Data mining of motor patterns recorded with wearable technology

Sensor Technologies to Manage the Physiological Traits of Chronic Pain: A Review

Non-oncologic chronic pain is a common high-morbidity impairment worldwide and acknowledged as a condition with significant incidence on quality of life. Pain intensity is largely perceived as a subjective experience, what makes challenging its objective measurement. However, the physiological traces of pain make possible its correlation with vital signs, such as heart rate variability, skin conductance, electromyogram, etc., or health performance metrics derived from daily activity monitoring or facial expressions, which can be acquired with diverse sensor technologies and multisensory approaches. As the assessment and management of pain are essential issues for a wide range of clinical disorders and treatments, this paper reviews different sensor-based approaches applied to the objective evaluation of non-oncological chronic pain. The space of available technologies and resources aimed at pain assessment represent a diversified set of alternatives that can be exploited to address the multidimensional nature of pain.

Building Wearables for Geology

Wearable devices have emerged in the last years with new applications that provide user convenience. Healthcare, sports, safety are some examples of applications embedded in thousands of devices released in the last years. Wearable operating systems with different focus emerged together with wearable applications in order to make adjustments and optimizations of software and hardware. This paper presents a wearable operating systems discussion and shows the current challenges and wearable operating system inuence. We developed a wearable appliance for geology. The wearable contains a Head Mounted Display (HMD) assembled with Google Cardboard API and sensors connected to developments boards. For each system component was used different operating systems according to hardware and software available. The results indicate some trends for wearable operating systems.

A novel data-mining approach leveraging social media to monitor consumer opinion of sitagliptin

A novel data mining method was developed to gauge the experience of the drug Sitagliptin (trade name Januvia) by patients with diabetes mellitus type 2. To this goal, we devised a two-step analysis framework. Initial exploratory analysis using self-organizing maps was performed to determine structures based on user opinions among the forum posts. The results were a compilation of user's clusters and their correlated (positive or negative) opinion of the drug. Subsequent modeling using network analysis methods was used to determine influential users among the forum members. These findings can open new avenues of research into rapid data collection, feedback, and analysis that can enable improved outcomes and solutions for public health and important feedback for the manufacturer.

Network-based modeling and intelligent data mining of social media for improving care

Intelligently extracting knowledge from social media has recently attracted great interest from the Biomedical and Health Informatics community to simultaneously improve healthcare outcomes and reduce costs using consumer-generated opinion. We propose a two-step analysis framework that focuses on positive and negative sentiment, as well as the side effects of treatment, in users' forum posts, and identifies user communities (modules) and influential users for the purpose of ascertaining user opinion of cancer treatment. We used a self-organizing map to analyze word frequency data derived from users' forum posts. We then introduced a novel network-based approach for modeling users' forum interactions and employed a network partitioning method based on optimizing a stability quality measure. This allowed us to determine consumer opinion and identify influential users within the retrieved modules using information derived from both word-frequency data and network-based properties. Our approach can expand research into intelligently mining social media data for consumer opinion of various treatments to provide rapid, up-to-date information for the pharmaceutical industry, hospitals, and medical staff, on the effectiveness (or ineffectiveness) of future treatments.

A novel data-mining platform leveraging social media to monitor outcomes of Januvia

A novel data-mining method was developed to gauge the experiences of the diabetes mellitus drug Januvia. Self-organizing maps were used to analyze forum posts numerically to infer user opinion of drug Januvia. Graph theory was used to discover influential users. The result is a word list compilation correlating positive and negative word cluster groups and a web of influential users on Januvia. The implications could open new research avenues into rapid data collection, feedback, and analysis that would enable improved solutions for public health.

Monitoring motor capacity changes of children during rehabilitation using body-worn sensors

BACKGROUND

Rehabilitation services use outcome measures to track motor performance of their patients over time. State-of-the-art approaches use mainly patients' feedback and experts' observations for this purpose. We aim at continuously monitoring children in daily life and assessing normal activities to close the gap between movements done as instructed by caregivers and natural movements during daily life. To investigate the applicability of body-worn sensors for motor assessment in children, we investigated changes in movement capacity during defined motor tasks longitudinally.

METHODS

We performed a longitudinal study over four weeks with 4 children (2 girls; 2 diagnosed with Cerebral Palsy and 2 with stroke, on average 10.5 years old) undergoing rehabilitation. Every week, the children performed 10 predefined motor tasks. Capacity in terms of quality and quantity was assessed by experts and movement was monitored using 10 ETH Orientation Sensors (ETHOS), a small and unobtrusive inertial measurement unit. Features such as smoothness of movement were calculated from the sensor data and a regression was used to estimate the capacity from the features and their relation to clinical data. Therefore, the target and features were normalized to range from 0 to 1.

RESULTS

We achieved a mean RMS-error of 0.15 and a mean correlation value of 0.86 (p < 0.05 for all tasks) between our regression estimate of motor task capacity and experts' ratings across all tasks. We identified the most important features and were able to reduce the sensor setup from 10 to 3 sensors. We investigated features that provided a good estimate of the motor capacity independently of the task performed, e.g. smoothness of the movement.

