Microsoft Radiomics App v1.0 is a software-only medical device intended for use by trained radiation oncologists, dosimetrists and physicists to derive optimal organ and tumor contours for input to radiation treatment planning. Supported image modalities are Computed Tomography and Magnetic Resonance. Radiomics App assists in the following scenarios:

· Load, save and display of medical images and contours for treatment evaluation and treatment planning.

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

· Localization and definition of both solid tumors and healthy anatomical structures.

· Fusion display of compatible images for treatment planning.
· Three-dimensional rendering of medical images and the segmented contours.

Images reviewed using the Radiomics App software should not be used for primary image interpretations. Radiomics App is not for use with digital mammography.

Device Description

Microsoft Radiomics App v1.0 is a software-only medical device intended for use by trained radiation oncologists, dosimetrists and medical physicists for radiation treatment planning.

Radiomics App stems from more than eight years of research in computerized medical image analysis, computer vision and machine learning. It applies well tested, state-of-the-art algorithms for the assisted delineation of anatomical structures of interest in three-dimensional, clinical radiological scans.

Radiomics App works on computed tomography (CT) and magnetic resonance (MR) scans, and is designed to contour/delineate both healthy anatomical structures as well as lesions such as solid tumors.

Radiomics App integrates into the clinical data network of radiation therapy treatment centers, receiving data from imaging devices such as CT and MR scanners. The purpose of the tool is to assist the expert user in producing segmentations (three-dimensional contours) of anatomical structures, for both solid tumors and healthy tissue structures. The following segmentation tools are provided:

Assisted Contouring. This module allows for the manual, user-guided segmentation of . structures of interest in both CT and MR images.
. Machine-learning based contouring. This module uses machine learning algorithms (ML) to provide an initial segmentation of certain structures of interest automatically. The user has the option to accept this initial segmentation or edit and refine it.
. Contour refinement. This module allows the user to edit and improve segmentations created by either the machine learning or the assisted contouring algorithms.

These segmentations are then exported back into the clinical data network, and subsequently utilized in a radiotherapy treatment planning system to generate a treatment plan for a patient.

AI/ML Overview

The provided document is a 510(k) Summary for the Microsoft Radiomics App v1.0. It describes the device, its intended use, and compares it to a predicate device (MIM 5.2). However, it does not contain detailed acceptance criteria or a specific study proving the device meets those criteria with performance metrics, sample sizes, expert qualifications, or ground truth details.

The document primarily focuses on software verification and validation, asserting that the software performs in accordance with specifications and that its performance is comparable to the predicate device, but without providing quantitative results from validation of specific features.

Therefore, for many of your requested items, the information is explicitly stated as "Not applicable" or is not provided in detail.

Here's a breakdown of the available information:

1. Table of Acceptance Criteria and Reported Device Performance

This information is not provided in this document. The document states that "Validation testing of the following functions of the Radiomics App demonstrated that the software meets user needs and intended uses and to support substantial equivalence," and lists functions like measurements, volumetric rendering, and contouring. However, it does not specify quantitative acceptance criteria (e.g., "accuracy > X%", "Dice score > Y%") nor does it report specific performance metrics against such criteria.

2. Sample size used for the test set and the data provenance

This information is not provided in this document. The document mentions "Validation Testing" but does not detail the number of cases or images used for these tests, nor the origin (country, retrospective/prospective) of any data used.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

This information is not provided in this document. While the device is intended for use by "trained radiation oncologists, dosimetrists and physicists," there is no mention of experts being used to establish ground truth for testing purposes.

4. Adjudication method for the test set

This information is not provided in this document.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

No, a multi-reader multi-case (MRMC) comparative effectiveness study was not done. The document explicitly states under "Clinical Study": "Not applicable. Clinical studies are not necessary to establish the substantial equivalence of this device." This device is an AI-assisted contouring tool, but the submission does not include a study on its comparative effectiveness with human readers.

6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done

The document mentions "Automatic Contouring - Validation Test" as one of the validation tests performed. This implies some level of standalone algorithm performance was assessed. However, no specific performance metrics or acceptance criteria for this standalone performance are provided. The device description also states: "Machine-learning based contouring. This module uses machine learning algorithms (ML) to provide an initial segmentation of certain structures of interest automatically. The user has the option to accept this initial segmentation or edit and refine it." This confirms a standalone component, but performance details are absent.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

This information is not explicitly detailed for the validation testing. Given the context of contouring for radiation treatment planning, it's highly probable that expert-generated contours would serve as ground truth, but the document does not confirm this or specify the method of ground truth establishment.

8. The sample size for the training set

This information is not provided in this document. The document mentions "Machine-learning based contouring" and that "Radiomics App stems from more than eight years of research in computerized medical image analysis, computer vision and machine learning," implying a training phase, but the sample size used for training is not disclosed.

9. How the ground truth for the training set was established

This information is not provided in this document.

Summary

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

December 27, 2017

Image /page/0/Picture/1 description: The image contains the logo of the U.S. Food and Drug Administration (FDA). The logo consists of two parts: the Department of Health & Human Services logo on the left and the FDA logo on the right. The FDA logo features the letters "FDA" in a blue square, followed by the words "U.S. FOOD & DRUG ADMINISTRATION" in blue text.

