Class Distribution Curve Based Discretization With Application to Wearable Sensors and Medical Monitoring

Nicholas Skapura; Guozhu Dong

doi:10.4018/ijmstr.2017100102

Class Distribution Curve Based Discretization With Application to Wearable Sensors and Medical Monitoring

International Journal of Monitoring and Surveillance Technologies Research ◽

10.4018/ijmstr.2017100102 ◽

2017 ◽

Vol 5 (4) ◽

pp. 23-37 ◽

Cited By ~ 1

Author(s):

Nicholas Skapura ◽

Guozhu Dong

Keyword(s):

Human Activities ◽

Pattern Mining ◽

Distribution Curve ◽

Wearable Sensors ◽

Main Idea ◽

Sensor Technology ◽

Medical Monitoring ◽

Wearable Sensor ◽

High Quality ◽

Class Distribution

Understanding diseases and human activities, and constructing highly accurate classifiers are two important tasks in bio-medicine, healthcare, and wearable sensor technology. Being able to mine high-quality patterns is useful here, as such patterns can help improve understanding and build accurate classifiers. However, most pattern mining algorithms only operate on discrete data; applying them often requires a binning step to discretize continuous attributes. This article presents a new discretization technique, called Class Distribution Curve based Binning (CDC Binning); the main idea is to use a so-called class distribution curve, which measures the class purity in sliding windows over an attribute's range, to construct binning intervals. Experiments show that (1) CDC Binning outperforms existing binning methods in discovering high-quality patterns, especially when the class distribution curve is complicated (e.g. when the two classes are two fairly similar human activities), and (2) it can outperform other binning methods by 10% in classification accuracy when using discovered patterns as features. CDC Binning is particularly useful for applications where the classes/activities to be distinguished are similar to each other. This is especially important in wearable sensor technology where detection of behavioral or activity changes in a person (e.g. fall detection) could indicate a significant medical event.

Download Full-text

Automated P-Wave Quality Assessment for Wearable Sensors

Proceedings ◽

10.3390/ecsa-5-05743 ◽

2018 ◽

Vol 4 (1) ◽

pp. 13

Author(s):

Diogo Tecelão ◽

Peter Charlton

Keyword(s):

At Risk ◽

Quality Assessment ◽

Wearable Sensors ◽

P Wave ◽

Sensor Data ◽

Wearable Sensor ◽

High Quality ◽

P Waves ◽

Ecg Signals ◽

Wave Morphology

Hospital patients recovering from major cardiac surgery are at risk of paroxysmal atrial fibrillation (AF), an arrhythmia which can be life-threatening. Wearable sensors are routinely used for electrocardiogram (ECG) monitoring in patients at risk of AF, providing real-time AF detection. However, wearable sensors could have greater impact if used to identify the subtle changes in P-wave morphology which precede AF. This would allow prophylactic treatment to be administered, potentially preventing AF. However, ECG signals acquired by wearable sensors are susceptible to artefact, making it difficult to distinguish between physiological changes in P-wave morphology, and changes due to noise. The aim of this study was to design and assess the performance of a novel automated P-wave quality assessment tool to identify high-quality P-waves, for AF prediction. We designed a two-stage algorithm which uses P-wave template-matching to assess quality. Its performance was assessed using the AFPDB, a database of wearable sensor ECG signals acquired from both healthy subjects and patients susceptible to AF. The algorithm’s quality assessments of 97,989 P-waves were compared to manual annotations. The algorithm identified high-quality P-waves with high sensitivity (93%) and good specificity (82%), indicating that it may have utility for identifying high-quality P-waves in wearable sensor data. Measurements of P-wave morphology derived from high-quality P-waves could be used to predict AF, improving patient outcomes, and reducing healthcare costs. Further studies assessing the clinical utility of the presented tool are warranted for validation.

Download Full-text

Patient perspectives of wearable sensors: a mixed methods evaluation (Preprint)

10.2196/preprints.18836 ◽

2020 ◽

Author(s):

Meera Joshi ◽

Stephanie Archer ◽

Abigail Morbi ◽

Sonal Arora ◽

Richard Kwasnicki ◽

...

Keyword(s):

Wearable Sensors ◽

Surgical Patients ◽

Sensor Technology ◽

Patient Questionnaire ◽

Continuous Variables ◽

Wearable Sensor ◽

Structured Interviews ◽

Patient Deterioration ◽

Mixed Methods Evaluation ◽

Interview Studies

BACKGROUND Continuous vital sign monitoring using wearable sensors may enable earlier detection of patient deterioration and sepsis. OBJECTIVE To explore patient experiences of wearable sensor technology and continuous monitoring through questionnaire and interview studies. METHODS All patients recruited for a wearable sensor study were asked to complete a study questionnaire. Patients were asked 9 questions with answers on Likert scale and scores were treated as continuous variables. A subgroup of surgical patients wearing the wearable sensor were invited to take part in semi-structured interviews. All interview data was analysed using thematic analysis. RESULTS A total of 453 patients completed the patient questionnaire (90.6% response rate). A high proportion of patients agreed the wearable sensor was comfortable to wear (n=427, 85.4%), they would wear the patch again when in hospital (n=429, 85.8%) and they would wear the wearable patch at home (n=398, 79.6%). Twelve surgical patients consented to interviews. Five main themes of interest to patients emerged from the interviews; 1) Centralised monitoring 2) enhanced feelings of patient safety, 3) impact on nursing staff 4) comfort & usability and 5) the future and views on technology. CONCLUSIONS Overall, the feedback from patients using wearable monitoring was strongly positive with relatively few concerns raised. Patients feel wearable sensors improve their sense of safety, may relieve pressure on healthcare staff and are a welcome part of future healthcare

