(277 days)
Yes
The device description explicitly states that it "utilizes artificial intelligence (AI) and computer vision algorithms" and that these algorithms are "trained on medical images".
No
Explanation: The device is an image processing software intended to aid in the analysis and measurement of radiographic abnormalities for diagnosis, not to provide therapy or treatment.
Yes
The device is intended to "aid medical professionals in the measurement of minimum joint space width; the assessment of the presence or absence of sclerosis, joint space narrowing, and osteophytes based on OARSI criteria for these parameters; and, the presence or absence of radiographic knee OA based on Kellgren & Lawrence Grading of standing, fixed-flexion radiographs of the knee." This explicitly states its role in assessing and grading conditions, which is a diagnostic function, even though it stipulates it should not be solely relied upon for diagnosis.
Yes
The device description explicitly states "Deep Xray is a standalone software device". It processes existing digital images and provides a report via a web browser, without including any hardware components.
Based on the provided information, this device is not an IVD (In Vitro Diagnostic).
Here's why:
- IVD Definition: In Vitro Diagnostics are tests performed on samples taken from the human body, such as blood, urine, or tissue, to detect diseases, conditions, or infections.
- Device Function: The Deep Xray device analyzes radiological images (CR and DX) of the knee. It does not analyze biological samples from the patient's body.
- Intended Use: The intended use is to aid medical professionals in the interpretation and measurement of features within medical images, not to perform a diagnostic test on a biological sample.
Therefore, Deep Xray falls under the category of a medical image processing software device, not an In Vitro Diagnostic device.
No
The letter does not explicitly state that the FDA has reviewed, approved, or cleared a PCCP for this specific device.
Intended Use / Indications for Use
Deep Xray is a radiological fully automated image processing software device of either computed (CR) or directly digital (DX) images intended to aid medical professionals in the measurement of minimum joint space width; the assessment of the presence or absence of sclerosis, joint space narrowing, and osteophytes based on OARSI criteria for these parameters; and, the presence or absence of radiographic knee OA 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, physicians and medical technicians.
Product codes
QIH
Device Description
Deep Xray is a standalone software device that utilizes artificial intelligence (AI) and computer vision algorithms to assist clinical professionals in analyzing and measuring radiographic abnormalities of knee osteoarthritis (OA) during review of posterior or anterior-posterior knee radiographs. DeepXray provides automated metric measurements of the joint space width and angular measurements of the femoral-tibial angle. DeepXray also performs assessments of knee osteoarthritis based on the Kellgren-Lawrence Grade (KL Grade), as well as individual radiographic features of osteoarthritis, including joint space narrowing, osteophyte and sclerosis based on the OARSI (Osteoarthritis Research Society International) grading criteria.
The output of DeepXray is rendered as a summary report and can be viewed on a web browser. Using this web interface, the user can verify the AI report side-by-side with the original radiograph using standard DICOM image tools and review each AI analysis result with the help of markup images overlaid with highlighted disease location or reference lines used for automated measurements. The web report also notifies the user for potential data quality issues. The clinical professionals can make modifications to the AI analysis results based on their professional judgement before saving and outputting the report.
Mentions image processing
Yes
Mentions AI, DNN, or ML
Yes
Input Imaging Modality
DICOM compliant images in either digitally computed (CR) or directly digital (DX) formats
Anatomical Site
Joint (knee)
Indicated Patient Age Range
Not Found
Intended User / Care Setting
trained professionals including, but not limited to, radiologists, physicians and medical technicians.
Description of the training set, sample size, data source, and annotation protocol
Patients: 3387 (1390 Male, 1997 Female, 1642 > 60 years old).
Ethnicity: 2675 White, 626 Black or African American, 25 Asian, 61 Unknown or not reported.
DICOM images: 18406 (11213 CR, 3438 DX, 3755 RG).
Visiting Timepoint: Baseline (3385), 12 month (3158), 24 month (2999), 36 month (2877), 48 month (2757), 72 month (1602), 96 month (1628).
X-ray Manufacturer: 3914 Agfa, 3030 Fujifilm, 3450 GE, 2055 Konica-Minolta, 66 Philips, 318 Siemens, 3873 Swissray, 1700 Others (Not reported).
Knees: 35217 (20536 KL 2, 21089 JSN Absent, 14128 JSN Present, 18997 Osteophyte Absent, 15533 Osteophyte Present, 25813 Sclerosis Absent, 8336 Sclerosis Present).
Data Source: from one of the five clinical sites of the Osteoarthritis Initiative (OAI), a multi-center, longitudinal study of men and women with or at increased risk for knee OA. Annotation protocol: Not Found.
