EUVIMEDEUVIMED entdecken
Crossref

A Novel DyDimNet-Based Dimension-Aware Deep Learning Framework for Automated Ovarian Tumor Classification Using Ultrasound Images

Aarti Parekh, Madhavi Desai

International Journal of Drug Delivery Technology · 29.06.2026 · 16 (59s)

DOI: 10.25258/ijddt.16.59s.79 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Drug Delivery Technology, online erschienen am 29.06.2026. Verfasst von Aarti Parekh und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate identification of ovarian tumors is crucial in increasing treatment options, reducing mortality, and facilitating successful treatment. Due to their non-invasive nature and low cost, ultrasound imaging is extensively used in the identification of ovarian tumors. Unfortunately, many challenges in diagnostics such as speckle noise, variations in anatomy, class imbalance, and subtle differences between classes can significantly negatively impact diagnostic performance. Most current deep learning algorithms use uniform convolutions and generalized augmentations that limit the ability to utilize dimension-specific spatial features found in ovarian ultrasound images. To overcome these limitations, this study proposes the DyDimNet framework with Dimension-Aware Adaptive Augmentation (DAAA) for automated classification of ovarian tumors. DyDimNet improves discriminative feature learning through the dynamic analysis of spatial dependency across the height, width and depth dimensions. The DAAA approach applies augmentation based on physical dimensions in order to preserve physical fidelity and improve the representation of the minor class. The framework was evaluated using experimental tests on the MMOTU ovarian ultrasound dataset for both binary (96.8% accuracy) and multi-class (92.6%) classification tasks in an efficient computational manner and real-time feasibility for the respective data representation

Zitieren

APA

Parekh, A., & Desai, M. (2026). A Novel DyDimNet-Based Dimension-Aware Deep Learning Framework for Automated Ovarian Tumor Classification Using Ultrasound Images. International Journal of Drug Delivery Technology, 16(59s). https://doi.org/10.25258/ijddt.16.59s.79

Vancouver

Parekh A, Desai M. A Novel DyDimNet-Based Dimension-Aware Deep Learning Framework for Automated Ovarian Tumor Classification Using Ultrasound Images. International Journal of Drug Delivery Technology. 2026;16(59s). doi:10.25258/ijddt.16.59s.79

BibTeX
@article{Parekh2026,
  author = {Parekh, Aarti and Desai, Madhavi},
  title = {A Novel DyDimNet-Based Dimension-Aware Deep Learning Framework for Automated Ovarian Tumor Classification Using Ultrasound Images},
  journal = {International Journal of Drug Delivery Technology},
  year = {2026},
  volume = {16},
  number = {59s},
  doi = {10.25258/ijddt.16.59s.79}
}
Kostenlos starten →