Download Full-text

The Use of Wearable Sensor Technology to Detect Shock Impacts in Sports and Occupational Settings: A Scoping Review

Sensors ◽

10.3390/s21154962 ◽

2021 ◽

Vol 21 (15) ◽

pp. 4962

Author(s):

Ingrid Eitzen ◽

Julie Renberg ◽

Hilde Færevik

Keyword(s):

Scoping Review ◽

Wearable Sensors ◽

Real Life ◽

Team Sports ◽

Sensor Technology ◽

Wearable Sensor ◽

Dominant Type ◽

Inner Organs ◽

Shock Impacts ◽

Occupational Settings

Shock impacts during activity may cause damage to the joints, muscles, bones, or inner organs. To define thresholds for tolerable impacts, there is a need for methods that can accurately monitor shock impacts in real-life settings. Therefore, the main aim of this scoping review was to present an overview of existing methods for assessments of shock impacts using wearable sensor technology within two domains: sports and occupational settings. Online databases were used to identify papers published in 2010–2020, from which we selected 34 papers that used wearable sensor technology to measure shock impacts. No studies were found on occupational settings. For the sports domain, accelerometry was the dominant type of wearable sensor technology utilized, interpreting peak acceleration as a proxy for impact. Of the included studies, 28 assessed foot strike in running, head impacts in invasion and team sports, or different forms of jump landings or plyometric movements. The included studies revealed a lack of consensus regarding sensor placement and interpretation of the results. Furthermore, the identified high proportion of validation studies support previous concerns that wearable sensors at present are inadequate as a stand-alone method for valid and accurate data on shock impacts in the field.

Download Full-text

It’s not about the capture, it’s about what we can learn”: a qualitative study of experts’ opinions and experiences regarding the use of wearable sensors to measure gait and physical activity

Journal of NeuroEngineering and Rehabilitation ◽

10.1186/s12984-021-00874-8 ◽

2021 ◽

Vol 18 (1) ◽

Author(s):

Alison Keogh ◽

Kristin Taraldsen ◽

Brian Caulfield ◽

Beatrix Vereijken

Keyword(s):

Physical Activity ◽

Clinical Practice ◽

Paradigm Shift ◽

Wearable Sensors ◽

Wearable Devices ◽

Healthcare Research ◽

Quality Data ◽

Sensor Technology ◽

Wearable Sensor ◽

Structured Interviews

Abstract Background The use of wearable sensor technology to collect patient health data, such as gait and physical activity, offers the potential to transform healthcare research. To maximise the use of wearable devices in practice, it is important that they are usable by, and offer value to, all stakeholders. Although previous research has explored participants’ opinions of devices, to date, limited studies have explored the experiences and opinions of the researchers who use and implement them. Researchers offer a unique insight into wearable devices as they may have access to multiple devices and cohorts, and thus gain a thorough understanding as to how and where this area needs to progress. Therefore, the aim of this study was to explore the experiences and opinions of researchers from academic, industry and clinical contexts, in the use of wearable devices to measure gait and physical activity. Methods Twenty professionals with experience using wearable devices in research were recruited from academic, industry and clinical backgrounds. Independent, semi-structured interviews were conducted, audio-recorded and transcribed. Transcribed texts were analysed using inductive thematic analysis. Results Five themes were identified: (1) The positives and negatives of using wearable devices in research, (2) The routine implementation of wearable devices into research and clinical practice, (3) The importance of compromise in protocols, (4) Securing good quality data, and (5) A paradigm shift. Researchers overwhelmingly supported the use of wearable sensor technology due to the insights that they may provide. Though barriers remain, researchers were pragmatic towards these, believing that there is a paradigm shift happening in this area of research that ultimately requires mistakes and significant volumes of further research to allow it to progress. Conclusions Multiple barriers to the use of wearable devices in research and clinical practice remain, including data management and clear clinical utility. However, researchers strongly believe that the potential benefit of these devices to support and create new clinical insights for patient care, is greater than any current barrier. Multi-disciplinary research integrating the expertise of both academia, industry and clinicians is a fundamental necessity to further develop wearable devices and protocols that match the varied needs of all stakeholders.

