Automated Detection of Brain Metastases on MRI (BrainMets AI)

Validation of an AI Software for Detection of Brain Metastases on MRI: AIDED Versus UNAIDED Performance

This reader study aims at assessing whether the radiologist aided with AI software has at least a non-inferior performance than without AI assistance in detecting brain metastases.

200 retrospective MRI images will be included in the study with 100 positive and 100 negative exams.

Assessment will be completed by 12 readers with varying level of experience. This is a Retrospective Multiple-Reader Multiple-Case (MRMC) randomised study.

Study Overview

• Status: Enrolling by invitation
• Conditions: Brain Metastases
• Intervention / Treatment: Device: Brain Metastasis AI

• Study Type: Observational
• Enrollment (Estimated): 12

Contacts and Locations

Study Locations

• Netherlands
  • Amsterdam, Netherlands
    • Netherlands Cancer Institute

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

Brain Metastasis

Description

Inclusion Criteria:

• subjects older than 18 years with a known or possible primary extracranial cancer who undergo MRI for diagnosis, treatment planning or follow-up of brain metastases.

Exclusion Criteria:

• subjects with primary intracranial tumor(s), with more than 10 brain metastases, with meningeal metastases, with radiation necrosis, or post brain surgery

Study Plan

How is the study designed?

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Detection accuracy	Demonstrate that the accuracy for detecting brain metastases by radiologists on MRI using the software (AIDED) is at least non-inferior to that of radiologists not using the software (UNAIDED).	3 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Reading time	Demonstrate that the AIDED radiologists reading time for detecting brain metastases is significantly lower than for UNAIDED radiologists.	3 months
Inter-reader agreement	Demonstrate that AIDED reading reduces detection variation between observers/readers in brain metastasis detection.	3 months

Collaborators and Investigators

• Sponsor: Robovision BV
• Collaborators: Netherlands Cancer Institute - NKI

Study record dates

Study Major Dates

• Study Start (Actual): January 24, 2025
• Primary Completion (Estimated): September 1, 2025
• Study Completion (Estimated): October 1, 2025

Study Registration Dates

• First Submitted: December 5, 2024
• First Submitted That Met QC Criteria: December 5, 2024
• First Posted (Actual): December 10, 2024

Study Record Updates

• Last Update Posted (Actual): August 3, 2025
• Last Update Submitted That Met QC Criteria: August 1, 2025
• Last Verified: August 1, 2025

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Brain Diseases; Central Nervous System Diseases; Nervous System Diseases; Pathologic Processes; Neoplasms by Site; Neoplasms; Neoplastic Processes; Nervous System Neoplasms; Central Nervous System Neoplasms; Neoplasm Metastasis; Brain Neoplasms
• Other Study ID Numbers: CI-2024-BrainMetsAI; IRBd20-090 (Other Identifier: IRB NKI-AVL)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No