Novel algorithms for reducing bladder volume estimation error caused by scanning positions

Bladder volume is an important and useful indicator to accurately diagnosis urinary diseases. Using the mechanical probe with single vibration source to estimate bladder volume is low-priced and portable. It also contains fast computation speed and low requirements to operators. However, the bladder volume estimated by this kind of probe will be influenced by improper scanning positions. To address this problem, we propose two localization approaches. One is based on geometry method, which has fast computation speed and high accuracy. The other one is based on neural network method, which contains high robustness and stable experiment results. After applying these two approaches to the standard phantom environment and human environment, we get a conclusion that the proposed approaches can effectively decrease the bladder volume estimation error caused by improper scanning positions.