Personalized HRTF Prediction Based on Light GBM Using Anthropometric Data

This paper proposes a personalized headrelated transfer function(HRTF)prediction method based on Light GBM using anthropometric data.Considering the overfitting problems of the current training-based prediction methods,we use Light GBM and a specific network structure to prevent over-fitting and enhance the prediction performance.By decomposing and combining the data to be predicted,we set up 90 Light GBM models to separately predict the 90instants of HRTF in log domain.At the same time,the method of 10-fold cross-validation is used to score the accuracy of the model.For models with scores below 80 points,Bayesian optimization is used to adjust model hyperparameters to obtain a better model structure.The results obtained by Light GBM are evaluated with spectral distortion(SD)which can show the fitting error between the prediction and the original data.The mean SD values of both ears on the whole test set are 2.32 d B and 2.28 d B respectively.Compared with the non-linear regression method and the latest method,SD value of Light GBM-based method relatively decreases by 83.8%and 48.5%.

Ergonomic Evaluation on Taxi Drivers' Compartment

Driving involves long hours of physical work within c onfined compartment. Taxi drivers usually work with prolonged working hours, add itional stress may likely be induced on particular body limbs. Occupational heal th may occur and working efficiency may potentially be affected resulting fr om fatigues, pains or diseases. These problems, however, could be remedied if mo re attention is paid on seating design, the workplace and driving postures adopt ed. Ergonomics design can provide better understanding of those concerned areas. A study was conducted to analyse occupational and health problems related to ta xi drivers. Through ergonomic evaluation of driver’s compartment, the authors m ake recommendations to improve health and safety aspects. Besides, the experienc e of gained from this study using ergonomic design can be extended to other occu pation or products.

Using computed tomography scans to develop an ex-vivo gastric model

The objective of this research was to use abdominal computed tomography (CT) scans to non-invasively quantify anthropometrical data of the human stomach and to concomitantly create an anatomically correct and distensible ex-vivo gastric model. Thirty-three abdominal CT scans of human subjects were obtained and were imported into reconstruction software to generate 3D models of the stomachs. Anthropometrical data such as gastric wall thickness, gastric surface area and gastric volume were subsequently quantified. A representative 3D computer model was exported into a selective laser sintering (SLS) rapid prototyping machine to create an anatomically correct solid gastric model. Subsequently, a replica wax template of the SLS model was created. A negative mould was offset around the wax template such that the offset distance was equivalent to that of the gastric wall thickness. A silicone with similar mechanical properties to the human stomach was poured into the offset. The lost wax manufacturing technique was employed to create a hollow distensible stomach model. 3D computer gastric models were generated from the CT scans. A hollow distensible silicone ex-vivo gastric model with similar compliance to that of the human stomach was created. The anthropometrical data indicated that there is no significant relationship between BMI and gastric surface area or gastric volume. There were inter- and intra-group differences between groups with respect to gastric wall thickness. This study demonstrates that abdominal CT scans can be used to both non-invasively determine gastric anthropometrical data as well as create realistic ex-vivo stomach models.