返回
Adaptive Osprey Optimization Algorithm with Vision Transformer Encoder For Automatic Psoriasis Disease Detection
DOI:10.1080/03772063.2025.2586175.png)
摘要
En 中文
The skin disease psoriasis affects crores of people across the globe and is identified by red, itchy patches. Initial detection of psoriasis is crucial for effective management and enhanced patient outcomes. Many attempts have been made to use picture processing and machine learning to automate the process of diagnosing skin disorders. The reason for this misdiagnosis is due to the margin of error in the methods doctors use to make the diagnosis. Furthermore, early psoriasis diagnostic models are less accurate and time-consuming. To address these shortcomings, we present an optimized vision transformer for the automatic diagnosis of psoriasis in this paper. Initially, the psoriasis disease pictures are preprocessed using the median filter. After preprocessing, the entire picture is embedded in a convolutional neural network (CNN) that embeds each scene transform encoder layer. In the next step, an optimal visual transform encoder is used to classify the input patch image. Here, the hyper parameters of the transformer are optimized employing the Adaptive Osprey Optimization Algorithm (AOSO). According to experimental results of this method, it can be concluded that the optimal vision transformer model proposed maximized classification accuracy by 98.3%.
Keyword:
Accuracy
AOSO
automatic disease detection
CNN
psoriasis
vision transformer
期刊
IF:
1.3
论文数:
261
被引数:
3.3K
机构
引用论文
Diagnostic clinical decision support based on deep learning and knowledge-based systems for psoriasis: From diagnosis to treatment options基于深度学习和知识系统的银屑病诊断临床决策支持: 从诊断到治疗方案
Optimized vision transformer encoder with cnn for automatic psoriasis disease detection具有cnn的优化视觉变压器编码器,用于自动牛皮癣疾病检测
Exploring psychosocial adaptation among people with chronic skin disease: A grounded theory study探索慢性皮肤病患者心理社会适应:一项扎根理论研究
Nursing Open
IF0

