Suk Jin Lee

Respiratory Motion Estimation With Hybrid Implementation of Extended Kalman Filter

The extended Kalman filter (EKF) can be used for the purpose of training nonlinear neural networks to perform desired input-output mappings. To improve the computational requirements of the EKF, Puskorius proposed the decoupled EKF (DEKF) as a practical remedy for the proper management of computational resources. This approach, however, sacrifices computational accuracy of estimates because it ignores the interactions between the estimates of mutually exclusive weights. To overcome such a limitation, therefore, we proposed hybrid implementation based on EKF (HEKF) for respiratory motion estimation, which uses the channel number for the mutually exclusive groups and the coupling technique to compensate the computational accuracy. Moreover, the authors restricted to a DEKF algorithm in which the weights connecting the inputs to a node are grouped together. If there are multiple input training sequences with respect to the time stamp, the complexity can increase by the power of the input channel number. To improve the computational complexity, we split the complicated neural network into a couple of simple neural networks to adjust separate input channels. The experimental results validated that the prediction overshoot of the proposed HEKF was improved by 62.95% in the average prediction overshoot values. The proposed HEKF showed a better performance of 52.40% improvement in the average of the prediction time horizon. We have evaluated that the proposed HEKF can outperform DEKF by comparing the prediction overshoot values, the performance of the tracking estimation value, and the normalized root-mean-squared error.

Tracking Human Motion With Multichannel Interacting Multiple Model

Tracking human motion with multiple body sensors has the potential to promote a large number of applications such as detecting patient motion, and monitoring for home-based applications. With multiple sensors, the tracking system architecture and data processing cannot perform the expected outcomes because of the limitations of data association. For the collaborative and intelligent applications of motion tracking (Polhemus Liberty AC magnetic tracker), we propose a human motion tracking system with multichannel interacting multiple model estimator (MC-IMME). To figure out interactive relationships among distributed sensors, we used a Gaussian mixture model (GMM) for clustering. With a collaborative grouping method based on GMM and expectation-maximization algorithm for distributed sensors, we can estimate the interactive relationship with multiple body sensors and achieve the efficient target estimation to employ a tracking relationship within a cluster. Using multiple models with filter divergence, the proposed MC-IMME can achieve the efficient estimation of the measurement and the velocity from measured datasets of human sensory data. We have newly developed MC-IMME to improve overall performance with a Markov switch probability and a proper grouping method. The experiment results shows that the prediction overshoot error can be improved on average by 19.31% by employing a tracking relationship.

Design of wireless nanosensor networks for intrabody application

Emerging nanotechnology presents great potential to change human society. Nanoscale devices are able to be included with Internet. This new communication paradigm, referred to as Internet of Nanothings (IoNT), demands very short-range connections among nanoscale devices. IoNT raises many challenges to realize it. Current network protocols and techniques may not be directly applied to communicate with nanosensors. Due to the very limited capability of nanodevices, the devices must have simple communication and simple medium sharing mechanism in order to collect the data effectively from nanosensors. Moreover, nanosensors may be deployed at organs of the human body, and they may produce large data. In this process, the data transmission from nanosensors to gateway should be controlled from the energy efficiency point of view. In this paper, we propose a wireless nanosensor network (WNSN) at the nanoscale that would be useful for intrabody disease detection. The proposed conceptual network model is based on On-Off Keying (OOK) protocol and TDMA framework. The model assumes hexagonal cell-based nanosensors deployed in cylindrical shape 3D hexagonal pole. We also present in this paper the analysis of the data transmission efficiency, for the various combinations of transmission methods, exploiting hybrid, direct, and multi-hop methods.

Human tracking with an infrared camera using a curve matching framework

The Kalman filter (KF) has been improved for a mobile robot to human tracking. The proposed algorithm combines a curve matching framework and KF to enhance prediction accuracy of target tracking. Compared to other methods using normal KF, the Curve Matched Kalman Filter (CMKF) method predicts the next movement of the human by taking into account not only his present motion characteristics, but also the previous history of target behavior patterns-the CMKF provides an algorithm that acquires the motion characteristics of a particular human and provides a computationally inexpensive framework of human-tracking system. The proposed method demonstrates an improved target tracking using a heuristic weighted mean of two methods, i.e., the curve matching framework and KF prediction. We have conducted the experimental test in an indoor environment using an infrared camera mounted on a mobile robot. Experimental results validate that the proposed CMKF increases prediction accuracy by more than 30% compared to normal KF when the characteristic patterns of target motion are repeated in the target trajectory.

