RESUMO
Although commercial motion-capture systems have been widely used in various applications, the complex setup limits their application scenarios for ordinary consumers. To overcome the drawbacks of wearability, human posture reconstruction based on a few wearable sensors have been actively studied in recent years. In this paper, we propose a deep-learning-based sparse inertial sensor human posture reconstruction method. This method uses bidirectional recurrent neural network (Bi-RNN) to build an a priori model from a large motion dataset to build human motion, thereby the low-dimensional motion measurements are mapped to whole-body posture. To improve the motion reconstruction performance for specific application scenarios, two fundamental problems in the model construction are investigated: training data selection and sparse sensor placement. The problem of deep-learning training data selection is to select independent and identically distributed (IID) data for a certain scenario from the accumulated imbalanced motion dataset with sufficient information. We formulate the data selection into an optimization problem to obtain continuous and IID data segments, which comply with a small reference dataset collected from the target scenario. A two-step heuristic algorithm is proposed to solve the data selection problem. On the other hand, the optimal sensor placement problem is studied to exploit most information from partial observation of human movement. A method for evaluating the motion information amount of any group of wearable inertial sensors based on mutual information is proposed, and a greedy searching method is adopted to obtain the approximate optimal sensor placement of a given sensor number, so that the maximum motion information and minimum redundancy is achieved. Finally, the human posture reconstruction performance is evaluated with different training data and sensor placement selection methods, and experimental results show that the proposed method takes advantages in both posture reconstruction accuracy and model training time. In the 6 sensors configuration, the posture reconstruction errors of our model for walking, running, and playing basketball are 7.25°, 8.84°, and 14.13°, respectively.
RESUMO
Dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) has become a powerful tool for the diagnosis of breast cancer in the clinical setting due to its high sensitivity and specificity. Pharmacokinetic parameters, including Ktrans and area under the curve (AUC), and texture features derived from DCE-MRI have been used to specify the characteristics inside tumors. In the present study, 56 patients (average age 45.3±11.1; range 25-69 years) with histopathologically proved breast tumors were analyzed using the pharmacokinetic parameters and texture features. Malignant tumors displayed higher Ktrans and AUC values than the benign, Ktrans exhibited a significantly difference between the malignant and benign tumors (P=0.001) compared with the AUC values (P=0.029); texture features from DCE-MRI images and pharmacokinetic parameter maps also showed a good diagnostic ability. Alongside the routine method, principal components analysis (PCA) and Fisher discriminant analysis (FDA) were employed on these texture features to differentiate the breast lesions automatically. The Factor-1 scores of PCA were used to divide the patients into two groups, and the diagnosing accuracies of the FDA method on the texture features from DCE-MRI images, Ktrans maps, AUC maps were 93, 98 and 98%, with a cross validation accuracies of 82, 77 and 77%, respectively. To conclude, pharmacokinetic parameters, texture features and the combined computer-assisted classification method were discussed. All method involved in this study may be a potential assisted tool for radiological analysis on breast.
