Department of Medical Physics, Faculty of Medicine, School of Health Sciences, University of Ioannina, Ioannina, Greece.
Department of Skin and Venereal Diseases, Faculty of Medicine, School of Health Sciences, University of Ioannina, Ioannina, Greece.
Skin Res Technol. 2019 Jul;25(4):538-543. doi: 10.1111/srt.12684. Epub 2019 Feb 14.
Actinic keratosis (AK) is a common premalignant skin lesion that can potentially progress to squamous cell carcinoma. Appropriate long-term management of AK requires close patient monitoring in addition to therapeutic interventions. Computer-aided diagnostic systems based on clinical photography might evolve in the future into valuable adjuncts to AK patient management. The present study proposes a late fusion approach of color-texture features (shallow features) and deep features extracted from pre-trained convolutional neural networks (CNN) to boost AK detection accuracy on clinical photographs.
System uses a sliding rectangular window of 50 × 50 pixels and a classifier that assigns the window region to either the AK or the healthy skin class. 6010 and 13 915 cropped regions of interest (ROI) of 50 × 50 pixels of AK and healthy skin, respectively, from 22 patients were used for system implementation. Different support vector machine (SVM) classifiers employing shallow or deep features and their late fusion using the max rule at decision level were compared with the McNemar test and Yule's Q-statistic.
Support vector machine classifiers based on deep and shallow features exhibited overall competitive performances with complementary improvements in detection accuracy. Late fusion yielded significant improvement (6%) in both sensitivity (87%) and specificity (86%) compared to single classifier performance.
The parallel improvement of sensitivity and specificity is encouraging, demonstrating the potential use of our system in evaluating AK burden. The latter might be of value in future clinical studies for the comparison of field-directed treatment interventions.
光化性角化病（AK）是一种常见的癌前皮肤病变，有可能进展为鳞状细胞癌。除了治疗干预外，对 AK 进行适当的长期管理还需要密切监测患者。基于临床摄影的计算机辅助诊断系统在未来可能会演变成 AK 患者管理的有价值的辅助手段。本研究提出了一种基于颜色-纹理特征（浅层特征）和从预训练卷积神经网络（CNN）提取的深度特征的晚期融合方法，以提高临床照片上 AK 的检测准确性。
系统使用 50×50 像素的滑动矩形窗口和一个分类器，将窗口区域分配到 AK 或健康皮肤类别。从 22 名患者中分别使用 6010 个和 13915 个 50×50 像素的 AK 和健康皮肤的感兴趣区域（ROI）进行系统实现。比较了不同的支持向量机（SVM）分类器，它们分别使用浅层或深层特征，并在决策级使用最大规则进行晚期融合，使用 McNemar 检验和 Yule's Q 统计量进行比较。
基于深度和浅层特征的支持向量机分类器表现出总体竞争力，检测准确性有互补性提高。与单个分类器性能相比，晚期融合在敏感性（87%）和特异性（86%）方面均有显著提高（6%）。
敏感性和特异性的平行提高令人鼓舞，表明我们的系统在评估 AK 负担方面具有潜在的应用价值。后者可能在未来的临床研究中用于比较现场定向治疗干预。
