Production-model training dataset size
4061 CT images
- EndpointImages used for final production models
- PopulationCT images from 31 clinical sites
- Tested versionContour ProtégéAI configuration described in K213976
FDA-cleared profile
MIM Software Inc. · Radiotherapy Planning Imaging · FDA-cleared (510(k))
Authorized purpose: Contour ProtégéAI creates machine-learning contours from CT and MR DICOM images to support quantitative analysis, adaptive therapy, radiotherapy planning, archiving, and follow-up; professionals must visualize, review, and edit contours, and the software is not intended to detect or contour lesions.
Overview status: Source-reviewed overview
Workflow role has not yet been reviewed for this navigation pilot.
A tag identifies a documented role; it does not establish compatibility or local acceptance. Methods and sources
Clinical takeaway. Clinical role: Contour ProtégéAI creates machine-learning contours from CT and MR DICOM images to support quantitative analysis, adaptive therapy, radiotherapy planning, archiving, and follow-up; professionals must visualize, review, and edit contours, and the software is not intended to detect or contour lesions.
Evidence takeaway. Evidence from the reviewed source record: The exact FDA document reports testing of the new 3.0.0 CT neural network models on 739 independent images from 12 institutions not represented in the training data, against ground-truth segmentations by a trained user reviewed and approved by a supervising physician. Per-structure Dice was compared with the MIM Atlas reference device, with equivalence demonstrated at p=0.05 marked for every listed structure; Contour ProtégéAI mean Dice ranged from 0.28 (Bowel_Large) to 0.96 (Brain). Results are as reported in the FDA submission, not independent validation.
Evidence context not reviewed for this FDA record.
Label constraints have not been reviewed for this exact FDA record. Unknown is not all ages or evidence of unsuitability.
Public-label constraints, not patient-level eligibility or a medical recommendation. Training and evaluation populations are separate. Read the original labeling and confirm the exact release and protocol.
Packet wording: "For the proposed Contour ProtégéAI device, the new 3.0.0 CT neural network models were trained on a pool of training data that did not include any patients from the same institution as the test subjects." … "The models were then evaluated on the test subjects from a pool of 739 independent images gathered from 12 institutions.". Read from the exact FDA decision document on 2026-11-14.
Evidence
Metrics are shown only when they are tied to a source, endpoint, population, and tested product version. Candidate literature without exact product and version linkage is not shown as product evidence.
Packet wording: "For the proposed Contour ProtégéAI device, the new 3.0.0 CT neural network models were trained on a pool of training data that did not include any patients from the same institution as the test subjects." … "The models were then evaluated on the test subjects from a pool of 739 independent images gathered from 12 institutions.". Read from the exact FDA decision document on 2026-11-14.
Sensitivity is not promoted as a product-wide endpoint for this device or software function. Compare only an exact endpoint, threshold, population, version, and source if later verified.
Specificity is not promoted as a product-wide endpoint for this device or software function. Compare only an exact endpoint, threshold, population, version, and source if later verified.
4061 CT images
31 sites
K213976
Manufacturer evidence submitted for the exact FDA decision. It is not independent clinical validation, and local technical, clinical, privacy, security, and workflow acceptance remains necessary.
Safety and lifecycle
Recall and adverse-event records are shown only after product matching. Adverse-event reports do not establish incidence or causality.
Exact-identifier review does not prove absence of all product-family safety signals.
Price not public here; enter your quote
Not publicly available: Reviewed sources do not publish an exact-product procurement price with a stated offer scope.
No separate named-product payment identified in reviewed public material
Centers for Medicare & Medicaid Services · United StatesThis is not a product-specific reimbursement determination. Build a local total-cost and value model and verify payer, site-of-service, coding, and coverage assumptions independently.
Match the AI release, system build, configuration and region. Source documentation does not establish acceptance of an installed system.
No specific system/version pairing has completed review for this FDA record. A broad DICOM or product-family statement does not establish a named viewer build's support.
Public source context, not tested interoperability. Each hop needs its own release/configuration evidence. Missing or unavailable routes are not documented incompatibility. All local tests are planned, not passed; no installed-system acceptance is recorded.
Acquisition/source → routing → AI processing → returned objects → viewer/worklist/report → support owner
No specific route context has been source-reviewed for this record. Obtain both endpoint builds and documents; this is not a finding of incompatibility.
Source context, not route validation: Supported CT or MR images, structures, prescriptions, plans, registrations, and treatment-session data.