Description of the test set, sample size, data source, and annotation protocol
Patients: 1121 (499 Male, 622 Female, 601 > 60 years old).
Ethnicity: 937 White, 159 Black or African American, 13 Asian, 12 Unknown or not reported.
DICOM images: 6125 (5783 CR, 342 DX).
Visiting Timepoint: Baseline (1121), 12 month (1058), 24 month (978), 36 month (939), 48 month (897), 72 month (558), 96 month (574).
X-ray Manufacturer: 4671 Agfa, 1144 Fujifilm, 310 GE.
Knees: 11816 (6850 KL 2, 6862 JSN Absent, 4954 JSN Present, 8018 Osteophyte Absent, 3538 Osteophyte Present, 8520 Sclerosis Absent, 2865 Sclerosis Present).
Data Source: independent data from one of the five clinical sites of the Osteoarthritis Initiative (OAI), a multi-center, longitudinal study of men and women with or at increased risk for knee OA. Annotation protocol: Not Found.
Summary of Performance Studies (study type, sample size, AUC, MRMC, standalone performance, key results)
Performance Data: DeepXray's clinical performance validation was performed on an independent data from one of the five clinical sites of the Osteoarthritis Initiative (OAI), a multi-center, longitudinal study. DeepXray's automatic quality control mechanism blocked 0.3% of test samples.
Key Results for performance of status indicators (Sensitivity/Specificity):
-
Kellgren-Lawrence Grade (KL Grade ≥2):
- Sample Number: 11775 knees/6114 DICOM/1121 subjects
- Sensitivity: 0.87 (0.86/0.88)
- Specificity: 0.84 (0.83/0.85)
-
Joint Space Narrowing (OARSI Grade ≥1):
- Sample Number: 11775 knees/6114 DICOM/1121 subjects
- Sensitivity: 0.88 (0.87/0.89)
- Specificity: 0.82 (0.81/0.83)
-
Osteophyte (OARSI Grade ≥1):
- Sample Number: 11518 knees/5993 DICOM/1121 subjects
- Sensitivity: 0.86 (0.85/0.87)
- Specificity: 0.80 (0.79/0.81)
-
Sclerosis (Presence/Absence):
- Sample Number: 11348 knees/5904 DICOM/1119 subjects
- Sensitivity: 0.84 (0.83/0.85)
- Specificity: 0.88 (0.87/0.89)
Performance of measurements for Joint Space Width (JSW) and Femoral-Tibial Angle (FTA) quantified by orthogonal linear regression against reference measurements from the OAI study:
-
Medial mJSW (mm):
- Sample Number: 7748 knees/4432 DICOM/862 subjects
- Slope: 1.02 (1.00, 1.03)
- Intercept: 0.04 (-0.03, 0.11)
-
Lateral mJSW (mm):
- Sample Number: 7605 knees/4377 DICOM/861 subjects
- Slope: 0.98 (0.95, 1.01)
- Intercept: 0.06 (-0.10, 0.26)
-
Femoral-Tibial Angle (degree°):
- Sample Number: 7546 knees/4310 DICOM/854 subjects
- Slope: 0.97 (0.96, 0.99)
- Intercept: -0.10 (-0.17, -0.04)
The performance data support that the knee OA assessments and measurements given by DeepXray are in good agreement with clinical professionals' labeling provided by the OAI study.
Key Metrics (Sensitivity, Specificity, PPV, NPV, etc.)
Sensitivity, Specificity, Slope, Intercept.
Predicate Device(s)
IB Lab GmbH's KOALA (K192109), Radiobotics ApS's RBknee (K203696)
Reference Device(s)
Not Found
Predetermined Change Control Plan (PCCP) - All Relevant Information
Not Found
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).
0
Image /page/0/Picture/0 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 seal on the left and the FDA acronym along with the full name of the agency on the right. The FDA part of the logo is in blue, with the acronym in a square and the full name written out to the right of the square.
Alpha Intelligence Manifolds, Inc. % Qingzong Tseng Director 2F, No. 170, Zhonghe Rd., Zhonghe District New Taipei City, 235068 TAIWAN
Re: K223621
September 8, 2023
Trade/Device Name: DeepXray Regulation Number: 21 CFR 892.2050 Regulation Name: Medical image management and processing system Regulatory Class: Class II Product Code: QIH Dated: August 14, 2023 Received: August 14, 2023
Dear Qingzong Tseng:
We have reviewed your Section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. Although this letter refers to your product as a device, please be aware that some cleared products may instead be combination products. The 510(k) Premarket Notification Database located at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm identifies combination product submissions. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading.
If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.
Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting of medical device-related adverse events) (21 CFR 803) for
1
devices or postmarketing safety reporting (21 CFR 4, Subpart B) for combination products (see https://www.fda.gov/combination-products/guidance-regulatory-information/postmarketing-safety-reportingcombination-products); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820) for devices or current good manufacturing practices (21 CFR 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.
Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR Part 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to https://www.fda.gov/medical-device-safety/medical-device-reportingmdr-how-report-medical-device-problems.
For comprehensive regulatory information about mediation-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
Jessica Lamb, PhD 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) K223621
Device Name DeepXray
Indications for Use (Describe)
Deep Xray is a radiological fully automated image processing software device of either computed (CR) or directly digital (DX) images intended to aid medical professionals in the measurement of minimum joint space width; the assessment of the presence or absence of sclerosis, joint space narrowing, and osteophytes based on OARSI criteria for these parameters; and, the presence or absence of radiographic knee OA 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, physicians and medical technicians.
| Type of Use (Select one or both, as applicable) |
|---|
| ------------------------------------------------- |
X Prescription Use (Part 21 CFR 801 Subpart D)
Over-The-Counter Use (21 CFR 801 Subpart C)
CONTINUE ON A SEPARATE PAGE IF NEEDED.
This section applies only to requirements of the Paperwork Reduction Act of 1995.
DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW.
The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:
Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff(@fda.hhs.gov
"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."
3
Image /page/3/Picture/0 description: The image shows the logo for Alpha Intelligence Manifolds. The logo consists of a red geometric shape on the left, followed by the letters "AIM" in a dark gray color. Below the letters is the text "Alpha Intelligence Manifolds" in a smaller font size and a lighter gray color. The geometric shape appears to be an abstract representation of a house or building.
K223621 510(k) SUMMARY
DeepXrav Alpha Intelligence Manifolds, Inc.
| Applicant: | Alpha Intelligence Manifolds, Inc.
2F, No.170, Zhonghe Road, Zhonghe District,
New Taipei City, 235068,
Taiwan
Telephone: +882-2-2240-6570 |
|----------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------|
| Date Prepared: | Aug 14, 2023 |
| Device Name: | DeepXray |
| Regulation Number: | 892.2050 |
| Regulation Name: | Medical Image Management and Processing System |
| Product Code: | QIH |
| Classification Name: | Automated Radiological Image Processing Software |
| Device Class: | Class II |
| Review Panel: | Radiology |
| Predicate Devices: | IB Lab GmbH's KOALA (K192109)
Radiobotics ApS's RBknee (K203696) |
Device Description
Deep Xray is a standalone software device that utilizes artificial intelligence (AI) and computer vision algorithms to assist clinical professionals in analyzing and measuring radiographic abnormalities of knee osteoarthritis (OA) during review of posterior or anterior-posterior knee radiographs. DeepXray provides automated metric measurements of the joint space width and angular measurements of the femoral-tibial angle. DeepXray also performs assessments of knee osteoarthritis based on the Kellgren-Lawrence Grade (KL Grade), as well as individual radiographic features of osteoarthritis, including joint space narrowing, osteophyte and sclerosis based on the OARSI (Osteoarthritis Research Society International) grading criteria.
The output of DeepXray is rendered as a summary report and can be viewed on a web browser. Using this web interface, the user can verify the AI report side-by-side with the original radiograph
4
Image /page/4/Picture/0 description: The image shows the logo for Alpha Intelligence Manifolds. The logo consists of a red geometric shape on the left, followed by the letters "AIM" in a dark gray sans-serif font. Below the letters is the full name of the company, "Alpha Intelligence Manifolds", in a smaller, lighter gray font. The geometric shape resembles a stylized house or building.
using standard DICOM image tools and review each AI analysis result with the help of markup images overlaid with highlighted disease location or reference lines used for automated measurements. The web report also notifies the user for potential data quality issues. The clinical professionals can make modifications to the AI analysis results based on their professional judgement before saving and outputting the report.
Intended Use / Indications for Use
Deep Xray is a radiological fully automated image processing software device of either computed (CR) or directly digital (DX) images intended to aid medical professionals in the measurement of minimum joint space width; the assessment of the presence of sclerosis, joint space narrowing, and osteophytes based on OARSI criteria for these parameters; and, the presence or absence of radiographic knee OA based on Kellgren & Lawrence Grading of standing, fixedflexion 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, orthopedists, physicians and medical technicians.
Comparison of Technological Characteristics
Deep Xray has the same technological characteristics as the predicate devices: KOALA (K192109) and RBknee (K203696). DeepXray and predicate devices all utilize computer vision (CV) as well as artificial intelligence (AI) algorithms trained on medical images to perform automated image processing tasks such as knee detection, landmark detection, and joint space detection. DeepXray and predicate devices all operate within Docker containers on a Linux server.