Microsoft Corp. % Donna-Bea Tillman, Ph.D. Senior Consultant Biologics Consulting Group, Inc. 1555 King Street. Suite 300 ALEXANDRIA VA 22314

Re: K173420

Trade/Device Name: Radiomics App v1.0 Regulation Number: 21 CFR 892.2050 Regulation Name: Picture archiving and communications system Regulatory Class: II Product Code: LLZ Dated: October 31, 2017 Received: November 1, 2017

Dear Dr. Tillman:

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. 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); good manufacturing practice requirements as set forth in the quality systems (OS) regulation (21 CFR Part 820); and if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.

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

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 http://www.fda.gov/MedicalDevices/Safety/ReportaProblem/default.htm for the CDRH's Office of Surveillance and Biometrics/Division of Postmarket Surveillance.

For comprehensive regulatory information about mediation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/) and CDRH Learn (http://www.fda.gov/Training/CDRHLearn). 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 (http://www.fda.gov/DICE) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).

Sincerely,

Michael D. O'Hara For

Robert Ochs, Ph.D. Director Division of Radiological Health Office of In Vitro Diagnostics and Radiological Health Center for Devices and Radiological Health

Enclosure

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

Indications for Use

510(k) Number (if known) K173420

Device Name Radiomics App v1.0

Indications for Use (Describe)

· Load, save and display of medical images and contours for treatment evaluation and treatment planning.

· Localization and definition of both solid tumors and healthy anatomical structures.

· Fusion display of compatible images for treatment planning.
· Three-dimensional rendering of medical images and the segmented contours.

Images reviewed using the Radiomics App software should not be used for primary image interpretations. Radiomics App is not for use with digital mammography.

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}------------------------------------------------

510(K) SUMMARY

1. SUBMITTER

Submitter:	Microsoft Corp1 Microsoft WayRedmond, WA 98052Phone: 425-538-9419
Contact Person:	Ivan TarapovIvan.Tarapov@microsoft.com
Submission Correspondent:	Donna-Bea Tillman, Ph.D, MPABiologics Consulting Group, Inc.1555 King Street, Suite 300Alexandria, VA 22314Phone: 410-531-6542dtillman@biologicsconsulting.com
Date Prepared:	October 31, 2017

2. DEVICE

Name of Device:	Radiomics App v1.0
Common or Usual Name:	Radiological Image Processing System
Classification Name:	Picture Archiving and Communications System21 CFR 892.2050
Regulatory Class:	Class II
Product Code:	LLZ

3. PREDICATE DEVICE

Predicate Device Name:	MIM 5.2 (BRACHY)
Manufacturer:	MIM Software Inc.
510(k) Number:	K103576
Reference Devices:	No reference devices were used in this submission.

DEVICE DESCRIPTION 4.

Microsoft Radiomics App v1.0 is a software-only medical device intended for use by trained radiation oncologists, dosimetrists and medical physicists for radiation treatment planning.

Radiomics App works on computed tomography (CT) and magnetic resonance (MR) scans, and is designed to contour/delineate both healthy anatomical structures as well as lesions such as solid tumors.

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

Assisted Contouring. This module allows for the manual, user-guided segmentation of . structures of interest in both CT and MR images.
. Machine-learning based contouring. This module uses machine learning algorithms (ML) to provide an initial segmentation of certain structures of interest automatically. The user has the option to accept this initial segmentation or edit and refine it.
. Contour refinement. This module allows the user to edit and improve segmentations created by either the machine learning or the assisted contouring algorithms.

These segmentations are then exported back into the clinical data network, and subsequently utilized in a radiotherapy treatment planning system to generate a treatment plan for a patient.

ട്. INDICATION FOR USE

Load, save and display of medical images and contours for treatment evaluation and treatment planning.
. Creation, transformation, and modification of contours for applications including, but not limited to: transferring contours to radiotherapy treatment planning systems, aiding adaptive therapy, and archiving contours for patient follow-up.
Localization and definition of both solid tumors and healthy anatomical structures.
. Fusion display of compatible images for treatment planning.
Three-dimensional rendering of medical images and the segmented contours.

Images reviewed using the Radiomics App software should not be used for primary image interpretations.

Radiomics App is not for use with digital mammography.

Comparison of Indications for Use

Both devices are only intended for use by trained medical professionals. As noted below, both devices are intended for use during radiation therapy planning. The subject device supports a subset of the image modalities supported by the predicate device. Both devices accept DICOM images generated elsewhere as inputs. Both devices provide the user with tools to contour medical images for use in radiation therapy planning and patient follow-up. Both devices can be used to localize and define normal tissues as well as solid tumors. Both devices support the fusion of different image modalities. Both devices provide three-dimensional image rendering. The subject device is not intended for primary image interpretation or for use with digital mammography. This is a subset of the indications of the predicate device.

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

The minor differences in indications of use between the subject Radiomics App and the predicate MIM software do not constitute a different intended use.