Scope: Exact FDA submission.
Confirm scanner/model, protocol, series completeness and both software builds.
Field checked 2026-09-04 · FDA 510(k) decision material K213976: Contour ProtégéAI
Source context, not route validation: The exact FDA package describes Contour ProtégéAI as an accessory to MIM software that processes CT and MR images as supported by ACR/NEMA DICOM 3.0, creating contours for uses including transfer to radiation therapy treatment planning systems and archiving. Interface versions and local compatibility require confirmation.
Scope: Exact FDA submission. Packet wording: "Trained medical professionals use Contour ProtégéAI as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0." … "Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.". Read from the exact FDA decision document on 2026-11-14; this is the release-level description, not current commercial configuration.
Name gateways, destinations, transfer rules and failure visibility; a DICOM statement does not prove a route.
Field checked 2026-11-14 · FDA 510(k) decision material K213976: Contour ProtégéAI
Source context, not route validation: The exact FDA package describes MIMcloud remote-server and local workstation or server options for the Contour ProtégéAI release; verify the configured mode, data flow, security, and service boundary locally.
Scope: Exact FDA submission. This is the release-level boundary described in the exact FDA package; current hosting, contracts, security, availability, and site configuration remain separate evidence.
Confirm processing location, queue ownership, exclusions and exact AI build.
Field checked 2026-09-04 · FDA 510(k) decision material K213976: Contour ProtégéAI
Source context, not route validation: Contours, treatment plans, adapted plans, dose or evaluation results, or session information requiring review and approval.
Scope: Exact FDA submission.
Confirm object classes, units, references and whether findings, overlays or measurements reach their intended destination.
Field checked 2026-09-04 · FDA 510(k) decision material K213976: Contour ProtégéAI
Unresolved route: Contours, treatment plans, adapted plans, dose or evaluation results, or session information requiring review and approval.
Scope: Exact FDA submission.
Confirm the named viewer/worklist/report build, display behavior and report transfer. Output support alone does not establish this endpoint. The displayed source describes output, not validated end-to-end delivery.
Field checked 2026-09-04 · FDA 510(k) decision material K213976: Contour ProtégéAI
Unresolved route: Not established for the exact route.
Scope: Scope not established.
Name the local, platform and AI supplier support owners and escalation path.
No tests below have been performed or accepted.
Agree the evidence and acceptance owner with your team and suppliers. Checkmarks apply to this page only and are not acceptance receipts.
This is the release-level boundary described in the exact FDA package; current hosting, contracts, security, availability, and site configuration remain separate evidence.
Packet wording: "Trained medical professionals use Contour ProtégéAI as a tool to assist in the automated processing of digital medical images of modalities CT and MR, as supported by ACR/NEMA DICOM 3.0." … "Creation of contours using machine-learning algorithms for applications including, but not limited to, quantitative analysis, aiding adaptive therapy, transferring contours to radiation therapy treatment planning systems, and archiving contours for patient follow-up and management.". Read from the exact FDA decision document on 2026-11-14; this is the release-level description, not current commercial configuration.
Vendor statement, not an attestation; exact-release scope is not established.
Monitoring guidance is buyer-facing implementation guidance derived from the exact cleared scope; it is not a claim that the FDA package provides a live monitoring service.
Map installed model, build, options, population, anatomy, inputs, outputs, and enabled functions to K213976 before acceptance.
exact submission · checked 2026-09-04Later product-family features are not evidence that this cleared release includes them.
The exact FDA package describes MIMcloud remote-server and local workstation or server options for the Contour ProtégéAI release; verify the configured mode, data flow, security, and service boundary locally.
exact submission · checked 2026-09-04FDA-described deployment boundaries do not establish current hosting contracts, security controls, uptime, or local qualification.
The exact public integration contract remains incomplete; establish and test supported input, output, identity, routing, archive, and downtime behavior locally.
exact submission · checked 2026-09-04Test representative identifiers, orientation, geometry, measurements, units, routing, failure handling, and round trips locally.
Use a signed local test set for the supported population and workflow. Include common and difficult cases, exclusions, acquisition variation, failed or low-quality inputs, reviewer overrides, and downstream effects. Generated contours require visualization, review, and editing; the exact labeled structures and CT or MR inputs determine applicability, and the software is not intended to detect or contour lesions.
exact submission · checked 2026-09-04Define stop-use, escalation, rollback, and revalidation triggers before production use.