The only differences with predicate devices are that, instead of outputting the knee osteoarthritis analysis report as DICOM images, DeepXray output a web report providing quality warnings, markup images as well as an editing interface for the clinical professionals to modify any automated analysis results they considered inappropriate. In addition to the disease status (KL Grade, JSN, osteophyte, sclerosis) and minimum joint space width value (mJSW) reported by the predicate devices, DeepXray also reports the quantitative measurement of the Femoral-Tibial Angle.
In general, the technological characteristics of the DeepXray is directly comparable to the predicate devices, KOALA and RBknee. A table comparing the key features of the subject and predicate devices is provided below:
5
Image /page/5/Picture/0 description: The image shows the logo for Alpha Intelligence Manifolds (AIM). The logo consists of a red geometric shape on the left, followed by the letters "AIM" in a dark gray sans-serif font. Below the letters, the full name of the company, "Alpha Intelligence Manifolds," is written in a smaller, lighter gray font.
| Feature | DeepXray
(Subject Device) | KOALA
(K192109,
Predicate Device) | RBknee
(K203696,
Predicate Device) |
|------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Classification Name
and Product Code | Automated Radiological
Image Processing
Software (QIH) | System, Image
Processing, Radiological
(LLZ) | System, Image
Processing, Radiological
(LLZ) |
| Anatomical Area | Joint (knee) | Joint (knee) | Joint (knee) |
| Image Input | DICOM compliant images
in either digitally
computed (CR) or directly
digital (DX) formats | DICOM compliant images
in either digitally
computed (CR) or directly
digital (DX) formats | DICOM compliant images
in either digitally
computed (CR) or directly
digital (DX) formats |
| Image Processing | Knee detection;
Landmark detection; Joint
space detection | Knee detection;
Landmark detection; Joint
space detection | Knee detection;
Landmark detection; Joint
space detection |
| Human Intervention
for interpretation | Required | Required | Required |
| Intended User | Trained professionals | Trained professionals | Trained professionals |
| Output Format | Web report with quality
warning, markup images
and editing interface | A single DICOM Image | Markup images and
textual report as static
DICOM images |
| Output Information | - Knee OA status:
KL grade ≥2 or ≤1
- JSN status:
Absent/Present - Osteophyte status:
Absent/Present - Sclerosis status:
Absent/Present - Minimum Joint Space
Width - Femoral-Tibial Angle | - Knee OA status:
KL grade ≥2 or ≤1 - JSN status:
Absent/Present - Osteophyte status:
Absent/Present - Sclerosis status:
Absent/Present - Minimum Joint Space
Width | - Knee OA status:
KL grade ≥2 or 60 | 601 (53.6%) | 1642 (48.5%) |
| Ethnicity | | |
| White | 937 (83.6%) | 2675 (79.0%) |
| Black or African American | 159 (14.2%) | 626 (18.5%) |
| Asian | 13 (1.2%) | 25 (0.8%) |
| Unknown or not reported | 12 (1.1%) | 61 (1.9%) |
| #DICOM | 6125 | 18406 |
| CR | 5783 (94.4%) | 11213 (60.9%) |
| DX | 342 (5.6%) | 3438 (18.7%) |
| RG | 0 (0%) | 3755 (20.4%) |
| Visiting Timepoint: | | |
| Baseline | 1121 (18.3%) | 3385 (18.4%) |
| 12 month | 1058 (17.3%) | 3158 (17.2%) |
| 24 month | 978 (16%) | 2999 (16.3%) |
| 36 month | 939 (15.3%) | 2877 (15.6%) |
| 48 month | 897 (14.6%) | 2757 (15.0%) |
| 72 month | 558 (9.1%) | 1602 (8.7%) |
| 96 month | 574 (9.4%) | 1628 (8.8%) |
| X-ray Manufacturer: | | |
| Agfa | 4671 (76.3%) | 3914 (21.3%) |
| Fujifilm | 1144 (18.7%) | 3030 (16.5%) |
| GE | 310 (5.1%) | 3450 (18.7%) |
| Konica-Minolta | 0 (0%) | 2055 (11.2%) |
| Philips | 0 (0%) | 66 (0.4%) |
| Siemens | 0 (0%) | 318 (1.7%) |
| Swissray | 0 (0%) | 3873 (21.0%) |
| Others (Not reported) | 0 (0%) | 1700 (9.2%) |
| #Knees: | 11816 | 35217 |
| KI 2 | 4966 (42%) | 14681 (41.7%) |
| JSN Absent | 6862 (58.1%) | 21089 (59.9%) |
| JSN Present | 4954 (41.9%) | 14128 (40.1%) |
| Osteophyte Absent | 8018 (69.4%) | 18997 (55.0%) |
| Osteophyte Present | 3538 (30.6%) | 15533 (45.0%) |
| Sclerosis Absent | 8520 (74.8%) | 25813 (75.6%) |
| Sclerosis Present | 2865 (25.2%) | 8336 (24.4%) |
7
Image /page/7/Picture/0 description: The image shows the logo for Alpha Intelligence Manifolds (AIM). The logo consists of a red geometric shape resembling a stylized house or building, followed by the letters "AIM" in a dark gray sans-serif font. Below the logo is the full name of the company, "Alpha Intelligence Manifolds," also in a sans-serif font.
