English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

The objective of this thesis is to develop computerized algorithms that can easily and reliably diagnose the non-tumorous skin pigmentation disorders and assess the severity of such conditions based on the facial image photographs of the patients. The topic is critical in the clinical management as the correct diagnosis can avoid improper treatments which may result in adverse effects while the severity assessment determines if the treatment method used is effective or not. For solving the diagnosis problem based on the image classification techniques, we propose a voting based probabilistic linear discriminant analysis (V-PLDA) method which addresses the large within-class variance problem in the non-tumorous skin pigmentation image data. The proposed V-PLDA method has been validated with a real-world dataset that contains totally 150 samples of five non-tumorous skin pigmentation disorders indlucing freckles, lentigines, Horis nevus, melasma and nevus of Ota. The results show that the proposed V-PLDA method obtains significantly higher classification accuracy than the original PLDA method as well as several state-of-the-art image classification methods. For the pigmentation severity assessment problem, we focus on a commonly observed non-tumorous skin pigmentation disorder, melasma, by estimating the clinical modified melasma area and severity index (mMASI) using the computerized methods. We first develop a complete system for the assessment, which includes several steps of the image processing, for example, illumination normalization, image segmentation, and darkness estimation, to obtain the estimated severity scores. An image set recorded from 29 patients at two time points is used for validating the proposed algorithm. The results show that the estimated scores using the proposed algorithm have a good correlation with the scores given by the dermatologists. With the severity assessment algorithm developed, the performance of the algorithm is improved by focusing on the image segmentation process. We propose an optimal hybrid thresholding method by determining the parameter that trades off between the local and the global information according to the image content. The optimal trade-off parameter is obtained by minimizing the distance between the input image and the threshold surface. The same melasma image dataset is used in the experiment and the results show that the proposed optimal hybrid

Computerized methods for classification and severity assessment of skin pigmentation disorders