Related Experiment Video
Updated: Oct 14, 2025
10:26
Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules
Published on: May 19, 2023
2.1K
Detection of COVID-19 from Chest CT Images Using CNN with MLP Hybrid Model.
Sakthi Jaya Sundar Rajasekar1, Vasumathi Narayanan2, Varalakshmi Perumal2
1Melmaruvathur Adhiparasakthi Institute of Medical Sciences and Research, Melmaruvathur, Tamil Nadu, India.
Studies in Health Technology and Informatics
|November 4, 2021
Summary
This study introduces a hybrid deep learning model for faster COVID-19 detection using CT scans. The model achieved 94.89% accuracy, outperforming traditional methods for early diagnosis.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computational Biology
Background:
- Early and accurate COVID-19 diagnosis is crucial to prevent hazardous spread and mortality.
- Current methods like RT-PCR have limitations including time, resources, and potential inaccuracies.
- Computed Tomography (CT) scans offer an alternative but often require significant manual effort and time.
Purpose of the Study:
- To develop an automated intelligent system for rapid COVID-19 detection from CT scan images.
- To improve diagnostic efficiency and reduce the burden on healthcare professionals.
- To evaluate the performance of a novel hybrid learning model against existing methods.
Main Methods:
- A hybrid learning model combining Convolutional Neural Network (CNN) for feature extraction and Multilayer Perceptron (MLP) for classification was developed.
- The model was trained and tested using CT scan images for COVID-19 identification.
- Performance metrics including Accuracy, F1-Score, Precision, and Recall were used for evaluation.
Main Results:
- The proposed Hybrid CNN-MLP model achieved a high accuracy of 94.89%.
- This performance significantly surpassed traditional CNN (86.95% accuracy) and MLP (80.77% accuracy) models.
- The hybrid approach demonstrated superior effectiveness in classifying COVID-19 infections from CT scans.
Conclusions:
- The Hybrid CNN-MLP model offers a promising automated solution for efficient COVID-19 detection using CT images.
- This approach can aid in early diagnosis, potentially saving lives and optimizing healthcare resources.
- Automated systems like this are vital for managing infectious disease outbreaks effectively.
