Ortho AI is an image-processing software indicated to assist in making measurements for a total hip arthroplasty, total knee arthroplasty, and lumbar spine fusion surgery.

It is intended to assist in the measurement of x-ray images by measuring lengths, angles and position of implants relative to the bone structures of interest provided, that the points of interest can be identified from radiology images.

The device allows for overlaying of digital annotations on radiological images and includes tools for performing measurements using the images and digital annotations. The software is not for primary image interpretation. The software is not for use on mobile phones.

Intended patient population: Adult patients >=22 years old, with appropriate imaging, undergoing primary hip replacement, primary knee replacement, and lumbar spine surgery.

Intended user population: orthopaedic surgeons who perform hip and knee replacement, and orthopaedic/neurosurgeons who perform lumbar spine surgery.

Device Description

Ortho AI is a software as a medical device (SaMD) system that provides preoperative planning data for hip replacement surgery, knee replacement surgery, and lumbar spinal fusion surgery using AI/ML models that are semi-automated and interpretable. The software guides the user through a predetermined workflow that begins with the use of preoperative radiographic images as input to the software. As part of this initial preoperative workflow, the software places digital annotations on these preoperative images, which can be modified by the user (semi-automated). The software additionally includes functionality to store user, patient, and case information.

AI/ML Overview

Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

Module	Acceptance Criteria (Dice Coefficient)	Reported Performance (Dice Coefficient)	Other Performance Metrics Reported
Overall	Minimum 0.85 for all algorithms and connected domains	All connected domains above 0.85	See specific model performance
Hip Model	Minimum 0.85 for all connected domains	All connected domains above 0.85	LLD: within +/- 1.96mm of human measurement Offset (global): within +/- 0.88mm of human measurement SFP angle: within +/- 1.05mm of human measurement
Hip-Spine Model	Minimum 0.85 for all connected domains	All connected domains above 0.85	SS, SPT, APPt, PI, LL, PI-LL: all within 2 degrees of human measurement (no statistical difference)
Knee Model	Minimum 0.85 for all connected domains	All connected domains above 0.85	LDFA, mPTA, aHKA, aJLOA: all within 2 degrees of human measurement (no statistical difference)

2. Sample Size for the Test Set and Data Provenance

Sample Size for Test Set: The acceptance criteria section states, "A test sample size of ≥ 150 samples". While the exact number of test images for each model's independent test set isn't explicitly detailed within the provided text, the document implies that this minimum was met or exceeded for the performance testing.
Data Provenance: The text does not explicitly state the country of origin of the data. It also does not specify whether the data was retrospective or prospective.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications

Number of Experts: 3
Qualifications of Experts: All 3 were fellowship-trained, ABOS board-certified orthopaedic surgeons, each with greater than 10 years of experience.

4. Adjudication Method for the Test Set

Adjudication Method: 2+1 truthing process. Two blinded orthopaedic surgeons independently reviewed each segmented image in the testing set and applied modifications. A final senior-level surgeon adjudicator then reviewed these modifications and made further adjustments if necessary.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

The provided text does not indicate that a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was done to evaluate how much human readers improve with AI vs. without AI assistance. The study focuses on the standalone performance of the algorithm against human measurements and ground truth.

6. Standalone Performance

Yes, standalone performance was evaluated. The section "Standalone algorithm testing" directly addresses this, providing specific measurement accuracies for the Hip, Hip-Spine, and Knee models (e.g., LLD, Offset, angles) and reporting Dice coefficients. The statement "no statistical difference between human vs. machine learning measurements" for Hip-spine and Knee models further confirms this.

7. Type of Ground Truth Used

Expert Consensus / Human Measurement: The ground truth for the test set was established through a "2+1 truthing process" by three qualified orthopaedic surgeons, which constitutes expert consensus. The device's performance is reported against "human measurement" for various parameters (e.g., LLD measurements within +/- 1.96mm of human measurement).

8. Sample Size for the Training Set

Hip Model: 1,367 images from 1,367 patients
Hip-Spine Model: 4,836 images from 4,836 patients
Knee Model: 4,536 images from 4,536 patients

9. How the Ground Truth for the Training Set Was Established

The text describes the "Truthing process" for the test set. While the methodology for the training set's ground truth is not explicitly detailed in the provided snippet, it is a standard practice that such large training datasets would also have their ground truth established by experts, likely following a similar or more extensive version of the described truthing process. However, the document only explicitly explains the ground truthing for the test set.

Summary

{0}------------------------------------------------

Image /page/0/Picture/0 description: The image contains the logos of the Department of Health and Human Services and the U.S. Food & Drug Administration (FDA). The Department of Health and Human Services logo is on the left, and the FDA logo is on the right. The FDA logo is a blue square with the letters "FDA" in white, followed by the words "U.S. Food & Drug Administration" in blue.