Model capital and recurring cost against net staff time after review, throughput, repeat imaging or rework, room time, consumables, infrastructure, training, service, downtime, upgrade, and replacement costs.
product family · checked 2026-09-04No exact-product price was found in the reviewed public material; productivity or value claims require local evidence.
Monitor K213976 by exact hardware, software, model, and algorithm version; patient population; anatomy; protocol; operator; input quality; output edits or overrides; failed studies; repeat acquisition or rework; downstream impact; downtime; incidents; recalls; and updates. Track these product-specific limits: Generated contours require visualization, review, and editing; the exact labeled structures and CT or MR inputs determine applicability, and the software is not intended to detect or contour lesions.
exact submission · checked 2026-09-04Revalidate after material software, hardware, model, protocol, interface, or clinical-policy changes. radiotherapy planning or image-processing software limits remain submission-specific.
Regulatory identity
The FDA listing establishes the regulatory identity. It does not by itself establish local workflow fit, pricing, security, or performance in your environment.
Research record
This audit distinguishes completed source review from fields that have not yet been researched.
The refreshed discovery snapshot contains no PubMed or ClinicalTrials.gov candidates; no external evidence was promoted. The current discovery snapshot contains 0 PubMed and 0 ClinicalTrials.gov candidate records; its dated, count-matched human screening remains current. Unlinked candidates remain quarantined. Native FDA recall identifiers produced 0 postmarket candidate records; 0 have been reviewed (0 published, 0 rejected) and 0 remain unreviewed.
Candidate leads were screened against the exact product and tested version. Unlinked records remain quarantined.
Source classes: fda ai list, fda decision summary, vendor documentation, fda decision material, peer reviewed literature, clinical trial registry, reimbursement policy, fda device recall, literature index, trial registry, pubmed, clinical trials, openfda device recall
Recorded state, verification date, notes and sources for fields without a publishable value. Conflicting records stay explicitly unsettled here; no plain settled value is shown for them.
No public list price for the exact cleared configuration was found. Obtain a written quote separating hardware or license basis, implementation, interfaces, infrastructure, training, service, upgrades, renewal, consumables, and exit costs.
7 sources - 3 history items
Scope: identity, purpose, clinicalUse, modality, anatomy, intendedUsers, careSetting, workflowRole, input, output, limitations, evidenceSummary
Scope: regulatoryIdentity, modalityContext
Scope: evidenceDiscovery
Scope: evidenceDiscovery, studyStatus
Scope: recallDiscovery, nativeSubmissionMatching
Scope: reimbursement, radiationTherapyPaymentContext
Scope: security
Regulatory record
510(k)
Research record
human reviewed
Editorial review
Source-reviewed editorial overlay
Alternatives
MIM Software Inc.
Authorized purpose: Contour ProtégéAI+ automatically creates contours on CT and T2 MR DICOM images as an accessory to MIM software for quantitative analysis, adaptive therapy, transfer, and archive; users must review and edit contours, and the device is not intended to detect or contour lesions.
MIM Software Inc.
Authorized purpose: Contour ProtegeAI assists trained medical professionals with automated processing of CT and MR images, creates machine-learning contours for quantitative analysis, adaptive therapy, transfer to radiation-treatment planning systems, and archiving, segments structures across CT locations, and segments the prostate, seminal vesicles, and urethra on T2-weighted MR images.
MIM Software Inc.
Authorized purpose: Contour ProtegeAI assists trained medical professionals with automated processing of CT and MR images, machine-learning contour creation for quantitative analysis, adaptive therapy, transfer to radiation-treatment planning systems, and archiving, segmentation across CT locations, and segmentation of the prostate, seminal vesicles, and urethra on T2-weighted MR images.
MVision AI
Authorized purpose: MVision AI Segmentation automatically generates initial contours from CT images for radiotherapy planning; the contours are transferred for professional visualization, review, modification, and approval, and the labeled software is not intended for lesion or tumor detection or real-time adaptive planning.
Copy this public context and add the field, proposed wording and original source before sending. Nothing here is submitted automatically. Form-specific prefill has not been verified.
Open existing correction destinationPermission to display a submission is not evidence verification. A submission enters editorial review; it does not approve a public change. Form availability, data entry, receipt, review and publication are separate steps. Data entry and receipt remain unverified.
No entry has been added to this scoped change history for this record. This is not proof that the record never changed.
Correction policy and limits · Public repository change history