Related Experiment Video
Updated: May 27, 2025
Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging
Published on: June 7, 2020
Deep Learning-Based Signal Amplification of T1-Weighted Single-Dose Images Improves Metastasis Detection in Brain MRI
Robert Haase1, Thomas Pinetz, Erich Kobler
1From the Department of Diagnostic and Interventional Neuroradiology, University Hospital Bonn, Bonn, Germany (R.H., E.K., Z.B., S.Z., A.-H.S., F.C.S., S.P., A.M.S., D.P., A.R., K.D.); Institute of Applied Mathematics, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany (T.P., A.E.); Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA (D.P.); Division of Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany (D.P., H.-P.S.); Department of Diagnostic and Interventional Radiology With Nuclear Medicine, Thoraxklinik at University Hospital Heidelberg, Heidelberg, Germany (M.F.-D., K.S., G.H., C.P.H.); Department of Neuroradiology, Heidelberg University Hospital, Heidelberg, Germany (M.F.-D.); Department of Diagnostic and Interventional Radiology, University Hospital Heidelberg, Heidelberg, Germany (V.W., C.P.H.); Institute for Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Essen, Germany (M.H., J.H., C.D.); Translational Lung Research Center Heidelberg (TLRC), Member of the German Center of Lung Research (DZL), Heidelberg, Germany (C.P.H.); Praxisnetz, Radiology and Nuclear Medicine, Bonn, Germany (M.V.); Department of Diagnostic and Interventional Radiology, University Hospital Bonn, Bonn, Germany (J.A.L.); German Center for Neurodegenerative Diseases (DZNE), Helmholtz Association of German Research Centers, Bonn, Germany (A.R., K.D.); and Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA (K.D.).
Deep learning enhances single-dose brain MRI to create artificial double-dose images, improving brain metastasis detection. This AI approach increases sensitivity, especially for smaller tumors, without compromising accuracy.
Area of Science:
- Radiology
- Artificial Intelligence
- Oncology
Background:
- Double-dose contrast-enhanced MRI improves brain tumor detection but raises safety concerns.
- Gadolinium-based contrast agents pose risks to patients and the environment.
- Developing safer, effective alternatives for contrast-enhanced MRI is crucial.
Purpose of the Study:
- To evaluate a deep learning (DL) method for creating artificial double-dose (A-DD) images from single-dose (T-SD) T1-weighted brain MRI.
- To assess the efficacy of A-DD images in detecting brain metastases compared to T-SD images.
Main Methods:
- A prospective, multicenter study involved 30 participants.
- A DL model was applied to T-SD brain MRI images to generate A-DD images.
- Four readers independently reviewed T-SD and A-DD images for metastases detection.
- Performance was compared using statistical analysis, including sensitivity and false-positive rates.
Main Results:
- All readers detected more metastases on A-DD images compared to T-SD images.
- Experienced readers showed significant sensitivity increases (up to 12.1%) with A-DD images.
- Less experienced readers achieved sensitivity levels comparable to experienced readers using A-DD images.
- Improved sensitivity was most notable for metastases measuring ≤5 mm.
- False-positive findings did not significantly increase, though descriptively higher.
Conclusions:
- Artificial double-dose (A-DD) imaging generated by deep learning enhances brain metastasis detection.
- This AI-driven approach offers improved sensitivity without significant loss of precision.
- A-DD imaging represents a promising advancement for routine single-dose brain MRI protocols.
