Pleural effusion is the pathologic accumulation of body fluids around the unilateral or bilateral lungs that is primarily caused by heart disease. A chest radiograph is a rapid examination technique used to provide a preliminary diagnosis of lung and heart diseases. Computer-Aided diagnosis with the digitalized image is an automated approach that addresses the drawbacks of manual inspection. In this study, two corner detectors along with a two-dimensional convolution process are used to enhance the chest X-ray image for an accurate extrapolation of the bilateral lung cavities. Based on bounding box pixel analysis, the pixel ratios of the lung anatomy between normal and abnormal conditions can be estimated to identify the pleural effusion size. Next, a smart drainage monitoring system is developed to improve the current functions of the traditional drainage tool and confirm the drainage safety, including (a) drainage volume and required time detection, (b) unplanned removal warning, and (c) physiological status monitoring. The experimental result will be used to determine the feasibility of the proposed effusion volume estimation algorithm and the efficiency of the smart drainage monitoring prototyping tool. The proposed smart drainage monitoring system and the computer-Aided method with digitalized images can be further applied in real clinical practice in the intensive care unit.
All Science Journal Classification (ASJC) codes
- Computer Science(all)
- Materials Science(all)