Multimedia analysis for medical applications

The advances in computing techniques, data acquisition technology, hardware, and networks have mutually promoted the development of multimedia analysis approaches. Many machine learning, signal/image processing, and data mining algorithms have been successfully developed for multimedia analysis. Recently, as an emerging topic in the multimedia domain, medical data analysis attracts much attention because of the increase of imaging modalities. This special issue serves as a forum for researchers all over the world to discuss their research work and recent advances in multimedia learning methods for various medical applications, including medical image representation, segmentation, classification, retrieval, and others. The special issue seeks for the original contributions that address the challenges of medical data. Submissions came from an open call for paper and with the assistance of professional referees. Nine papers are finally selected out from in total 25 submissions after two rounds of rigorous peer review. These accepted papers cover several popular topics of medical image analysis and applications, including 3D image reconstruction, image segmentation, image representation, image registration, etc. We summarize these accepted papers as follows. In the paper entitled “Patch-Wise Label Propagation for MR Brain Segmentation Based on Multi-Atlas Images”, Wang et al. propose a patch-wise label propagation method based on multiple atlases for MR brain segmentation using a sparse coding scheme, where the weight of each sample is driven by the sparse coding procedure. The segmentation performance is evaluated based on the ADNI dataset for hippocampus segmentation. The experimental results show that the proposed method can achieve better segmentation performance, compared with conventional multi-atlas-based methods. The paper entitled “Automatic Stenosis Detection Using SVM from CTA Projection Images” presents an automated support vector machine (SVM)-based approach that detects the branches and stenosis in 2D projection images obtained from different rotation angles of computed tomography angiography (CTA) image of the heart. Different SVM models have been built for branch and stenosis detection using geometric and shape-based features obtained from the sliding window regions. The proposed system was evaluated in terms of Precision and Recall using CTA images obtained from Billroth Hospitals, Chennai, India, and the experimental results are encouraging. Recently, optical coherence tomography (OCT) has been widely employed for the evaluation of age-related macular degeneration (AMD). In the paper entitled “Automated Segmentation of Choroidal Neovascularization in Optical Coherence Tomography Images Using Multi-scale Convolutional Neural Networks with Structure Prior”, Xi et al. propose multi-scale convolutional neural networks with structure prior (MS-CNN-SP) to segment choroidal neovascularization (CNV) from OCT data. The proposed method was evaluated on 15 spectral domain OCT data with CNV. The experimental results also demonstrate the effectiveness of the proposed method. To automatically obtain the measurement of the proliferating behavior of cells in vitro, Su et al. propose a new feature representation for mitotic event detection in timelapse phase-contrast microscopy image sequences of stem cell populations. In paper entitled “Pooled Time Series Representation for Mitosis Event Recognition”, an imaging model-based microscopy image segmentation method is implemented for mitotic candidate extraction. Then, a new feature representation framework based on time series pooling is proposed for sequential events. Finally, a Support Vector Machine classifier is utilized for mitotic cell modeling * Jun Zhang xdzhangjun@gmail.com