Abstract:
CAD-Skin is an advanced system designed for the detection and classification of skin disor ders using deep learning convolutional neural networks and encoder-decoder units. Integrating
Internet of Things (IoT)-assisted skin cancer recognition, CAD-Skin incorporates connected
devices and sensors for primary analysis and monitoring of skin conditions. Recent advance ments in IoT-based skin cancer recognition utilizing deep learning (DL) have significantly en hanced early analysis and monitoring of skin cancer. The CAD-Skin system employs a modern
preprocessing approach, combining multi-scale retinex, gamma correction, unsharp masking,
and contrast-limited adaptive histogram equalization, resulting in highly detailed and accurate
anatomical structure depictions. To address unbalanced datasets, data augmentation techniques
are applied, improving the model’s robustness across various pigmented skin diseases and pre venting overfitting. Additionally, a quantum support vector machine (QSVM) algorithm is inte grated for final-stage classification, redefining dermatological diagnostics. CAD-Skin enhances
category recognition for different skin disease severities, including actinic keratosis, malignant
melanoma, and other skin cancers. Utilizing the PAD-UFES-20-Modified, ISIC-2018, and ISIC 2019 datasets, the system achieved accuracy rates of 98%, 99%, and 99%, respectively. For cer tain strict skin disorder diagnoses, it achieved 97.43% accuracy. This research demonstrates that
CAD-Skin provides precise diagnoses and timely detection of skin abnormalities, diversifying
options for doctors and enhancing patient satisfaction during medical practice.
x
Keywords: Skin lesion classification, Artificial intelligence in healthcare, Deep learning, Convo lutional neural networks (CNN), Encoder-decoder units, Data augmentation, Quantum support
vector machine (QSVM), Internet of Things (IoT)
