A computer-assisted navigation system to be used for total knee arthroplasties (TKAs) was reported to improve the accuracy of bone resection and result in precise implant placement, but the concomitant surgical invasion and time consumption are clinical problems. We developed a computed tomography (CT)-based navigation system (NNS) to be used for minimally invasive TKA. It requires only the ref...

Image-guided radiation therapy aims to improve the accuracy of treatment delivery by tracking tumor position and compensating for observed movement. Due to system latency it is sometimes necessary to predict tumor trajectory evolution in order to facilitate changes in beam delivery. Neural networks (NNs) have previously been investigated for predicting future tumor position because of their abi...

An artificial neural network is an important model for training features of voice conversion (VC) tasks. Typically, neural networks (NNs) are very effective in processing nonlinear features, such as Mel Cepstral Coefficients (MCC), which represent the spectrum features. However, a simple representation of fundamental frequency (F0) is not enough for NNs to deal with emotional voice VC. This is ...

BACKGROUND AND OBJECTIVES There is a huge need for blood transfusion in the newborn particularly due to the reduced marrow activity in the neonatal period. Despite widely use of blood products in the neonatal period, there is paucity of local data on the pattern, indications and reactions to blood transfusions in Nigerian newborns. This study evaluates the blood transfusion indications and patt...

The nonnutritive sweetener (NNS) acesulfame potassium (Ace-K) elicits a bitter off-taste that varies among adults due to polymorphisms in a bitter taste receptor gene. Whether polymorphisms affect liking for Ace-K by children, who live in different sensory worlds, is unknown. We examined hedonic response to Ace-K among children compared to adults, and whether response was related to common vari...

In this paper, performance oriented control laws are synthesized for a class of single-input-single-output (SISO) n-th order nonlinear systems in a normal form by integrating the neural networks (NNs) techniques and the adaptive robust control (ARC) design philosophy. All unknown but repeatable nonlinear functions in the system are approximated by the outputs of NNs to achieve a better model co...

BACKGROUND AND OBJECTIVES Survey items of the Japan National Nutrition Survey (J-NNS) have changed over time. Several papers on dietary surveys have been published; however, to date, there are no in-depth papers regarding physical examinations. Therefore, we investigated changes in the survey items in the physical examinations performed in the J-NNS and the National Health and Nutrition Survey ...

Why heavily parameterized neural networks (NNs) do not overfit the data is an important long standing open question. We propose a phenomenological model of NN training to explain this non-overfitting puzzle. Our linear frequency principle (LFP) accounts for key dynamical feature NNs: they learn low frequencies first, irrespective microscopic details. Theory based on our LFP shows that dominance...