Improving Biomedical Signal Search Results in Big Data Case-Based Reasoning Environments.

Time series subsequence matching has importance in a variety of areas in healthcare informatics. These include case-based diagnosis and treatment as well as discovery of trends among patients. However, few medical systems employ subsequence matching due to high computational and memory complexities. This manuscript proposes a randomized Monte Carlo sampling method to broaden search criteria with minimal increases in computational and memory complexities over R-NN indexing. Information gain improves while producing result sets that approximate the theoretical result space, query results increase by several orders of magnitude, and recall is improved with no signi cant degradation to precision over R-NN matching.

Forelimb EMG-based trigger to control an electronic spinal bridge to enable hindlimb stepping after a complete spinal cord lesion in rats.

BACKGROUND: A complete spinal cord transection results in loss of all supraspinal motor control below the level of the injury. The neural circuitry in the lumbosacral spinal cord, however, can generate locomotor patterns in the hindlimbs of rats and cats with the aid of motor training, epidural stimulation and/or administration of monoaminergic agonists. We hypothesized that there are patterns of EMG signals from the forelimbs during quadrupedal locomotion that uniquely represent a signal for the "intent" to step with the hindlimbs. These observations led us to determine whether this type of "indirect" volitional control of stepping can be achieved after a complete spinal cord injury. The objective of this study was to develop an electronic bridge across the lesion of the spinal cord to facilitate hindlimb stepping after a complete mid-thoracic spinal cord injury in adult rats. METHODS: We developed an electronic spinal bridge that can detect specific patterns of EMG activity from the forelimb muscles to initiate electrical-enabling motor control (eEmc) of the lumbosacral spinal cord to enable quadrupedal stepping after a complete spinal cord transection in rats. A moving window detection algorithm was implemented in a small microprocessor to detect biceps brachii EMG activity bilaterally that then was used to initiate and terminate epidural stimulation in the lumbosacral spinal cord. We found dominant frequencies of 180-220 Hz in the EMG of the forelimb muscles during active periods, whereas these frequencies were between 0-10 Hz when the muscles were inactive. RESULTS AND CONCLUSIONS: Once the algorithm was validated to represent kinematically appropriate quadrupedal stepping, we observed that the algorithm could reliably detect, initiate, and facilitate stepping under different pharmacological conditions and at various treadmill speeds.

High performance biomedical time series indexes using salient segmentation.

The advent of remote and wearable medical sensing has created a dire need for efficient medical time series databases. Wearable medical sensing devices provide continuous patient monitoring by various types of sensors and have the potential to create massive amounts of data. Therefore, time series databases must utilize highly optimized indexes in order to efficiently search and analyze stored data. This paper presents a highly efficient technique for indexing medical time series signals using Locality Sensitive Hashing (LSH). Unlike previous work, only salient (or interesting) segments are inserted into the index. This technique reduces search times by up to 95% while yielding near identical search results.

A Monte Carlo Approach to Biomedical Time Series Search.

Time series subsequence matching (or signal searching) has importance in a variety of areas in health care informatics. These areas include case-based diagnosis and treatment as well as the discovery of trends and correlations between data. Much of the traditional research in signal searching has focused on high dimensional R-NN matching. However, the results of R-NN are often small and yield minimal information gain; especially with higher dimensional data. This paper proposes a randomized Monte Carlo sampling method to broaden search criteria such that the query results are an accurate sampling of the complete result set. The proposed method is shown both theoretically and empirically to improve information gain. The number of query results are increased by several orders of magnitude over approximate exact matching schemes and fall within a Gaussian distribution. The proposed method also shows excellent performance as the majority of overhead added by sampling can be mitigated through parallelization. Experiments are run on both simulated and real-world biomedical datasets.

Aggregated Indexing of Biomedical Time Series Data.

Remote and wearable medical sensing has the potential to create very large and high dimensional datasets. Medical time series databases must be able to efficiently store, index, and mine these datasets to enable medical professionals to effectively analyze data collected from their patients. Conventional high dimensional indexing methods are a two stage process. First, a superset of the true matches is efficiently extracted from the database. Second, supersets are pruned by comparing each of their objects to the query object and rejecting any objects falling outside a predetermined radius. This pruning stage heavily dominates the computational complexity of most conventional search algorithms. Therefore, indexing algorithms can be significantly improved by reducing the amount of pruning. This paper presents an online algorithm to aggregate biomedical times series data to significantly reduce the search space (index size) without compromising the quality of search results. This algorithm is built on the observation that biomedical time series signals are composed of cyclical and often similar patterns. This algorithm takes in a stream of segments and groups them to highly concentrated collections. Locality Sensitive Hashing (LSH) is used to reduce the overall complexity of the algorithm, allowing it to run online. The output of this aggregation is used to populate an index. The proposed algorithm yields logarithmic growth of the index (with respect to the total number of objects) while keeping sensitivity and specificity simultaneously above 98%. Both memory and runtime complexities of time series search are improved when using aggregated indexes. In addition, data mining tasks, such as clustering, exhibit runtimes that are orders of magnitudes faster when run on aggregated indexes.

Dynamic self-adaptive remote health monitoring system for diabetics.

