Fall Risk Estimation with Relaxation of the Strong Constraint on Depth Camera Location

Fall accidents are among the most serious problems at medical centers as numerous accidents occur at the bedside of a patient. To resolve this problem, the authors have proposed a method using a point cloud dataset (PCD) with images obtained from a depth camera to calculate the fall risks of hospitalized patients. However, in the conventional method, there is a strong constraint that the depth camera must be placed at a particular position above the bed; thus, it is occasionally difficult to install the camera using the conventional system. To remove this restriction, this study adds a method for correcting an original PCD to the conventional method and proposes it as a new system. Through verification, it was demonstrated that this method could convert the PCD measured at the bedside of a patient into a bed-top-view dataset. Moreover, even after the correction of the PCD, the precision of fall risk estimation remains unchanged; therefore, a system that relaxes the constraint of depth camera location and is compatible with various hospital rooms has been realized.