EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Optimized Brain Tumor Detection: A Dual- Module Approach for MRI Image Enhancement and Tumor Classification

B. Kalpana, G.V.S. Ananthnath

International Journal of Scientific Research in Science, Engineering and Technology · 24.05.2025 · 12 (3) · 304-311

DOI: 10.32628/ijsrset251241 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Scientific Research in Science, Engineering and Technology, online erschienen am 24.05.2025. Verfasst von B. Kalpana und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The images and their segmentations obtained from the MR images are critical in making an early diagnosis and treatment of brain tumors. In this project, an attempt is made to develop a highly advanced automatic framework employing deep learning neural architectures for classification and segmentation of tumors based on MobileNet and UNET as the two main underlying architectures enhancing accuracy and computational efficiency in detection. In other words, the good feature of MobileNet is that it is light in design and offers a real-time implementation possibility when the performance itself is based on a model that is less complex. On the other hand, classifies anatomically-specific features across its densely connected layers for maximum accuracy and robustness. The system will be responsible for classifying MRI brain images into tumors and non-tumors. The classification networks are designed on MobileNet and U-net architectures to maximize accuracy and minimize computing power. MobileNet offers an optimized lightweight architecture suitable for edge and mobile implementations, with fast inference, while it improves detection accuracy with good gradient flow. The framework can also be extended to include segmentation methods to localize the tumor sites in the human brain. The integration of such a model would push the diagnostic capacity of an automated, reliable, and accurate tumor detection system to support clinical decision- making, thus having the potential to enhance the diagnosis while reducing invasion techniques. From here, one could foresee the advancement to real-time diagnostic systems in clinical hospitals. The proposed methods will be validated on benchmark datasets, with the performance metrics for validation in real medical imaging scenarios taking accuracy, precision, recall, and segmentation quality as their goals.

Zitieren

APA

Kalpana, B., & Ananthnath, G. (2025). Optimized Brain Tumor Detection: A Dual- Module Approach for MRI Image Enhancement and Tumor Classification. International Journal of Scientific Research in Science, Engineering and Technology, 12(3), 304-311. https://doi.org/10.32628/ijsrset251241

Vancouver

Kalpana B, Ananthnath G. Optimized Brain Tumor Detection: A Dual- Module Approach for MRI Image Enhancement and Tumor Classification. International Journal of Scientific Research in Science, Engineering and Technology. 2025;12(3):304-311. doi:10.32628/ijsrset251241

BibTeX
@article{Kalpana2025,
  author = {Kalpana, B. and Ananthnath, G.V.S.},
  title = {Optimized Brain Tumor Detection: A Dual- Module Approach for MRI Image Enhancement and Tumor Classification},
  journal = {International Journal of Scientific Research in Science, Engineering and Technology},
  year = {2025},
  volume = {12},
  number = {3},
  pages = {304--311},
  doi = {10.32628/ijsrset251241}
}
Kostenlos starten →