EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Learning–Based Image Processing for Medical Diagnostics in Low Resource Environments

Moeed Yusuf, Samina Yasmeen

International Journal of Computer Application and Applied Sciences · 31.12.2022 · 1 (1) · 22-28

DOI: 10.66400/ijcaas3 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computer Application and Applied Sciences, online erschienen am 31.12.2022. Verfasst von Moeed Yusuf und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In low-resource healthcare settings, access to high-quality medical imaging and trained radiologists is often limited, delaying or preventing accurate disease diagnosis. This paper proposes a deep learning–based image processing framework tailored for medical diagnostics in such environments. The framework leverages a lightweight convolutional neural network (CNN) architecture chained with image enhancement modules to improve low resolution or noisy scans, such as those from low-field MRI or portable X-ray devices. To reduce computational load, we employ model compression techniques (pruning and quantization) and optimize the model for inference on resource-constrained hardware. We evaluate our system on datasets emulating low-resource conditions—including degraded chest X-rays for tuberculosis (TB) detection and low-quality MRI scans and demonstrate that our method significantly improves classification accuracy, sensitivity, and AUC compared to baseline models. The system achieves diagnostic performance close to high-end models while running efficiently on edge devices, thereby offering a practical tool for underserved healthcare environments.

Zitieren

APA

Yusuf, M., & Yasmeen, S. (2022). Deep Learning–Based Image Processing for Medical Diagnostics in Low Resource Environments. International Journal of Computer Application and Applied Sciences, 1(1), 22-28. https://doi.org/10.66400/ijcaas3

Vancouver

Yusuf M, Yasmeen S. Deep Learning–Based Image Processing for Medical Diagnostics in Low Resource Environments. International Journal of Computer Application and Applied Sciences. 2022;1(1):22-28. doi:10.66400/ijcaas3

BibTeX
@article{Yusuf2022,
  author = {Yusuf, Moeed and Yasmeen, Samina},
  title = {Deep Learning–Based Image Processing for Medical Diagnostics in Low Resource Environments},
  journal = {International Journal of Computer Application and Applied Sciences},
  year = {2022},
  volume = {1},
  number = {1},
  pages = {22--28},
  doi = {10.66400/ijcaas3}
}
Kostenlos starten →