Medical image foundation models in assisting diagnosis of brain tumors: a pilot study
- 16.04.2024
- Imaging Informatics and Artificial Intelligence
Erschienen in:
- Verfasst von
- Mengyao Chen
- Meng Zhang
- Lijuan Yin
- Lu Ma
- Renxing Ding
- Tao Zheng
- Qiang Yue
- Su Lui
- Huaiqiang Sun
- Erschienen in
- European Radiology | Ausgabe 10/2024
Abstract
Objectives
To build self-supervised foundation models for multicontrast MRI of the whole brain and evaluate their efficacy in assisting diagnosis of brain tumors.
Methods
In this retrospective study, foundation models were developed using 57,621 enhanced head MRI scans through self-supervised learning with a pretext task of cross-contrast context restoration with two different content dropout schemes. Downstream classifiers were constructed based on the pretrained foundation models and fine-tuned for brain tumor detection, discrimination, and molecular status prediction. Metrics including accuracy, sensitivity, specificity, and area under the ROC curve (AUC) were used to evaluate the performance. Convolutional neural networks trained exclusively on downstream task data were employed for comparative analysis.
Results
The pretrained foundation models demonstrated their ability to extract effective representations from multicontrast whole-brain volumes. The best classifiers, endowed with pretrained weights, showed remarkable performance with accuracies of 94.9, 92.3, and 80.4%, and corresponding AUC values of 0.981, 0.972, and 0.852 on independent test datasets in brain tumor detection, discrimination, and molecular status prediction, respectively. The classifiers with pretrained weights outperformed the convolutional classifiers trained from scratch by approximately 10% in terms of accuracy and AUC across all tasks. The saliency regions in the correctly predicted cases are mainly clustered around the tumors. Classifiers derived from the two dropout schemes differed significantly only in the detection of brain tumors.
Conclusions
Foundation models obtained from self-supervised learning have demonstrated encouraging potential for scalability and interpretability in downstream brain tumor-related tasks and hold promise for extension to neurological diseases with diffusely distributed lesions.
Clinical relevance statement
The application of our proposed method to the prediction of key molecular status in gliomas is expected to improve treatment planning and patient outcomes. Additionally, the foundation model we developed could serve as a cornerstone for advancing AI applications in the diagnosis of brain-related diseases.
Anzeige
- Titel
- Medical image foundation models in assisting diagnosis of brain tumors: a pilot study
- Verfasst von
-
Mengyao Chen
Meng Zhang
Lijuan Yin
Lu Ma
Renxing Ding
Tao Zheng
Qiang Yue
Su Lui
Huaiqiang Sun
- Publikationsdatum
- 16.04.2024
- Verlag
- Springer Berlin Heidelberg
- Erschienen in
-
European Radiology / Ausgabe 10/2024
Print ISSN: 0938-7994
Elektronische ISSN: 1432-1084 - DOI
- https://doi.org/10.1007/s00330-024-10728-1
Dieser Inhalt ist nur sichtbar, wenn du eingeloggt bist und die entsprechende Berechtigung hast.
Dieser Inhalt ist nur sichtbar, wenn du eingeloggt bist und die entsprechende Berechtigung hast.