DeepXray's automatic quality control mechanism blocked 0.3% of test samples and did not give analysis results on those samples. Of the remaining test samples, the performance for the status indicators (expressed in Sensitivity/Specificity) of the radiographic findings are shown in the table below:
| DeepXray
Output | Sample Number | Performance Metric | Result
(95% C.I.) |
|--------------------------------|----------------------------------------------|-----------------------------------|----------------------|
| Kellgren-
Lawrence
Grade | 11775 knees/
6114 DICOM/
1121 subjects | Sensitivity
(KL Grade ≥2) | 0.87 (0.86/0.88) |
| | | Specificity
(KL Grade ≥2) | 0.84 (0.83/0.85) |
| Joint Space
Narrowing | 11775 knees/
6114 DICOM/
1121 subjects | Sensitivity
(OARSI Grade ≥1) | 0.88 (0.87/0.89) |
| | | Specificity
(OARSI Grade ≥1) | 0.82 (0.81/0.83) |
| Osteophyte | 11518 knees/
5993 DICOM/
1121 subjects | Sensitivity
(OARSI Grade ≥1) | 0.86 (0.85/0.87) |
| | | Specificity
(OARSI Grade ≥1) | 0.80 (0.79/0.81) |
| Sclerosis | 11348 knees/
5904 DICOM/
1119 subjects | Sensitivity
(Presence/Absence) | 0.84 (0.83/0.85) |
| | | Specificity
(Presence/Absence) | 0.88 (0.87/0.89) |
The performance of the measurements for the Joint Space Width (JSW) and the Femoral-Tibial Angle (FTA) by the DeepXray were quantified by orthogonal linear regression against reference measurements from the OAI study. The performance test results are summarized below:
| DeepXray
| Output | Sample Number | Orthogonal linear regression | Result (95% C.I.) |
|---|---|---|---|
| Medial | |||
| mJSW (mm) | 7748 knees/ | ||
| 4432 DICOM/ | |||
| 862 subjects | Slope | 1.02 (1.00, 1.03) | |
| Medial | |||
| mJSW (mm) | 7748 knees/ | ||
| 4432 DICOM/ | |||
| 862 subjects | Intercept | 0.04 (-0.03, 0.11) | |
| Lateral | |||
| mJSW (mm) | 7605 knees/ | ||
| 4377 DICOM/ | |||
| 861 subjects | Slope | 0.98 (0.95, 1.01) | |
| Lateral | |||
| mJSW (mm) | 7605 knees/ | ||
| 4377 DICOM/ | |||
| 861 subjects | Intercept | 0.06 (-0.10, 0.26) | |
| Femoral-Tibial Angle | |||
| (degree°) | 7546 knees/ | ||
| 4310 DICOM/ | |||
| 854 subjects | Slope | 0.97 (0.96, 0.99) | |
| Femoral-Tibial Angle | |||
| (degree°) | 7546 knees/ | ||
| 4310 DICOM/ | |||
| 854 subjects | Intercept | -0.10 (-0.17, -0.04) |
The performance data support that the knee OA assessments and measurements given by DeepXray are in good agreement with clinical professionals' labeling provided by the OAL study.
8
Image /page/8/Picture/0 description: The image contains the logo for Alpha Intelligence Manifolds. The logo consists of a red geometric shape resembling a stylized house or network node, followed by the letters "AIM" in a dark gray sans-serif font. Below the logo is the full name of the company, "Alpha Intelligence Manifolds," also in a dark gray sans-serif font, but smaller than the "AIM" letters.
Substantial Equivalence
DeepXray has the same intended use, similar indications, technological characteristics, and principles of operation as its predicate devices. The differences between DeepXray and its predicate devices 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. Performance data demonstrates that DeepXray performs as intended. Thus, DeepXray is substantially equivalent to its predicate devices.