Prediction and Classification of Respiratory Motion

This book describes recent radiotherapy technologies including tools for measuring target position during radiotherapy and tracking-based delivery systems. This book presents a customized prediction of respiratory motion with clustering from multiple patient interactions. The proposed method contributes to the improvement of patient treatments by considering breathing pattern for the accurate dose calculation in radiotherapy systems. Real-time tumor-tracking, where the prediction of irregularities becomes relevant, has yet to be clinically established. The statistical quantitative modeling for irregular breathing classification, in which commercial respiration traces are retrospectively categorized into several classes based on breathing pattern are discussed as well. The proposed statistical classification may provide clinical advantages to adjust the dose rate before and during the external beam radiotherapy for minimizing the safety margin. In the first chapter following the Introduction to this book, we review three prediction approaches of respiratory motion: model-based methods, model-free heuristic learning algorithms, and hybrid methods. In the following chapter, we present a phantom studyprediction of human motion with distributed body sensorsusing a Polhemus Liberty AC magnetic tracker. Next we describe respiratory motion estimation with hybrid implementation of extended Kalman filter. The given method assigns the recurrent neural network the role of the predictor and the extended Kalman filter the role of the corrector. After that, we present customized prediction of respiratory motion with clustering from multiple patient interactions. For the customized prediction, we construct the clustering based on breathing patterns of multiple patients using the feature selection metrics that are composed of a variety of breathing features. We have evaluated the new algorithm by comparing the prediction overshoot and the tracking estimation value. The experimental results of 448 patients breathing patterns validated the proposed irregular breathing classifier in the last chapter.

The Minimum Scheduling Time for Convergecast in Wireless Sensor Networks

We study the scheduling problem for data collection from sensor nodes to the sink node in wireless sensor networks, also referred to as the convergecast problem. The convergecast problem in general network topology has been proven to be NP-hard. In this paper, we propose our heuristic algorithm (finding the minimum scheduling time for convergecast (FMSTC)) for general network topology and evaluate the performance by simulation. The results of the simulation showed that the number of time slots to reach the sink node decreased with an increase in the power. We compared the performance of the proposed algorithm to the optimal time slots in a linear network topology. The proposed algorithm for convergecast in a general network topology has 2.27 times more time slots than that of a linear network topology. To the best of our knowledge, the proposed method is the first attempt to apply the optimal algorithm in a linear network topology to a general network topology.

Virtual source-based minimum interference path multicast routing in optical virtual private networks

Virtual Private Network (VPN) services over the Internet are gaining increased acceptance due to the economic benefits and flexibility. However, with difficulties of providing sufficient transmission capacity for value-added and mission-critical services, the Optical VPN (OVPN) deploying Dense Wavelength-Division Multiplexing (DWDM) technology has been seen as a favorable approach for realizing the future VPN services. In an OVPN, the Routing and Wavelength Assignment problem plays a key role for capacity utilization and therefore the Multicast Routing and Wavelength Assignment problem has been the dominant issue in a DWDM-based OVPN. In this paper, using Virtual Source (VS) nodes that have splitting and wavelength conversion capabilities, we propose a new Multicast Routing and Wavelength Assignment method for multicast sessions. The algorithm combines the VS-based tree generation approach with Multi-Wavelength Minimum Interference Path Routing (MW-MIPR) that chooses a path that does not interfere too much with potential future multicast session requests.

Review: Prediction of Respiratory Motion

Radiation therapy is a cancer treatment method that employs high-energy radiation beams to destroy cancer cells by damaging the ability of these cells to reproduce [1].

A New Data Aggregation Algorithm for Clustering Distributed Nodes in Sensor Networks

The sensor nodes in sensor networks are limited in power, computational capacities, and memory. In order to fulfill these limitations an appropriate strategy is needed. Data aggregation is one of the power saving strategies in sensor networks, combining the data that comes from many sensor nodes into a set of the meaningful information. This paper proposes a new data aggregation algorithm named DAUCH (Data Aggregation algorithm Using DAG rooted at the Cluster Head) for clustering distributed nodes in sensor networks, combining the random cluster head election technique in LEACH with DAG in TORA. The proposed algorithm outperforms LEACH due to the less transmission power. Our simulation reveals that approximately a 4% improvement is accomplished comparing to the number of nodes alive with LEACH.

Customized prediction of respiratory motion with clustering from multiple patient interaction

Information processing of radiotherapy systems has become an important research area for sophisticated radiation treatment methodology. Geometrically precise delivery of radiotherapy in the thorax and upper abdomen is compromised by respiratory motion during treatment. Accurate prediction of the respiratory motion would be beneficial for improving tumor targeting. However, a wide variety of breathing patterns can make it difficult to predict the breathing motion with explicit models. We proposed a respiratory motion predictor, that is, customized prediction with multiple patient interactions using neural network (CNN). For the preprocedure of prediction for individual patient, we construct the clustering based on breathing patterns of multiple patients using the feature selection metrics that are composed of a variety of breathing features. In the intraprocedure, the proposed CNN used neural networks (NN) for a part of the prediction and the extended Kalman filter (EKF) for a part of the correction. The prediction accuracy of the proposed method was investigated with a variety of prediction time horizons using normalized root mean squared error (NRMSE) values in comparison with the alternate recurrent neural network (RNN). We have also evaluated the prediction accuracy using the marginal value that can be used as the reference value to judge how many signals lie outside the confidence level. The experimental results showed that the proposed CNN can outperform RNN with respect to the prediction accuracy with an improvement of 50%.