technological characteristics, and principles of operation as its predicate device. The minor differences between subject and predicate device in indications do not alter the intended use of the device and do not raise new or different questions regarding its safety and effectiveness when used as labeled.

The software verification and validation testing data, including the standalone software performance assessment study data, support the safety of the devices and demonstrate that the CoLumbo software performs as intended in the specified use conditions.

Therefore, the CoLumbo software is substantially equivalent.