• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference for Phoenixes on Emerging Current Trends in Engineering and Management >Psoriasis Severity Assessment of 2-D Psoriasis Skin Images
【24h】

Psoriasis Severity Assessment of 2-D Psoriasis Skin Images

机译：牛皮癣严重程度评估2-D牛皮癣皮肤图像

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Psoriasis, meaning "itchy condition", is a chronic skin disease that is characterized by scaly, reddened patches. It is a recurring disease with varying severity ranging from slight limited flakes to entire body. Psoriasis Area and Severity Index (PASI) is the most conventional method for measuring the severity of this disease. It computes the PASI score, which ranges from 0 to 72, by combining the severity of lesions and area affected into a single computational score. But these scores are not reliable as they vary for the same psoriatic lesion among different physicians and suffer from inter- and intra-observer difference. This paper mainly focuses on assessing the severity index of 2D digital images of psoriasis by removing erythema (redness) from the selected image, thereby considering other skin cells for analysis. It makes use of "Feature Space Scaling" algorithm that relies on color contrast and image texture along with a combination of Support Vector Machine (SVM) classification filters and Markov Random Fields (MRF) to come up with a treatment solution. This algorithm is tested on different psoriasis affected skin images under various lighting conditions and is proved to be reliable.
机译：牛皮癣，意味着“痒状况”，是一种慢性皮肤病，其特征在于鳞片状，红色斑块。它是一种经常性疾病，严重程度不同于微小的整体薄片。牛皮癣区和严重程度指数（PASI）是测量该疾病严重程度的最常规方法。它通过将影响单一计算得分的病变和面积的严重程度组合来计算PASI评分，其范围为0至72。但这些分数不可靠，因为它们不同于不同的医生之间相同的银屑病病变并遭受间歇性和观察者的差异。本文主要侧重于通过从所选图像中除去红斑（发红）来评估牛皮癣的2D数字图像的严重性指数，从而考虑其他皮肤细胞进行分析。它利用“特征空间缩放”算法依赖于颜色对比度和图像纹理以及支持向量机（SVM）分类滤波器和Markov随机字段（MRF）的组合来提出处理溶液。在各种照明条件下，该算法在不同的牛皮癣上测试了影响的皮肤图像，并被证明是可靠的。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号