Faculty of Computer Science and Engineering, University Ss. Cyril and Methodius, Skopje, Macedonia.
Department of Knowledge Technologies, Jožef Stefan Institute, Ljubljana, Slovenia.
Comput Med Imaging Graph. 2015 Jan;39:14-26. doi: 10.1016/j.compmedimag.2014.06.005. Epub 2014 Jun 19.
In this paper, we present the approach that we applied to the medical modality classification tasks at the ImageCLEF evaluation forum. More specifically, we used the modality classification databases from the ImageCLEF competitions in 2011, 2012 and 2013, described by four visual and one textual types of features, and combinations thereof. We used local binary patterns, color and edge directivity descriptors, fuzzy color and texture histogram and scale-invariant feature transform (and its variant opponentSIFT) as visual features and the standard bag-of-words textual representation coupled with TF-IDF weighting. The results from the extensive experimental evaluation identify the SIFT and opponentSIFT features as the best performing features for modality classification. Next, the low-level fusion of the visual features improves the predictive performance of the classifiers. This is because the different features are able to capture different aspects of an image, their combination offering a more complete representation of the visual content in an image. Moreover, adding textual features further increases the predictive performance. Finally, the results obtained with our approach are the best results reported on these databases so far.
在本文中，我们介绍了在 ImageCLEF 评估论坛上应用于医学模态分类任务的方法。更具体地说，我们使用了 2011、2012 和 2013 年的 ImageCLEF 竞赛中的模态分类数据库，这些数据库由四种视觉类型和一种文本类型的特征以及它们的组合描述。我们使用局部二值模式、颜色和边缘方向描述符、模糊颜色和纹理直方图以及尺度不变特征变换（及其变体对手 SIFT）作为视觉特征，并结合 TF-IDF 加权的标准词袋文本表示。广泛的实验评估结果确定 SIFT 和对手 SIFT 特征是模态分类的最佳表现特征。接下来，视觉特征的低级融合提高了分类器的预测性能。这是因为不同的特征能够捕捉图像的不同方面，它们的组合提供了图像中视觉内容的更完整表示。此外，添加文本特征进一步提高了预测性能。最后，我们的方法所获得的结果是迄今为止在这些数据库中报告的最佳结果。
