HiGMA-DADCN: Hirudinaria granulosa multitropic algorithm optimised double attention enabled deep convolutional neural network for psoriasis classification
Psoriasis, a chronic skin condition, holds particular importance in medical classification due to its impact on both physical and mental health. Its visible symptoms such as red, dry patches and persistent itchiness can severely affect a person’s quality of life. Although many researchers have worke...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/21681163.2025.2485100 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Psoriasis, a chronic skin condition, holds particular importance in medical classification due to its impact on both physical and mental health. Its visible symptoms such as red, dry patches and persistent itchiness can severely affect a person’s quality of life. Although many researchers have worked on psoriasis classification, existing methods often fall short. Common limitations include inaccurate classification, small or insufficient datasets, and high rates of false positives and false negatives. To overcome these challenges, this study introduces a novel approach: the Hirudinaria granulosa multitropic algorithm-based Double Attention Deep Convolutional Neural Network (HiGMA-DADCN). This model integrates two key components the HiGMA algorithm and a double attention module designed to enhance segmentation and boost overall classification performance. The HiGMA algorithm plays a crucial role in identifying and extracting the most relevant regions of affected skin through optimal segmentation. Experimental validation was conducted using the DermNet dataset. The proposed model achieved promising results: an accuracy of 95.80%, sensitivity of 95.15%, F1-score of 92.40%, MCC of 0.94, false positive rate of 0.04, and true positive rate of 0.96. Additionally, the model reduces computational complexity while significantly improving classification accuracy, outperforming existing techniques and offering a more reliable approach for psoriasis detection and diagnosis. |
|---|---|
| ISSN: | 2168-1163 2168-1171 |