Related Experiment Video
Updated: Oct 13, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
3.0K
Deep neural models for automated multi-task diagnostic scan management-quality enhancement, view classification and
1Healthcare Analytics and Language Engineering (HALE) Lab, Department of Information Technology, National Institute of Technology Karnataka, Surathkal, Mangalore 575025, India.
Biomedical Physics & Engineering Express
|November 18, 2021
Summary
This study introduces advanced deep learning models to enhance medical image quality and automate report generation. These innovations improve diagnostic accuracy and efficiency in healthcare settings.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computer Vision
Background:
- Medical imaging quality is often compromised by human error and equipment issues.
- Inconsistent documentation of scan orientations and unstructured data hinder effective medical image management.
- Lack of standardized data processing limits applications like patient retrieval and disease prediction.
Purpose of the Study:
- To develop multi-task neural models for enhancing medical image quality and automating diagnostic report generation.
- To improve the accuracy and efficiency of medical image analysis for better patient care.
- To establish a robust medical image management system supporting intelligent diagnosis.
Main Methods:
- Utilized ESRGAN (Enhanced Super-Resolution Generative Adversarial Network) for improving chest X-ray image quality and visualization.
- Developed ViewNet, a Convolutional Neural Network (CNN) model, for predicting X-ray view orientation.
- Employed Xception net for automated medical report generation from diagnostic scans.
Main Results:
- Proposed models demonstrated excellent performance, validated by BRISQUE, PIQE, and BLEU scores.
- Image enhancement significantly improved diagnostic potential, automatic detection, and region-of-interest segmentation.
- Deep learning approaches reduced diagnosis time and showed potential for intelligent diagnosis applications.
Conclusions:
- The developed multi-task neural models effectively address challenges in medical image quality and data management.
- The hybrid architecture offers a promising solution for enhancing diagnostic accuracy and efficiency.
- These advancements pave the way for improved healthcare outcomes through intelligent medical image analysis.