COMPARISON OF TECHNOLOGICAL CHARACTERISTICS 6. WITH THE PREDICATE DEVICE

	Proposed Device	Predicate Device
510(k) Number	TBD	K103576
Applicant	Microsoft	MIM Software
Device Name	Radiomics App	MIM 5.2
ClassificationRegulation	892.2050 - Picture Archiving andCommunication System	892.2050 – Picture Archiving andCommunication System
Product Code	LLZ	LLZ
Indications for Use	Microsoft Radiomics AdvancedImage Contouring v1.0 (RadiomicsApp) is a software-only medicaldevice intended for use by trainedradiation oncologists, dosimetristsand physicists to derive optimal organand tumor contours for input toradiation treatment planning.Supported image modalities areComputed Tomography and MagneticResonance. Radiomics App assists inthe following scenarios:Load, save and display of medicalimages and contours for treatmentevaluation and treatment planning.Creation, transformation, andmodification of contours forapplications including, but not limitedto: transferring contours toradiotherapy treatment planningsystems, aiding adaptive therapy, andarchiving contours for patient follow-up.Localization and definition of bothsolid tumors and healthy anatomicalstructures.Fusion display of compatible imagesfor treatment planning.Three-dimensional rendering ofmedical images and the segmentedcontours.Images reviewed using the Radiomics	MIM 5.2 software is used by trainedmedical professionals as a tool to aid inevaluation and information managementof digital medical images. The medicalimage modalities include, but are notlimited to, CT, MRI, CR, DX, MG, US,SPECT, PET and XA as supported byACRINEMA DICOM 3.0. MIM 5.2assists in the following indicationsReceive, transmit, store, retrieve, display,print, and process medical images andDICOM objects.Create, display and print reports frommedical images.Registration, fusion display, and reviewof medical images for diagnosis,treatment evaluation, and treatmentplanning.Evaluation of cardiac left ventricularfunction and perfusion, including leftventricular end-diastolic volume, end-systolic volume, and ejection fraction.Localization and definition of objectssuch as tumors and normal tissues inmedical images.Creation, transformation, andmodification of contours for applicationsincluding, but not limited to, quantitativeanalysis, aiding adaptive therapy,transferring contours to radiation therapytreatment planning systems, andarchiving contours for patient follow-up.
	Proposed Device	Predicate Device
	App software should not be used forprimary image interpretations.Radiomics App is not for use withdigital mammography.	and management.Quantitative and statistical analysis ofPET/SPECT brain scans by comparing toother registered PET/SPECT brain scans.
		Planning and evaluation of permanentimplant brachytherapy procedures.Lossy compressed mammographicimages and digitized film screen imagesmust not be reviewed for primary imageinterpretations. Images that are printed tofilm must be printed using a FDA-approved printer for the diagnosis ofdigital mammography images.Mammographic images must be viewedon a display system that has been clearedby the FDA for the diagnosis of digitalmammography images. The software isnot to be used for mammography CAD.
Intended users	Healthcare providers	Healthcare providers
Contouring modes	Assisted and automatic	Assisted and automatic
Types of tissuecontoured	Tumors and normal tissues	Tumors and normal tissues
Measurements	2D distance measurement, averagetissue density within a region (forCT), segmentation volume	2D distance measurement, segmentationvolume, average tissue density within aregion (for CT)
Image Fusion	Fuse only two 3D images, CT andMR	Fuse medical images from multiplemodalities
3D imagerendering	Yes	Yes
Cardiacapplications	No	Yes
Dose visualizationand manipulation	No	Yes
Image modalities	CT and MR	CT, MRI, CR, DX, MG, US, SPECT,PET and XA
Platform	Stand-alone package which operateson Microsoft Windows operatingsystems only.	Stand-alone package which operates onboth Windows and Mac computersystems.

The table below provides a side-by-same comparison of the technological characteristics of the proposed Radiomics App and the predicate MIM Software.

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

7. PERFORMANCE DATA

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

Biocompatibility

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

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

Electrical Safety and electromagnetic compatibility (EMC)

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

Software Verification and Validation Testing

Software verification and validation testing were conducted and documentation was provided as recommended by FDA's Guidance for Industry and FDA Staff, "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices." The software for this device was considered as a moderate level of concern.

Verification Testing

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

Validation Testing

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

Measurements - Validation Test
Volumetric Rendering - Validation Test
Multi-Planar Reconstruction Validation Test ●
Assisted Contouring Validation Test ●
Automatic Contouring - Validation Test

Animal Study

Not applicable. Animal studies are not necessary to establish the substantial equivalence of this device.

Clinical Study

Not applicable. Clinical studies are not necessary to establish the substantial equivalence of this device.

CONCLUSIONS 8.

The minor differences in indications of use between the subject Radiomics App and the predicate MIM software do not constitute a different intended use. The technological characteristics of the Radiomics App are similar to those of the MIM Software, with the biggest differences being that the Radiomics App offers a subset of the features provided by the MIM software. Results of software verification and validation testing demonstrate that the Radiomics App performs in accordance with specifications and that the performance is comparable to that of the predicate device. Therefore, the Radiomics App can be found to be substantially equivalent to the predicate MIM Software.

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).