Related Experiment Video
Updated: May 10, 2025
09:37
Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition
Published on: August 18, 2022
2.2K
CAD-Skin: A Hybrid Convolutional Neural Network-Autoencoder Framework for Precise Detection and Classification of
Abdullah Khan1, Muhammad Zaheer Sajid1, Nauman Ali Khan1
1Department of Computer Software Engineering, Military College of Signals, National University of Science and Technology, Islamabad 44000, Pakistan.
Bioengineering (Basel, Switzerland)
|April 26, 2025
Summary
This study introduces CAD-Skin, a novel deep learning system for accurate skin cancer detection. It achieves high accuracy in classifying skin lesions, aiding dermatologists and improving patient outcomes.
Area of Science:
- Dermatology
- Artificial Intelligence
- Medical Imaging
Background:
- Accurate skin lesion diagnosis is challenging due to morphological similarities between benign and malignant growths.
- Deep learning algorithms show promise, matching dermatologist diagnostic efficacy in image-based skin lesion analysis.
Purpose of the Study:
- To propose a novel deep learning algorithm, CAD-Skin, for enhanced skin cancer detection and classification.
- To improve the classification efficiency of skin lesions using deep convolutional neural networks and autoencoders.
Main Methods:
- Developed CAD-Skin system integrating deep convolutional neural networks and autoencoders.
- Employed advanced preprocessing: multi-scale retinex, gamma correction, unsharp masking, and adaptive histogram equalization.
- Utilized data augmentation for unbalanced datasets and integrated Quantum Support Vector Machine (QSVM) for classification.
Main Results:
- Achieved high accuracy rates of 98% on PAD-UFES-20-Modified, 99% on ISIC-2018, and 99% on ISIC-2019 datasets.
- Demonstrated minimum accuracy of 97.43% for specific skin disorders.
- Outperformed existing state-of-the-art methods in skin lesion classification accuracy.
Conclusions:
- The CAD-Skin system provides precise diagnosis and timely detection of skin abnormalities.
- The proposed system enhances category recognition for various skin disease severities, including melanoma.
- CAD-Skin offers improved diagnostic options for clinicians, potentially enhancing patient satisfaction.
