Classification of Psoriasis Using Vision Transformer: A Deep Learning-Based Approach
Bibliographic record
Abstract
Psoriasis is a chronic autoimmune skin condition causing complex symptoms, often complicating accurate diagnosis. While convolutional neural networks (CNN) have potential in dermatological image classification, they struggle to adapt to the subtle variations in skin tones, lesion textures and lighting condition compromising the accuracy in psoriasis classification. In addition, traditional methods have difficulty modelling the various factors associated with psoriasis symptoms. This research provides an enhanced method based on the Vision Transformer (ViT) architecture, which includes data augmentation and transfer learning, to overcome the difficulties of limited data access and class imbalance in medical image classification. The dataset used comprises 1307 images categorized into two classes representing psoriatic and non-psoriatic images. The self-attention mechanism in the ViT effectively captures both local and global picture information, enabling precise identification of patterns unique to psoriatic lesions. The proposed model achieved 97.53% accuracy, a 0.98 F1-score, and a strong Pearson correlation of 0.962 (p<0.001), with high sensitivity (98.37%) and specificity (97.84%), outperforming traditional CNN-based methods. The model’s improved accuracy and reliability has the potential of early diagnosis of psoriasis and assisting dermatologists for making informed decisions. Furthermore, ViT’s ability to effectively capture lesion variation in different skin tones, textures and lighting conditions, highlighting its advantage over CNN-based approaches. These findings indicate that ViT can efficiently manage the complicated nature of dermatological image analysis for timely and improved personalized treatment plans.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".