Related Experiment Video
Updated: May 10, 2025
08:20
Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
Published on: October 27, 2023
1.2K
A Hybrid CNN Framework DLI-Net for Acne Detection with XAI
Shaila Sharmin1, Fahmid Al Farid2, Md Jihad3
1Department of Computer Science, American International University-Bangladesh, Dhaka 1229, Bangladesh.
Journal of Imaging
|April 25, 2025
Summary
This study presents a novel hybrid deep learning model for accurate acne detection and classification. The AI model achieved 97% accuracy, significantly improving dermatological care and patient outcomes.
Area of Science:
- Dermatology
- Artificial Intelligence
- Medical Imaging
Background:
- Acne is a common skin condition affecting well-being.
- Accurate acne detection is vital for effective dermatological care.
- Deep learning shows promise for improving acne diagnosis speed and accuracy.
Purpose of the Study:
- To introduce a novel hybrid deep learning model for acne detection and classification.
- To enhance diagnostic accuracy and efficiency in dermatology.
- To improve clinical decision-making and patient outcomes through AI.
Main Methods:
- Developed a hybrid model combining DeepLabV3 for segmentation and InceptionV3 for classification.
- Trained the model on a custom dataset.
- Utilized Grad-CAM for visualization to enhance model interpretability.
Main Results:
- Achieved exceptional performance with 97% validation and test accuracy.
- Obtained an F1 score, precision, and recall of 0.97.
- Outperformed existing baseline models in acne detection and classification.
Conclusions:
- The hybrid AI model offers a robust and accurate solution for acne detection.
- AI has transformative potential in dermatology, improving clinical decisions and patient care.
- Enhanced model interpretability through Grad-CAM provides transparent insights.
