An Hyperparameter Optimization Study of Brain Tumor Medical Image Segmentation Using U-net
Main Article Content
Abstract
TheClinicians take a significant amount of time in interpreting medical images in general and brain MRI
in particular to minimize the error rate and identify the exact location of lesions and their types. The emergence of deep learning and the image segmentation performance have called us to apply it for brain tumors identification and thus assisting clinicians to solve this problem. Through the literature review, we can see that the U-Net method is one of the most promising methods already applied to biomedical images in general and on MRI segmentation in particular. But few works have studied the hyperparameters of this technique in order to make a comparison and choose the best one. The existing works focus on the application of the basic U-Net model only without trying to modify these parameters. In this paper, we have made a state of the art of existing U-Net approaches to study the different parametric configurations used and then we have proposed other configurations, based on other deep learning models. A final comparison between the initial approach and the proposed approach gave as good results.
Downloads
Metrics
Article Details
Licensing
TURCOMAT publishes articles under the Creative Commons Attribution 4.0 International License (CC BY 4.0). This licensing allows for any use of the work, provided the original author(s) and source are credited, thereby facilitating the free exchange and use of research for the advancement of knowledge.
Detailed Licensing Terms
Attribution (BY): Users must give appropriate credit, provide a link to the license, and indicate if changes were made. Users may do so in any reasonable manner, but not in any way that suggests the licensor endorses them or their use.
No Additional Restrictions: Users may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.