Image segmentation for silicosis diagnosis in grid environment
Ran Zheng, Hai Jin, Qin Zhang, Liang Zhang · 2009
Silicosis is one of the most serious occupational diseases in China. Radiologist information comes from chest radiographs, in which the most important parts are lung fields and ribs. It is necessary to distinguish lung fields and ribs from other organs. The anatomical knowledge-based lung segmentation can improve the accuracy of segmentation and solve the problems of opened region and lost clavicle region. The curve model-based rib segmentation can acquire more accurate rib location information, and provide more efficient orientation for latter diagnosis. These two methods are also deployed in grid environment to get higher efficiency when processing massive and distributed medical data for silicosis diagnosis. Experiments show that the two proposed algorithms can meet the latter medical requirements for chest radiograph with higher accuracy.