CONCLUSIONS

We found that children's task capacity can be assessed from wearable sensors and that some of the calculated features provide a good estimate of movement capacity over different tasks. This indicates the potential of using the sensors in daily life, when little or no information on the task performed is available. For the assessment, the use of three sensors on both wrists and the hip suffices. With the developed algorithms, we plan to assess children's motor performance in daily life with a follow-up study.

Wearable sensors and systems. From enabling technology to clinical applications

It is now more than 50 years since the time when clinical monitoring of individuals in the home and community settings was first envisioned. Until recently, technologies to enable such vision were lacking. However, wearable sensors and systems developed over the past decade have provided the tools to finally implement and deploy technology with the capabilities required by researchers in the field of patients' home monitoring. As discussed, potential applications of these technologies include the early diagnosis of diseases such as congestive heart failure, the prevention of chronic conditions such as diabetes, improved clinical management of neurodegenerative conditions such as Parkinson's disease, and the ability to promptly respond to emergency situations such as seizures in patients with epilepsy and cardiac arrest in subjects undergoing cardiovascular monitoring. Current research efforts are now focused on the development of more complex systems for home monitoring of individuals with a variety of preclinical and clinical conditions. Recent research on the clinical assessment of wearable technology promises to deliver methodologies that are expected to lead to clinical adoption within the next five to ten years. In particular, combining home robots and wearable technology is likely to be a key step toward achieving the goal of effectively monitoring patients in the home. These efforts to merge home robots and wearable technology are expected to enable a new generation of complex systems with the ability to monitor subjects' status, facilitate the administration of interventions, and provide an invaluable tool to respond to emergency situations.

A novel automatic method for monitoring Tourette motor tics through a wearable device

The aim of this study was to propose a novel automatic method for quantifying motor-tics caused by the Tourette Syndrome (TS). In this preliminary report, the feasibility of the monitoring process was tested over a series of standard clinical trials in a population of 12 subjects affected by TS. A wearable instrument with an embedded three-axial accelerometer was used to detect and classify motor tics during standing and walking activities. An algorithm was devised to analyze acceleration data by: eliminating noise; detecting peaks connected to pathological events; and classifying intensity and frequency of motor tics into quantitative scores. These indexes were compared with the video-based ones provided by expert clinicians, which were taken as the gold-standard. Sensitivity, specificity, and accuracy of tic detection were estimated, and an agreement analysis was performed through the least square regression and the Bland-Altman test. The tic recognition algorithm showed sensitivity = 80.8% ± 8.5% (mean ± SD), specificity = 75.8% ± 17.3%, and accuracy = 80.5% ± 12.2%. The agreement study showed that automatic detection tended to overestimate the number of tics occurred. Although, it appeared this may be a systematic error due to the different recognition principles of the wearable and video-based systems. Furthermore, there was substantial concurrency with the gold-standard in estimating the severity indexes. The proposed methodology gave promising performances in terms of automatic motor-tics detection and classification in a standard clinical context. The system may provide physicians with a quantitative aid for TS assessment. Further developments will focus on the extension of its application to everyday long-term monitoring out of clinical environments.

In vivo laboratory validation of the physiometer: a measurement system for long-term recording of posture and movements in the workplace

Posture and movement are thought to be important risk factors for the development of work-related musculoskeletal disorders. Whole day occupational exposure assessment has typically used self-report or observation techniques, but the need for more accurate measurement is now recognised. The aim of this study was to compare the kinematic recordings of a frequently used field system (physiometer) with two laboratory-based systems (Fastrak and Peak) in vivo. Head, thorax and right arm kinematics were recorded simultaneously by the three systems whilst a subject performed 27 single and multiple plane physiological and simulated daily living task movement trials. Errors observed in the Fastrak and Peak data included gimbal lock and quadrant errors. Physiometer data errors included undervalues, overvalues and temporal errors of slow response and resonance. All three systems showed some cross-talk. Agreement between the physiometer and the other systems was generally high for physiological movements (R(2) > 0.8) and less for functional movements (R(2) > 0.5). STATEMENT OF RELEVANCE: The physiometer recording device can provide an indication of posture across time in the workplace; however, its accuracy is limited, particularly during functional movements. Further technology should be developed to unobtrusively capture accurate all day 3-D kinematics.

Bluetooth-based sensor networks for remotely monitoring the physiological signals of a patient

Integrating intelligent medical microsensors into a wireless communication network makes it possible to remotely collect physiological signals of a patient, release the patient from being tethered to monitoring medical instrumentations, and facilitate the patient's early hospital discharge. This can further improve life quality by providing continuous observation without the need of disrupting the patient's normal life, thus reducing the risk of infection significantly, and decreasing the cost of the hospital and the patient. This paper discusses the implementation issues, and describes the overall system architecture of our developed Bluetooth sensor network for patient monitoring and the corresponding heart activity sensors. It also presents our approach to developing the intelligent physiological sensor nodes involving integration of Bluetooth radio technology, hardware and software organization, and our solutions for onboard signal processing.

