Download PDFOpen PDF in browser

Automated Tumor Detection in Medical Imaging Using Deep Learning

EasyChair Preprint 13783

17 pagesDate: July 2, 2024

Abstract

The rapid advancements in deep learning have revolutionized the field of medical imaging, offering unprecedented opportunities for improving diagnostic accuracy and efficiency. This research focuses on the development and implementation of automated tumor detection systems using deep learning algorithms, aiming to assist radiologists in early and precise tumor identification across various imaging modalities, including MRI, CT, and X-rays.

The study begins with an extensive review of the current state-of-the-art deep learning techniques in medical image analysis, highlighting key models such as Convolutional Neural Networks (CNNs), Generative Adversarial Networks (GANs), and U-Net architectures. These models are evaluated based on their accuracy, computational efficiency, and robustness in detecting different types of tumors, including brain, lung, and breast cancers.

A comprehensive dataset comprising annotated medical images is curated from multiple sources to train and validate the proposed models. Data augmentation techniques and transfer learning are employed to enhance the model’s performance and generalization capability, addressing the challenges posed by limited annotated medical data.

The core of this research involves the design and optimization of a deep learning pipeline that integrates pre-processing, segmentation, and classification stages. Advanced image preprocessing techniques such as normalization, noise reduction, and contrast enhancement are utilized to improve image quality and model input. The segmentation stage leverages fully convolutional networks (FCNs) to accurately delineate tumor boundaries, while the classification stage employs deep CNNs to differentiate between benign and malignant tumors.

Keyphrases: Classification, Convolutional Neural Networks, Explainable AI., Medical Imaging, Segmentation, Tumor Detection, deep learning

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
@booklet{EasyChair:13783,
  author    = {Kayode Sheriffdeen},
  title     = {Automated Tumor Detection in Medical Imaging Using Deep Learning},
  howpublished = {EasyChair Preprint 13783},
  year      = {EasyChair, 2024}}
Download PDFOpen PDF in browser