Development of a motion capturing and load analyzing system for caregivers aiding a patient to sit up in bed.

This research was carried out to analyze the actions of caregivers when aiding a patient to sit up in bed. The new system showed that three dimensional analysis could be performed even on points on the subjects' bodies that were hidden from view. We also developed a method to estimate the load on the lumbar region of caregivers based on the kinetic analysis of the human body. Using this system we were able to evaluate the performance of both lay and professional caregivers. We found a clear difference between the performances of the two types of caregivers, and noted that the professional adopted a posture that was safe and did not stress the lumbar vertebrae, whereas the layperson tended to adopt an unsafe posture.

Automated detection of healthy and diseased aortae from images obtained by contrast-enhanced CT scan.

PURPOSE: We developed the next stage of our computer assisted diagnosis (CAD) system to aid radiologists in evaluating CT images for aortic disease by removing innocuous images and highlighting signs of aortic disease. MATERIALS AND METHODS: Segmented data of patient's contrast-enhanced CT scan was analyzed for aortic dissection and penetrating aortic ulcer (PAU). Aortic dissection was detected by checking for an abnormal shape of the aorta using edge oriented methods. PAU was recognized through abnormally high intensities with interest point operators. RESULTS: The aortic dissection detection process had a sensitivity of 0.8218 and a specificity of 0.9907. The PAU detection process scored a sensitivity of 0.7587 and a specificity of 0.9700. CONCLUSION: The aortic dissection detection process and the PAU detection process were successful in removing innocuous images, but additional methods are necessary for improving recognition of images with aortic disease.