An unsupervised image segmentation algorithm for coronary angiography.

Computer visual systems can rapidly obtain a large amount of data and automatically process them with ease. These characteristics constitute advantages for the application of such systems in the automatic analysis of medical images, as well as in processing technology. The precision of image segmentation, which plays a critical role in computer visual systems, directly affects the quality of processing results. Coronary angiographs feature various background colors, complex patterns, and blurry edges. The image areas containing blood vessels cannot be precisely segmented through regular methods. Therefore, this study proposed an unsupervised learning algorithm that uses regional parameter expansion (RPE). This method was derived from the flood fill algorithm, which can effectively segment image areas containing blood vessels despite a complex background or uneven light and shadow. An optimal cover tree (OCT) algorithm was proposed for the establishment of coronary arteries and the estimation of vessel diameter. Through the region growing method, spanning trees were used to record the cover length of adjacent connections, thereby establishing vessel paths, and the length can be used to track changes in vessel diameter.

Use of photo diary and focus group to explore needs for digital disease management program among community older adults with chronic disease.

As technology advanced, new e-health solutions are evolved to empower people to manage their care at home. This study explored the needs for disease management in activity tracking using photo diary through older adults' subjective perspective. It further aimed to suggest which lifestyle measures, symptoms and behaviours would be meaningful to include in such a digital diseases care management program for technology design. Both photo diary and focus group discussion were used, 11 older adults with multiple metabolism-related chronic diseases (Mean age, 72.5 ± 6.14 years) were recruited and asked to carry out the photo diary to trace their living situation and needs using a tablet camera. A focus group discussion was applied to identify the needs of chronic disease management, based on the results of living context tracing. Five themes, regular physical activity, smart management of healthy behaviors, healthy diet, regular daily routine and social connection, were identified by content analysis from photo diary and the focus group discussion. The results indicated that the photo diary program can raise awareness and promotes positive behavior changes. It is believed that the E-approach can be applied to the effectively enhance older adults' self-management by monitoring their health status and their daily routine activities.

Interactive web service system for exploration of biological pathways.

OBJECTIVE: Existing bioinformatics databases such as KEGG (Kyoto Encyclopedia of Genes and Genomes) provide a wealth of information. However, they generally lack a user-friendly and interactive interface. METHODOLOGY: The study proposes a web service system for exploring the contents of the KEGG database in an intuitive and interactive manner. In the proposed system, the requested pathways are uploaded from the KEGG database and are converted from a static format into an interactive format such that their contents can be more readily explored. The system supports two basic functions, namely an exhaustive search for all possible reaction paths between two specified genes in a biological pathway, and the identification of similar reaction sequences in different biological pathways. RESULTS: The feasibility of the proposed system is evaluated by means of an initial pilot study involving 10 students with varying degrees of experience of the KEGG website and its operations. The results indicate that the system provides a useful learning tool for investigating biological pathways. CONCLUSION: A system is proposed for converting the static pathway maps in KEGG into interactive maps such that they can be explored at will. The results of a preliminary trial confirm that the system is straightforward to use and provides a versatile and effective tool for examining and comparing biological pathways.

Novel algorithm for coexpression detection in time-varying microarray data sets.

When analyzing the results of microarray experiments, biologists generally use unsupervised categorization tools. However, such tools regard each time point as an independent dimension and utilize the Euclidean distance to compute the similarities between expressions. Furthermore, some of these methods require the number of clusters to be determined in advance, which is clearly impossible in the case of a new dataset. Therefore, this study proposes a novel scheme, designated as the Variation-based Coexpression Detection (VCD) algorithm, to analyze the trends of expressions based on their variation over time. The proposed algorithm has two advantages. First, it is unnecessary to determine the number of clusters in advance since the algorithm automatically detects those genes whose profiles are grouped together and creates patterns for these groups. Second, the algorithm features a new measurement criterion for calculating the degree of change of the expressions between adjacent time points and evaluating their trend similarities. Three real-world microarray datasets are employed to evaluate the performance of the proposed algorithm.