Processing of Biomedical Signals
Much recent research has focused on biomedical signals that are obtained from the human body, such as brain waves or fMRI. Our lab is also working in this direction. For example, we examine the biomedical response of subjects exposed to a multimedia signal (e.g. image or song) as a stimulus. In this way, we are investigating solutions to problems in the conventional multimedia signal processing field by introducing new approaches based on biomedical signal-based technology.
Medical Image Analysis
Recently, medical experts discovered that H. pylori infection is strongly related to stomach cancer. More specifically, subjects that have not contracted H. pylori are in a very low risk group for developing stomach cancer. Also, medical experts discovered that it is possible to diagnose H. pylori infection from X-ray images of the stomach. Therefore, we are researching highly effective methods of automatically determining H. pylori infection from X-ray images of the stomach. This will support doctors when examining large volumes of X-ray images of the stomach.
Biomedical Image Processing
Biomedical engineering focuses on discovering the construction and functions of biological organisms, and then applying that to solving engineering problems. Since this requires the study of extremely small constructions, the volume of images obtained by scanning electron microscopes (SEM) is increasing rapidly. In order to efficiently manage a large collection of such images, it is necessary to automatically classify the imaged organisms. In our research, we focus on features of images obtained by SEM and the relationships between organisms, and investigate methods for automatically classifying organisms.
Functional Correlation Library
It is possible to access a wide range of data by providing properties of the desired data as inputs (queries) to a search engine. However, the user performing the search may not always be able to construct a clear query. Therefore, search engines that can function without a clear query are required, and researchers are focusing on new search systems that support the user’s ideas.