Related Experiment Video
Updated: Oct 4, 2025
08:20
Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
Published on: October 27, 2023
1.8K
A Novel Deep Learning-Based Black Fungus Disease Identification Using Modified Hybrid Learning Methodology
S Karthikeyan1, G Ramkumar2, S Aravindkumar3
1Department of Electronics and Communication Engineering, Sathyabama Institute of Science and Technology, Chennai, Tamil Nadu, India.
Contrast Media & Molecular Imaging
|February 4, 2022
Summary
A novel artificial intelligence (AI) strategy, the hybrid learning-based neural network classifier (HLNNC), effectively detects black fungus (mucormycosis) in patients with COVID-19 using eye images. This AI approach aids in early diagnosis of this rare but severe fungal infection.
Area of Science:
- Medical Artificial Intelligence
- Infectious Diseases
- Ophthalmology
Background:
- COVID-19 pandemic has led to increased incidence of opportunistic infections, including mucormycosis (black fungus).
- Mucormycosis, a rare but devastating fungal infection, disproportionately affects immunocompromised individuals, particularly those with COVID-19.
- Early and accurate diagnosis of black fungus is crucial for effective treatment and improved patient outcomes.
Purpose of the Study:
- To develop and evaluate a novel artificial intelligence (AI) strategy for detecting black fungus infection.
- To design a hybrid learning-based neural network classifier (HLNNC) integrating Convolutional Neural Network (CNN) and Support Vector Machine (SVM) principles.
- To utilize a dataset of eye photographs from patients with and without black fungus for AI model training and testing.
Main Methods:
- Implementation of a hybrid learning-based neural network classifier (HLNNC) for AI-driven diagnosis.
- Application of image processing techniques including image acquisition, preprocessing, feature extraction, and classification.
- Training and testing the HLNNC model on a curated dataset of eye images from COVID-19 patients with and without black fungus.
Main Results:
- The proposed HLNNC scheme demonstrated efficacy in identifying black fungus infection from eye photographs.
- Performance analysis confirmed the accuracy and reliability of the AI-based diagnostic approach.
- Graphical representation of results provided clear specifications of the model's performance.
Conclusions:
- The developed HLNNC model offers a promising AI-powered tool for the early detection of black fungus in COVID-19 patients.
- This AI strategy can aid clinicians in diagnosing mucormycosis, potentially reducing mortality and morbidity.
- Further research and validation are warranted to integrate this AI tool into clinical practice for managing post-COVID complications.
