Beat-to-beat variability of fetal myocardial performance index.

OBJECTIVES: To determine whether there is beat-to-beat (BTB) variability in the fetal left myocardial performance index (MPI), as evaluated by an automated system, and whether there is a correlation between MPI and fetal heart rate (FHR). METHODS: This was a prospective cross-sectional study of uncomplicated, morphologically normal, singleton pregnancies at 20-38 weeks' gestation. Multiple cineloops for left MPI measurement were acquired during a single examination of each fetus. Raw cineloop data were analyzed by our automated MPI system (intraclass correlation coefficient of 1.0 for any given waveform) to produce a set of MPIs. The corresponding instantaneous FHR was measured for each individual cardiac cycle for which MPI was calculated. RESULTS: Data from 29 fetuses were analyzed; mean MPI was 0.52, mean FHR was 150 beats per min and the median number of cardiac cycles examined per fetus was 70 (interquartile range, 31-115). Marked BTB variability was noted; median coefficient of variation was 10% (range, 5.5-13.9%). FHR was weakly correlated with absolute MPI (r = 0.22; P < 0.05). BTB variation in MPI as a percentage of the mean MPI was not significantly correlated with FHR (r = 0.031; P = 0.146). When standard error of the mean of all MPI values was divided by the mean for each case, it showed that at least four cardiac cycles should be averaged to reduce MPI variability to approximately ± 5%. CONCLUSION: There is significant BTB variability in fetal left MPI, which has an overall weak correlation with FHR. This could be a factor affecting the consistency of MPI values reported by different research groups. Variability would be reduced by averaging 4-5 cardiac cycles per fetus. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd.

Unintended Consequences of Wearable Sensor Use in Healthcare. Contribution of the IMIA Wearable Sensors in Healthcare WG.

OBJECTIVES: As wearable sensors take the consumer market by storm, and medical device manufacturers move to make their devices wireless and appropriate for ambulatory use, this revolution brings with it some unintended consequences, which we aim to discuss in this paper. METHODS: We discuss some important unintended consequences, both beneficial and unwanted, which relate to: modifications of behavior; creation and use of big data sets; new security vulnerabilities; and unforeseen challenges faced by regulatory authorities, struggling to keep pace with recent innovations. Where possible, we proposed potential solutions to unwanted consequences. RESULTS: Intelligent and inclusive design processes may mitigate unintended modifications in behavior. For big data, legislating access to and use of these data will be a legal and political challenge in the years ahead, as we trade the health benefits of wearable sensors against the risk to our privacy. The wireless and personal nature of wearable sensors also exposes them to a number of unique security vulnerabilities. Regulation plays an important role in managing these security risks, but also has the dual responsibility of ensuring that wearable devices are fit for purpose. However, the burden of validating the function and security of medical devices is becoming infeasible for regulators, given the many software apps and wearable sensors entering the market each year, which are only a subset of an even larger 'internet of things'. CONCLUSION: Wearable sensors may serve to improve wellbeing, but we must be vigilant against the occurrence of unintended consequences. With collaboration between device manufacturers, regulators, and end-users, we balance the risk of unintended consequences occurring against the incredible benefit that wearable sensors promise to bring to the world.

What Does Big Data Mean for Wearable Sensor Systems? Contribution of the IMIA Wearable Sensors in Healthcare WG.

OBJECTIVES: The aim of this paper is to discuss how recent developments in the field of big data may potentially impact the future use of wearable sensor systems in healthcare. METHODS: The article draws on the scientific literature to support the opinions presented by the IMIA Wearable Sensors in Healthcare Working Group. RESULTS: The following is discussed: the potential for wearable sensors to generate big data; how complementary technologies, such as a smartphone, will augment the concept of a wearable sensor and alter the nature of the monitoring data created; how standards would enable sharing of data and advance scientific progress. Importantly, attention is drawn to statistical inference problems for which big datasets provide little assistance, or may hinder the identification of a useful solution. Finally, a discussion is presented on risks to privacy and possible negative consequences arising from intensive wearable sensor monitoring. CONCLUSIONS: Wearable sensors systems have the potential to generate datasets which are currently beyond our capabilities to easily organize and interpret. In order to successfully utilize wearable sensor data to infer wellbeing, and enable proactive health management, standards and ontologies must be developed which allow for data to be shared between research groups and between commercial systems, promoting the integration of these data into health information systems. However, policy and regulation will be required to ensure that the detailed nature of wearable sensor data is not misused to invade privacies or prejudice against individuals.

Development of a standard fall data format for signals from body-worn sensors : the FARSEEING consensus.

