Department of Medical Physics, Faculty of Medicine, School of Health Sciences, University of Ioannina, University Campus, 45110 Ioannina, Greece.
Department of Skin and Venereal Diseases, Faculty of Medicine, School of Health Sciences, University of Ioannina, University Campus, 45110 Ioannina, Greece.
Comput Biol Med. 2017 Sep 1;88:50-59. doi: 10.1016/j.compbiomed.2017.07.001. Epub 2017 Jul 3.
Actinic keratoses (AK) are common premalignant skin lesions that can progress to invasive skin squamous cell carcinoma (sSCC). The subtle accumulation of multiple AK in aging individuals increases the risk of sSCC development, and this underscores the need for efficient treatment and patient follow-up. Our objectives were to develop a method based on color texture analysis of standard clinical photographs for the discrimination of AK from healthy skin and subsequently to test the developed approach in the quantification of field-directed treatment interventions.
AK and healthy skin in clinical photographs of 22 patients were demarcated by experts and regions of interest (ROIs) of 50 × 50 pixels were cropped. The data set comprised 6010 and 13915 ROIs from AK and healthy skin, respectively. Color texture features were extracted using local binary patterns (LBP) or texton frequency histograms and evaluated employing a support vector machine (SVM) classifier. Classifier evaluation was performed using a leave-one-patient-out scheme in RGB, YIQ and CIE-Lab color spaces. The best configuration of the SVM model was tested using 157 AK and 216 healthy skin rectangular regions of arbitrary size. AK treatment outcome was evaluated in an additional group of eight patients with 32 skin lesions.
The best configuration of the discrimination model was achieved by employing LBP color texture descriptors estimated from the Y and I components of the YIQ color space. The sensitivity and specificity of the SVM model were 80.1% and 81.1% at ROI level and 89.8% and 91.7% at region level, respectively. Based on the classifier results the quantitative AK reduction was 83.6%.
It is important that patients with AK seek evaluation for treatment to reduce the risk of disease progression. Efficient patient follow-up and treatment evaluation require cost-effective and easy to use approaches. The proposed SVM discrimination model based on LBP color texture analysis renders clinical photography a practical, widely available and cost-effective tool for the evaluation of AK burden and treatment efficacy.
光化性角化病（AK）是一种常见的癌前皮肤病变，可进展为侵袭性皮肤鳞状细胞癌（sSCC）。在衰老个体中，多个 AK 的轻微累积会增加 sSCC 发展的风险，这凸显了对有效治疗和患者随访的需求。我们的目标是开发一种基于标准临床照片的颜色纹理分析方法，用于区分 AK 与健康皮肤，并随后在量化定向治疗干预的效果方面测试所开发的方法。
由专家对 22 名患者的临床照片中的 AK 和健康皮肤进行标记，并裁剪出 50×50 像素的感兴趣区域（ROI）。该数据集包含分别来自 AK 和健康皮肤的 6010 和 13915 个 ROI。使用局部二值模式（LBP）或纹理频率直方图提取颜色纹理特征，并使用支持向量机（SVM）分类器进行评估。在 RGB、YIQ 和 CIE-Lab 颜色空间中，采用每位患者留一法进行分类器评估。使用 157 个 AK 和 216 个健康皮肤的任意大小矩形 ROI 测试 SVM 模型的最佳配置。在另外 8 名患者的 32 个皮肤病变中评估 AK 治疗效果。
采用 YIQ 颜色空间的 Y 和 I 分量估计的 LBP 颜色纹理描述符的 SVM 模型最佳配置，在 ROI 水平的灵敏度和特异性分别为 80.1%和 81.1%，在区域水平的灵敏度和特异性分别为 89.8%和 91.7%。基于分类器结果，AK 的定量减少率为 83.6%。
患有 AK 的患者寻求治疗评估以降低疾病进展的风险非常重要。有效的患者随访和治疗评估需要成本效益高且易于使用的方法。基于 LBP 颜色纹理分析的 SVM 判别模型使临床摄影成为评估 AK 负担和治疗效果的实用、广泛可用且具有成本效益的工具。