Diabetes is the seventh leading cause of death in the United States. In 2010, about 1.9 million new cases of diabetes were diagnosed in people aged 20 years or older. Remote health monitoring systems can help diabetics and their healthcare professionals monitor health-related measurements by providing real-time feedback. However, data-driven methods to dynamically prioritize and generate tasks are not well investigated in the remote health monitoring. This paper presents a task optimization technique used in WANDA (Weight and Activity with Blood Pressure and Other Vital Signs); a wireless health project that leverages sensor technology and wireless communication to monitor the health status of patients with diabetes. WANDA applies data analytics in real-time to improving the quality of care. The developed algorithm minimizes the number of daily tasks required by diabetic patients using association rules that satisfies a minimum support threshold. Each of these tasks maximizes information gain, thereby improving the overall level of care. Experimental results show that the developed algorithm can reduce the number of tasks up to 28.6% with minimum support 0.95, minimum confidence 0.97 and high efficiency.

Dynamic Task Optimization in Remote Diabetes Monitoring Systems.

Diabetes is the seventh leading cause of death in the United States, but careful symptom monitoring can prevent adverse events. A real-time patient monitoring and feedback system is one of the solutions to help patients with diabetes and their healthcare professionals monitor health-related measurements and provide dynamic feedback. However, data-driven methods to dynamically prioritize and generate tasks are not well investigated in the domain of remote health monitoring. This paper presents a wireless health project (WANDA) that leverages sensor technology and wireless communication to monitor the health status of patients with diabetes. The WANDA dynamic task management function applies data analytics in real-time to discretize continuous features, applying data clustering and association rule mining techniques to manage a sliding window size dynamically and to prioritize required user tasks. The developed algorithm minimizes the number of daily action items required by patients with diabetes using association rules that satisfy a minimum support, confidence and conditional probability thresholds. Each of these tasks maximizes information gain, thereby improving the overall level of patient adherence and satisfaction. Experimental results from applying EM-based clustering and Apriori algorithms show that the developed algorithm can predict further events with higher confidence levels and reduce the number of user tasks by up to 76.19 %.

Salient Segmentation of Medical Time Series Signals.

Searching and mining medical time series databases is extremely challenging due to large, high entropy, and multidimensional datasets. Traditional time series databases are populated using segments extracted by a sliding window. The resulting database index contains an abundance of redundant time series segments with little to no alignment. This paper presents the idea of "salient segmentation". Salient segmentation is a probabilistic segmentation technique for populating medical time series databases. Segments with the lowest probabilities are considered salient and are inserted into the index. The resulting index has little redundancy and is composed of aligned segments. This approach reduces index sizes by more than 98% over conventional sliding window techniques. Furthermore, salient segmentation can reduce redundancy in motif discovery algorithms by more than 85%, yielding a more succinct representation of a time series signal.

A remote patient monitoring system for congestive heart failure.

Congestive heart failure (CHF) is a leading cause of death in the United States affecting approximately 670,000 individuals. Due to the prevalence of CHF related issues, it is prudent to seek out methodologies that would facilitate the prevention, monitoring, and treatment of heart disease on a daily basis. This paper describes WANDA (Weight and Activity with Blood Pressure Monitoring System); a study that leverages sensor technologies and wireless communications to monitor the health related measurements of patients with CHF. The WANDA system is a three-tier architecture consisting of sensors, web servers, and back-end databases. The system was developed in conjunction with the UCLA School of Nursing and the UCLA Wireless Health Institute to enable early detection of key clinical symptoms indicative of CHF-related decompensation. This study shows that CHF patients monitored by WANDA are less likely to have readings fall outside a healthy range. In addition, WANDA provides a useful feedback system for regulating readings of CHF patients.

Missing data imputation for remote CHF patient monitoring systems.

Congestive heart failure (CHF) is a leading cause of death in the United States. WANDA is a wireless health project that leverages sensor technology and wireless communication to monitor the health status of patients with CHF. The first pilot study of WANDA showed the system's effectiveness for patients with CHF. However, WANDA experienced a considerable amount of missing data due to system misuse, nonuse, and failure. Missing data is highly undesirable as automated alarms may fail to notify healthcare professionals of potentially dangerous patient conditions. In this study, we exploit machine learning techniques including projection adjustment by contribution estimation regression (PACE), Bayesian methods, and voting feature interval (VFI) algorithms to predict both non-binomial and binomial data. The experimental results show that the aforementioned algorithms are superior to other methods with high accuracy and recall. This approach also shows an improved ability to predict missing data when training on entire populations, as opposed to training unique classifiers for each individual.

Linear Frequency Estimation Technique for Reducing Frequency Based Signals.

This paper presents a linear frequency estimation (LFE) technique for data reduction of frequency-based signals. LFE converts a signal to the frequency domain by utilizing the Fourier transform and estimates both the real and imaginary parts with a series of vectors much smaller than the original signal size. The estimation is accomplished by selecting optimal points from the frequency domain and interpolating data between these points with a first order approximation. The difficulty of such a problem lies in determining which points are most significant. LFE is unique in the fact that it is generic to a wide variety of frequency-based signals such as electromyography (EMG), voice, and electrocardiography (ECG). The only requirement is that spectral coefficients are spatially correlated. This paper presents the algorithm and results from both EMG and voice data. We complete the paper with a description of how this method can be applied to pattern types of recognition, signal indexing, and compression.