Ortho AI LLC Jonathan Vigdorchik Orthopaedic Surgeon 155 East 76th Street Apt 2H New York, NY 10021

January 2, 2025

Re: K241696

Trade/Device Name: Ortho AI Regulation Number: 21 CFR 892.2050 Regulation Name: Medical Image Management And Processing System Regulatory Class: Class II Product Code: QIH Dated: November 26, 2024 Received: November 27, 2024

Dear Jonathan Vigdorchik:

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"

{1}------------------------------------------------

(https://www.fda.gov/media/99812/download) and "Deciding When to Submit a 510(k) for a Software Change to an Existing Device" (https://www.fda.gov/media/99785/download).

Your device is also subject to, among other requirements, the Quality System (QS) regulation (21 CFR Part 820), which includes, but is not limited to, 21 CFR 820.30, Design controls; 21 CFR 820.90, Nonconforming product; and 21 CFR 820.100, Corrective and preventive action. Please note that regardless of whether a change requires premarket review, the QS regulation requires device manufacturers to review and approve changes to device design and production (21 CFR 820.30 and 21 CFR 820.70) and document changes and approvals in the device master record (21 CFR 820.181).

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

All medical devices, including Class I and unclassified devices and combination product device constituent parts are required to be in compliance with the final Unique Device Identification System rule ("UDI Rule"). The UDI Rule requires, among other things, that a device bear a unique device identifier (UDI) on its label and package (21 CFR 801.20(a)) unless an exception or alternative applies (21 CFR 801.20(b)) and that the dates on the device label be formatted in accordance with 21 CFR 801.18. The UDI Rule (21 CFR 830.300(a) and 830.320(b)) also requires that certain information be submitted to the Global Unique Device Identification Database (GUDID) (21 CFR Part 830 Subpart E). For additional information on these requirements, please see the UDI System webpage at https://www.fda.gov/medical-device-advicecomprehensive-regulatory-assistance/unique-device-identification-system-udi-system.

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

For comprehensive regulatory information about medical devices and radiation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/medicaldevices/device-advice-comprehensive-regulatory-assistance) and CDRH Learn (https://www.fda.gov/training-and-continuing-education/cdrh-learn). Additionally, you may contact the Division of Industry and Consumer Education (DICE) to ask a question about a specific regulatory topic. See the DICE website (https://www.fda.gov/medical-device-advice-comprehensive-regulatory

{2}------------------------------------------------

assistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).

Sincerely,

Jessica Lamb

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

{3}------------------------------------------------

Indications for Use

Submission Number (if known)

K241696

Device Name

Ortho Al

Indications for Use (Describe)

Ortho AI is an image-processing software indicated to assist in making measurements for a total hip arthroplasty, total knee arthroplasty, and lumbar spine fusion surgery.

Intended patient population: Adult patients >=22 years old, with appropriate imaging, undergoing primary hip replacement, primary knee replacement, and lumbar spine surgery.

Intended user population: orthopaedic surgeons who perform hip and knee replacement, and orthopaedic/neurosurgeons who perform lumbar spine surgery.

Type of Use (Select one or both, as applicable)

| Prescription Use (Part 21 CFR 801 Subpart D)

Over-The-Counter Use (21 CFR 801 Subpart C)

CONTINUE ON A SEPARATE PAGE IF NEEDED.

This section applies only to requirements of the Paperwork Reduction Act of 1995.

DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW.

The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:

Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff(@fda.hhs.gov

"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."

{4}------------------------------------------------

510(k) Summary

Device Trade Name:	Ortho AI
Manufacturer:	Ortho AI155 East 76th Street Apt 2HNew York, NY 10021Phone: 314-662-3222
Contact:	Jonathan VigdorchikOrtho AI LLC155 East 76th StreetApt 2HNew York, NY 10021Phone: 314-662-3222jonnyvmd@gmail.com
Prepared By:	MCRA, LLC803 7th St NWWashington, DC 20001Office: 202.552.5800
Date Prepared:	November 25, 2024
Device Trade Name:	Ortho AI
Classification:	892.2050 Medical image management and processing system
Class:	Class II
Product Code:	QIH
Predicate Device:	KOALA - K192109
Classification:	892.2050 Medical image management and processing system
Class:	Class II
Product Code:	QIH

Indications for Use:

Ortho AI is an image-processing software indicated to assist in making measurements for a total hip arthroplasty, total knee arthroplasty, and lumbar spine fusion surgery.