Advances in the application of technology to epilepsy: the CIMIT/NIO Epilepsy Innovation Summit

In 2008, a group of clinicians, scientists, engineers, and industry representatives met to discuss advances in the application of engineering technologies to the diagnosis and treatment of patients with epilepsy. The presentations also provided a guide for further technological development, specifically in the evaluation of patients for epilepsy surgery, seizure onset detection and seizure prediction, intracranial treatment systems, and extracranial treatment systems. This article summarizes the discussions and demonstrates that cross-disciplinary interactions can catalyze collaborations between physicians and engineers to address and solve many of the pressing unmet needs in epilepsy.

A data-mining approach for investigating social and economic geographical dynamics of beta-thalassemia's spread

Beta-thalassemia is an anemic genetic disorder that remains a major global health issue, especially in the globalized era where public health, economics, and education are tightly interwoven. Previous studies have examined the disease's rate and heredity. This study analyzed beta-thalassemia's socioeconomic geography and how it affects the afflicted population. We processed survey data and performed data mining using self-organizing maps to identify underlying data structure. We hypothesized that certain variables mark subgroups within the affected population and we aimed at identifying these subgroups and used a correlation-based measure to assess the variable's importance to the subgroup's distinction. The population's education level was one of the major factors that divided it into different subgroups. Our study showed that recurring patterns of specific variables separated the affected population into disparate subgroups based on their response to questionnaires. Future studies can use such tools to delve deeper into how other variables (e.g. socioeconomic and genomic) can identify subgroups within larger affected populations.

A wearable pelvic sensor design for drop foot treatment in post-stroke patients

A novel wearable pelvic sensor design for gait analysis was developed and evaluated in both normal and pathological gait. The device is a hip worn fusion of gyroscopes and accelerometers which allows for monitoring of the periodic vertical rotation of the pelvis during the walking cycle and uses this information as a predictor of gait events such as Heel-Strike (HS) and Toe-Off (TO). The gait pattern of two age and gender-matched groups (40-65 years) of 10 healthy subjects (5 male, 5 female) and 10 subjects with hemiplegic drop foot were examined. The pelvic sensor method was correlated against an optical motion system and footswitches, to evaluate the technique's efficacy at detecting foot contact events in walking and hip pattern. Data analysis showed the device was able to predict foot contact events from recorded maximum and minimum pelvic angle (TO: Healthy - 130ms Hemiplegic - 95ms; HS: Healthy - 127ms, Hemiplegic - 96ms). This ability to detect gait events would allow this sensor design to be used in simplifying drop foot stimulation systems. The proposed method also records the relative range of motion of the pelvis from which useful information on gait symmetry can be obtained and used in ambulatory monitoring or treatment intervention analysis.

Extraction of information on elder motor ability from clinical and biomechanical data through data mining

This study aimed at evaluating the additional knowledge provided by a biomechanical test coupled with clinical tests for motor ability assessment. A database including clinical test scores and sit-to-stand test variables obtained from 110 medically stable elderly subjects was submitted to data mining by searching for association rules. The presence of rules revealed some redundancies due to sample homogeneity, as mainly observed in the joint analysis of a questionnaire for daily activities assessment (Nottingham test) and the sit-to-stand, and due to similar evaluated information, as resulted from the joint analysis of a balance and gait scale (Tinetti test) and the sit-to-stand. Conversely, when no association rules were found, the tests carried unrelated information. The associations mined while analysing these clinical tests encouraged the integration of biomechanical tests, increasing significantly its clinical applicability and reducing the information redundancy. The information extracted also allowed to highlight rules typical of elderly persons that may serve as a knowledge-based tool for the detection of possible deviation from normality.

Knowledge discovery in databases of biomechanical variables: application to the sit to stand motor task

ABSTRACT : BACKGROUND : The interpretation of data obtained in a movement analysis laboratory is a crucial issue in clinical contexts. Collection of such data in large databases might encourage the use of modern techniques of data mining to discover additional knowledge with automated methods. In order to maximise the size of the database, simple and low-cost experimental set-ups are preferable. The aim of this study was to extract knowledge inherent in the sit-to-stand task as performed by healthy adults, by searching relationships among measured and estimated biomechanical quantities. An automated method was applied to a large amount of data stored in a database. The sit-to-stand motor task was already shown to be adequate for determining the level of individual motor ability. METHODS : The technique of search for association rules was chosen to discover patterns as part of a Knowledge Discovery in Databases (KDD) process applied to a sit-to-stand motor task observed with a simple experimental set-up and analysed by means of a minimum measured input model. Selected parameters and variables of a database containing data from 110 healthy adults, of both genders and of a large range of age, performing the task were considered in the analysis. RESULTS : A set of rules and definitions were found characterising the patterns shared by the investigated subjects. Time events of the task turned out to be highly interdependent at least in their average values, showing a high level of repeatability of the timing of the performance of the task. CONCLUSIONS : The distinctive patterns of the sit-to-stand task found in this study, associated to those that could be found in similar studies focusing on subjects with pathologies, could be used as a reference for the functional evaluation of specific subjects performing the sit-to-stand motor task.