Download Full-text

Sociometric wearable devices for studying human behavior in corporate and healthcare workplaces

BioTechniques ◽

10.2144/btn-2020-0160 ◽

2021 ◽

Author(s):

Asami Ito-Masui ◽

Eiji Kawamoto ◽

Ryo Esumi ◽

Hiroshi Imai ◽

Motomu Shimaoka

Keyword(s):

Wearable Sensors ◽

Wearable Devices ◽

Healthcare Research ◽

Human Interaction ◽

Sensor Technology ◽

Wearable Sensor ◽

Social Constructs ◽

Medical Settings ◽

Human Interactions ◽

Study Designs

Wearable sensor technology enables objective data collection of direct human interactions. The authors review sociometric wearable devices (SWD) and their application in healthcare. Human interactions captured by wearable sensors have been shown to correlate with social constructs such as teamwork and productivity in the office. Application of SWD in the field of healthcare requires special considerations: validation studies have shown technological disadvantages in acute medical settings. Application of SWD in healthcare should be considered based on the strengths and weaknesses of the methodology. SWD can also play an important role in investigation of human interaction and epidemic spread. When study designs and methodologies are carefully considered, incorporation of SWD in healthcare research has promising potential for new insights.

Download Full-text

Voodoo Machine Learning for Clinical Predictions

10.1101/059774 ◽

2016 ◽

Cited By ~ 17

Author(s):

Sohrab Saeb ◽

Luca Lonini ◽

Arun Jayaraman ◽

David C. Mohr ◽

Konrad P. Kording

Keyword(s):

Machine Learning ◽

Prediction Accuracy ◽

Cross Validation ◽

Clinical Decision Making ◽

Wearable Sensors ◽

Clinical Decision ◽

Machine Learning Algorithms ◽

Sensor Technology ◽

Wearable Sensor ◽

Rapid Accumulation

AbstractThe availability of smartphone and wearable sensor technology is leading to a rapid accumulation of human subject data, and machine learning is emerging as a technique to map that data into clinical predictions. As machine learning algorithms are increasingly used to support clinical decision making, it is important to reliably quantify their prediction accuracy. Cross-validation is the standard approach for evaluating the accuracy of such algorithms; however, several cross-validations methods exist and only some of them are statistically meaningful. Here we compared two popular cross-validation methods: record-wise and subject-wise. Using both a publicly available dataset and a simulation, we found that record-wise cross-validation often massively overestimates the prediction accuracy of the algorithms. We also found that this erroneous method is used by almost half of the retrieved studies that used accelerometers, wearable sensors, or smartphones to predict clinical outcomes. As we move towards an era of machine learning based diagnosis and treatment, using proper methods to evaluate their accuracy is crucial, as erroneous results can mislead both clinicians and data scientists.

Download Full-text

Recommendations for Assessment of the Reliability, Sensitivity, and Validity of Data Provided by Wearable Sensors Designed for Monitoring Physical Activity (Preprint)

10.2196/preprints.9341 ◽

2017 ◽

Author(s):

Peter Düking ◽

Franz Konstantin Fuss ◽

Hans-Christer Holmberg ◽

Billy Sperlich

Keyword(s):

Physical Activity ◽

Wearable Sensors ◽

Wearable Devices ◽

Insurance Companies ◽

Sensor Technology ◽

Wearable Sensor ◽

Scientific Evaluation ◽

Reliability Sensitivity ◽

The Individual ◽

Recommendations For Assessment

UNSTRUCTURED Although it is becoming increasingly popular to monitor parameters related to training, recovery, and health with wearable sensor technology (wearables), scientific evaluation of the reliability, sensitivity, and validity of such data is limited and, where available, has involved a wide variety of approaches. To improve the trustworthiness of data collected by wearables and facilitate comparisons, we have outlined recommendations for standardized evaluation. We discuss the wearable devices themselves, as well as experimental and statistical considerations. Adherence to these recommendations should be beneficial not only for the individual, but also for regulatory organizations and insurance companies.

Download Full-text

The Promise and Perils of Wearable Sensors in Organizational Research

Organizational Research Methods ◽

10.1177/1094428115617004 ◽

2016 ◽

Vol 20 (1) ◽

pp. 3-31 ◽

Cited By ~ 59

Author(s):

Daniel Chaffin ◽

Ralph Heidl ◽

John R. Hollenbeck ◽

Michael Howe ◽

Andrew Yu ◽

...

Keyword(s):

Large Scale ◽

Wearable Sensors ◽

Active Role ◽

Sensor Technology ◽

Wearable Sensor ◽

Organizational Research ◽

High Resolution Data ◽

Human Interactions ◽

Wearable Electronic ◽

The Individual

Rapid advances in mobile computing technology have the potential to revolutionize organizational research by facilitating new methods of data collection. The emergence of wearable electronic sensors in particular harbors the promise of making the large-scale collection of high-resolution data related to human interactions and social behavior economically viable. Popular press and practitioner-oriented research outlets have begun to tout the game-changing potential of wearable sensors for both researchers and practitioners. We systematically examine the utility of current wearable sensor technology for capturing behavioral constructs at the individual and team levels. In the process, we provide a model for performing validation work in this new domain of measurement. Our findings highlight the need for organizational researchers to take an active role in the development of wearable sensor systems to ensure that the measures derived from these devices and sensors allow us to leverage and extend the extant knowledge base. We also offer a caution regarding the potential sources of error arising from wearable sensors in behavioral research.

Download Full-text