It is intended to assist in the measurement of x-ray images by measuring lengths, angles and position of implants relative to the bone structures of interest, provided that the points of interest can be identified from radiology images.

{5}------------------------------------------------

Intended patient population: Adult patients >= 22 years old, with appropriate imaging, undergoing primary hip replacement, primary knee replacement, and lumbar spine surgery.

Intended user population: orthopaedic surgeons who perform hip and knee replacement, and orthopaedic/neurosurgeons who perform lumbar spine surgery.

Device Description:

Software/firmware input

As a non-invasive software as a medical device (SaMD) system, Ortho AI requires, as input, preoperative radiographic images of the pelvis, knee, or spine. Input mages may be one of a large number of standard image formats that are supported (.jpg, .bmp, etc.), a DICOM image. The image can also be uploaded intraoperatively from a fluoroscopic c-arm via a direct cable connection. Images must be uploaded by a healthcare professional.

Software/firmware output

The software output will be an HTML page that displays the processed image in JPEG format. The output displays the measurements of angles and length of anatomical structures that are essential for preoperative planning for the anatomical site displayed in the image. The output can only be viewed by a healthcare professional.

Software functions

Ortho AI is a software-only device intended to be used by healthcare professionals for the preoperative and intraoperative planning of total hip replacement surgery, partial/total knee replacement surgery, or lumbar spine surgery. The software semi-automates templating/planning for surgery. In the standard of care pre-operative planning procedure, a healthcare provider templates manually or digitally on a radiograph by drawing lines, making angular measurements, and sizing implants using acetate or digital templates. The Ortho AI software will semi-automate the above listed tasks by taking a radiological image as input and delivering key measurements of the surgeon's interest through an AI algorithm. These landmarks are then editable by the user.

Feature	Subject Device	Predicate Device	Comparison
510(k) Number	K241696	K192109	N/A
Name	Ortho AI	IB Lab's KOALASoftware	N/A
ClassificationName andProduct Code	System, ImageProcessing,Radiological (QIH)	System, ImageProcessing,Radiological (LLZ)	Same

Substantial Equivalence:

{6}------------------------------------------------

Runs on Server	Yes	Yes	Same
Image input	JPEG, PNG, BMP,DICOM	DICOM compliantimages collected in otherdevices in either digitallycomputed (CR) ordirectly digital (DX)formats.	Similar. The subjectdevice is validated toadditionally analyze otherstandard image formatsand does not raise newquestions about safety oreffectiveness.
Imagingprocessing	Landmark detection, usercan edit landmark	Knee detection;Landmark detection;Joint space detection	Same
AnatomicalArea	Knee, Hip, Lumbar Spine	Leg	Similar. The subjectdevice is intended to beused on the knee, hip, andlumbar spine while thepredicate device is onlyfor the leg. Thisdifference in specificanatomic locations doesnot raise new types ofquestions for safety oreffectiveness.
Measurement	Length and angle	Length and angle	Same
Way ofMeasurement	Semi-automatic (AIdriven)	Automatic (AI driven)	Similar. The subjectdevice requires the userto review and edit thedevice output asnecessary while thepredicate device does notas a fully automaticdevice.
Intended User	Trained Professional	Trained Professional	Same
Indications forUse	Ortho AI is an image-processing softwareindicated to assist inmaking measurements fora total hip arthroplasty,total knee arthroplasty,and lumbar spine fusionsurgery.It is intended to assist inthe measurement of x-rayimages by measuringlengths, angles andposition of implantsrelative to the bonestructures of interest,provided that the pointsof interest can be	IB Lab KOALA is aradiological fully-automated imageprocessing softwaredevice of either computed(CR) or directly digital(DX) images intended toaid medical professionalsin the measurement ofminimum joint spacewidth; the assessment ofthe presence or absenceof sclerosis, joint spacenarrowing, andosteophytes basedOARSI criteria for theseparameters; and thepresence or absence of	The subject deviceperforms measurementsof lengths and angles onhip, knee, and lumbarspine images. Thepredicate device performslength and anglemeasurements on legimages. This difference inspecific anatomiclocations does not raisenew questions for safetyor effectiveness andtherefore does not inducechanges in the intendeduse

{7}------------------------------------------------

identified from radiology	radiographic knee OA
images. The device allows for overlaying of digital annotations on radiological images and includes tools for performing measurements using the images and digital annotations. The software is not for primary image interpretation. The software is not for use on mobile phones. Intended patient population: Adult patients >=22 years of age, with appropriate imaging, undergoing primary hip replacement, primary knee replacement, and lumbar spine surgery. Intended user population: orthopaedic surgeons who perform hip and knee replacement, and orthopaedic/neurosurgeon s who perform lumbar spine surgery.	based on Kellgren & Lawrence Grading of standing, fixed-flexion radiographs of the knee. It should not be used in- lieu of full patient evaluation or solely relied upon to make or confirm a diagnosis. The system is to be used by trained professionals including, but not limited to, radiologists, orthopedics, physicians, and medical technicians.

