Thi Hong Nhan Vu

Online discovery of Heart Rate Variability patterns in mobile healthcare services

Recent years, advances in day-to-day wearable sensors have led to the development of low powered physiological sensor platforms, which can be integrated in body area networks, a new enabling technology for real-time health monitoring. The bottleneck in health state awareness is the algorithm that has to interpret the sensor data. Nowadays Coronary Heart Disease (CHD) is still the leading cause of death. Many classification techniques such as decision tree and neural networks proposed for an early detection of individual at risk for CHD are not able to continuously detect heart state based on sensor data stream. In this study, we propose an online three-layer neural network to recognize Heart Rate Variability (HRV) patterns related to CHD risk in consideration of daily activities. ECG sensor data is preprocessed using Poincare plot encoding. Incremental learning is utilized to train the network with new data without forgetting the previously learned patterns. The algorithm is named Poincare-based HRV patterns discovering Incremental Artificial neural Network (PHIAN). When a sample is presented, the nodes in the hidden layer of PHIAN compete for determining the node with the highest similarity to the input. Error variables associated with the neuron units are used as criteria for new node insertion in hopes of allowing the network to learn new patterns and reducing classification error. However, the node insertion has to be stopped in the overlapping decision areas. We suppose that the overlaps between classes have lower probability than the centric part of the classes. Therefore, after a period of learning we remove the nodes with no neighbor. Plus, the error probability density is taken into account instead of input probability density. Finally, the predictive capability of PHIAN is compared with three previous classification models, namely Self-Organizing Map (SOM), Growing Neural Gas (GNG), and Multilayer Perceptron (MLP) in terms of classification error and network structure. The results show that PHIAN outperforms the existing techniques. Our proposed model can be efficiently applied to early detection of abnormal conditions and prevent the abnormal becoming serious.

Heart rate variability pattern recognition in ambulatory environments

Coronary Heart Disease (CHD) is the leading cause of death in the world. Heart Rate Variability (HRV) has been known as a measure of cardiac electrophysiology that independently predicts sudden death in CHD patients. In this paper, we develop a Poincaré encoding based HRV patterns discovering Incremental Artificial neural Network (PHIAN). Long-term Electrocardiogram (ECG) recordings are made in various daily activities under the controlled heart rate and breathing frequency. Incremental training is exploited to achieve the ability to train the classifier model with new data without destroying the previously learned patterns. The hidden layer of the network is updated by new node insertions based on local errors. However, the insertion in the overlapping decision areas has to be stopped when it on longer improves the performance of the classification model. The algorithm effectiveness is finally assessed in terms of classification error in relation to the data set structure and network structure. The results manifest that PHIAN with a satisfactory number of nodes outperforms the previous techniques.

Discovery of Frequent Sequence Pattern in Moving Object Databases

The converge of location-aware devices, GIS functionalities and the increasing accuracy and availability of positioning technologies pave the way to a range of new types of location-based services. The field of spatiotemporal data mining where relationships are defined by spatial and temporal aspect of data is encountering big challenges since the increased search space of knowledge. Therefore, we aim to propose algorithms for mining spatiotemporal patterns in mobile environment in this paper. Moving patterns are generated utilizing two algorithms called All_MOP and Max_MOP. The first one mines all frequent patterns and the other discovers only maximal frequent patterns. Our proposed approach is able to reduce consuming time through comparison with DFS_MINE algorithm. In addition, our approach is applicable to location-based services such as tourist service, traffic service, and so on.