Related Experiment Video
Updated: Nov 18, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
3.1K
Wavelet Transform and Deep Convolutional Neural Network-Based Smart Healthcare System for Gastrointestinal Disease
Subhashree Mohapatra1, Janmenjoy Nayak2, Manohar Mishra3
1Department of Computer Science and Engineering, Siksha O Anusandhan (Deemed to Be University), Bhubaneswar, 751030, India.
Interdisciplinary Sciences, Computational Life Sciences
|February 10, 2021
Summary
A novel smart healthcare system accurately detects gastrointestinal (GI) abnormalities using time-frequency analysis and convolutional neural networks (CNNs). This AI-driven approach achieves high accuracy in classifying GI tract images, outperforming existing methods.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Gastroenterology
Background:
- Gastrointestinal (GI) abnormalities require accurate and timely detection for effective treatment.
- Existing diagnostic methods may have limitations in sensitivity and specificity.
- Advancements in artificial intelligence (AI) offer potential for improved diagnostic accuracy.
Purpose of the Study:
- To develop and evaluate a smart healthcare system for detecting various GI abnormalities.
- To leverage time-frequency analysis and convolutional neural networks (CNNs) for enhanced image analysis.
- To validate the system's performance using the KVASIR V2 dataset.
Main Methods:
- Image pre-processing and approximate discrete wavelet transform (ADWT) coefficient extraction.
- Training and testing of CNN models on eight classes of GI-tract images from the KVASIR V2 dataset.
- Two-level classification strategy to recognize predicted values and performance evaluation using accuracy, precision, recall, specificity, and F1 score.
Main Results:
- The proposed system achieved 97.25% accuracy at the first classification level and 93.75% at the second level.
- The system demonstrated strong performance across various performance metrics.
- Comparative analysis showed the proposed approach outperformed contemporary methods on the KVASIR V2 dataset.
Conclusions:
- The developed smart healthcare system effectively detects GI abnormalities with high accuracy.
- The integration of time-frequency analysis and CNNs shows significant promise for AI-driven gastroenterology diagnostics.
- This AI-based system offers a potential advancement in the early and accurate diagnosis of GI conditions.