identified from radiology

radiographic knee OA

images. The device allows for overlaying of digital annotations on radiological images and includes tools for performing measurements using the images and digital annotations. The software is not for primary image interpretation. The software is not for use on mobile phones. Intended patient population: Adult patients >=22 years of age, with appropriate imaging, undergoing primary hip replacement, primary knee replacement, and lumbar spine surgery. Intended user population: orthopaedic surgeons who perform hip and knee replacement, and orthopaedic/neurosurgeon s who perform lumbar spine surgery.

based on Kellgren & Lawrence Grading of standing, fixed-flexion radiographs of the knee. It should not be used in- lieu of full patient evaluation or solely relied upon to make or confirm a diagnosis. The system is to be used by trained professionals including, but not limited to, radiologists, orthopedics, physicians, and medical technicians.

Performance Testing Summary:

Software Validation and Verification was performed.

Acceptance Criteria and Sample size

The overall aim was to achieve a Dice coefficient greater than 0.85 (Dice > 0.85). A test sample size of ≥ 150 samples and observing an effect size of ≥ 0.01370, we can be confident achieving a statistical power of ≥ 80%.

Acceptance criteria for a mean Dice coefficient at a minimum of 0.85 for all algorithms and all connected domains.

Model summary:

Data independence: We ensured full data independence by adopting a patient-level partitioning process, where each patient contributed only a single x-ray image to the dataset, eliminating any risk of intersample dependencies or data leakage between the training, validation, and testing phases.

{8}------------------------------------------------

Truthing process: 3 fellowship-trained ABOS board-certified orthopaedic surgeons with each greater than 10 years of experience, followed a 2+1 truthing process where two blinded orthopaedic surgeons reviewed each segmented image in the testing set and applied modifications. After the reviews from each blinded surgeon, a final senior-level surgeon adjudicator reviewed the modifications and added further modifications to the segmentations, if necessary.

Hip model:

1,367 images from 1,367 patients -
Average age: 62 -
Female: 52% -
Ethnicity: 78% white, 18% Black/African American, 1% Other (non-white), 3% Asian -

Hip-spine model:

4,836 images from 4,836 patients -
Average age: 62 -
Female: 61% -
-Ethnicity: 78% white, 18% Black/African American, 1% Other (non-white), 3% Asian

Knee model:

-4,536 images from 4,536 patients
Average age: 61 -
-Female: 52%
Ethnicity: 78% white, 18% Black/African American, 1% Other (non-white), 3% Asian -

Standalone algorithm testing

Hip model:

LLD measurements within +/- 1.96mm of human measurement -
-Offset (global) within +/- 0.88mm of human measurement
SFP angle within +/- 1.05mm of human measurement -
Overall performance as measured by Dice coefficient shows all connected domains to be above our acceptance criteria of 0.85.
Subgroup analysis by x-ray machine also shows that all connected domains, across all x-ray machine types, to be above our acceptance criteria of 0.85.

Hip-spine model:

SS, SPT, APPt, PI, LL, PI-LL all within 2 degrees of human measurement -
No statistical difference between human vs. machine learning measurements -
Overall performance as measured by Dice coefficient shows all connected domains to be above our acceptance criteria of 0.85.
Subgroup analysis by x-ray machine also shows that all connected domains, across all x-ray machine types, to be above our acceptance criteria of 0.85.

Knee model:

LDFA , mPTA, aHKA, aJLOA all within 2 degrees of human measurement -
-No statistical difference between human vs. machine learning measurements

{9}------------------------------------------------

Overall performance as measured by Dice coefficient shows all connected domains to be above our acceptance criteria of 0.85.
Subgroup analysis by x-ray machine also shows that all connected domains, across all x-ray machine types, to be above our acceptance criteria of 0.85.

Conclusion

Ortho AI demonstrates substantial equivalence to the predicate device. The subject device has the same intended use and principles of operation; furthermore, it has similar indications and technological characteristics as its predicate device. The minor differences between subject and predicate device in indications do not alter the intended use of the subject device and do not raise new or different questions regarding its safety and effectiveness when used as labeled. Performance data demonstrate that the device performs as intended. Verification and validation testing, including the standalone software performance test, supports the safety of the device and demonstrates that Ortho AI performs as intended. Therefore, Ortho AI demonstrates substantial equivalence to the predicate device.

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).