Objective measurement of real-world fall events by using body-worn sensor devices can improve the understanding of falls in older people and enable new technology to prevent, predict, and automatically recognize falls. However, these events are rare and hence challenging to capture. The FARSEEING (FAll Repository for the design of Smart and sElf-adapaive Environments prolonging INdependent livinG) consortium and associated partners strongly argue that a sufficient dataset of real-world falls can only be acquired through a collaboration of many research groups. Therefore, the major aim of the FARSEEING project is to build a meta-database of real-world falls. To establish this meta-database, standardization of data is necessary to make it possible to combine different sources for analysis and to guarantee data quality. A consensus process was started in January 2012 to propose a standard fall data format, involving 40 experts from different countries and different disciplines working in the field of fall recording and fall prevention. During a web-based Delphi process, possible variables to describe participants, falls, and fall signals were collected and rated by the experts. The summarized results were presented and finally discussed during a workshop at the 20th Conference of the International Society of Posture and Gait Research 2012, in Trondheim, Norway. The consensus includes recommendations for a fall definition, fall reporting (including fall reporting frequency, and fall reporting variables), a minimum clinical dataset, a sensor configuration, and variables to describe the signal characteristics.

Queensland Lung Cancer Screening Study: rationale, design and methods.

BACKGROUND: Lung cancer is the leading cause of cancer-related mortality in Australia. Screening using low-dose computed tomography (LDCT) can reduce lung cancer mortality. The feasibility of screening in Australia is unknown. This paper describes the rationale, design and methods of the Queensland Lung Cancer Screening Study. AIMS: The aim of the study is to describe the methodology for a feasibility study of lung cancer screening by LDCT in Australia. METHODS: The Queensland Lung Cancer Screening Study is an ongoing, prospective observational study of screening by LDCT at a single tertiary institution. Healthy volunteers at high risk of lung cancer (age 60-74 years; smoking history ≥30 pack years, current or quit within 15 years; forced expiratory volume in 1s ≥50% predicted) are recruited from the general public through newspaper advertisement and press release. Participants receive a LDCT scan of the chest at baseline, year 1 and year 2 using a multidetector helical computed tomography scanner and are followed up for a total of 5 years. Feasibility of screening will be assessed by cancer detection rates, lung nodule prevalence, optimal management strategies for lung nodules, economic costs, healthcare utilisation and participant quality of life. CONCLUSIONS: Studying LDCT screening in the Australian setting will help us understand how differences in populations, background diseases and healthcare structures modulate screening effectiveness. This information, together with results from overseas randomised studies, will inform and facilitate local policymaking.

Assessing fall risk using wearable sensors: a practical discussion. A review of the practicalities and challenges associated with the use of wearable sensors for quantification of fall risk in older people.

Identification of older people most at risk of falling may facilitate early preventative intervention to reduce the likelihood of falls occurring. While many clinical fall risk assessment techniques exist, they often require subjective assessor interpretation, or are not appropriate for unsupervised screening of larger populations owing to a number of issues including safety, ability to reliably perform the assessment, and requirements for unwieldy apparatus. Researchers have more recently attempted to address some of these deficits by instrumenting new or existing physical fall risk assessments with wearable motion sensors to make such assessments more objective, quicker to administer, and potentially more appropriate for deployment for unsupervised use in the community. The objective of this paper is to discuss various practical questions involving sensor-based fall risk assessment (SFRA). Many of the issues discussed contribute to answering the important question of whether SFRA should or can be used in either a supervised or an unsupervised manner, and what possible deployment scenarios exist for it.

Electrocardiogram signal quality measures for unsupervised telehealth environments.

The use of telehealth paradigms for the remote management of patients suffering from chronic conditions has become more commonplace with the advancement of Internet connectivity and enterprise software systems. To facilitate clinicians in managing large numbers of telehealth patients, and in digesting the vast array of data returned from the remote monitoring environment, decision support systems in various guises are often utilized. The success of decision support systems in interpreting patient conditions from physiological data is dependent largely on the quality of these recorded data. This paper outlines an algorithm to determine the quality of single-lead electrocardiogram (ECG) recordings obtained from telehealth patients. Three hundred short ECG recordings were manually annotated to identify movement artifact, QRS locations and signal quality (discrete quality levels) by a panel of three experts, who then reconciled the annotation as a group to resolve any discrepancies. After applying a published algorithm to remove gross movement artifact, the proposed method was then applied to estimate the remaining ECG signal quality, using a Parzen window supervised statistical classifier model. The three-class classifier model, using a number of time-domain features and evaluated using cross validation, gave an accuracy in classifying signal quality of 78.7% (κ = 0.67) when using fully automated preprocessing algorithms to remove gross motion artifact and detect QRS locations. This is a similar level of accuracy to the reported human inter-scorer agreement when generating the gold standard annotation (accuracy = 70-89.3%, κ = 0.54-0.84). These results indicate that the assessment of the quality of single-lead ECG recordings, acquired in unsupervised telehealth environments, is entirely feasible and may help to promote the acceptance and utility of future decision support systems for remotely managing chronic disease conditions.

Signal quality measures for unsupervised blood pressure measurement.

Accurate systolic and diastolic pressure estimation, using automated blood pressure measurement, is difficult to achieve when the transduced signals are contaminated with noise or interference, such as movement artifact. This study presents an algorithm for automated signal quality assessment in blood pressure measurement by determining the feasibility of accurately detecting systolic and diastolic pressures when corrupted with various levels of movement artifact. The performance of the proposed algorithm is compared to a manually annotated reference scoring (RS). Based on visual representations and audible playback of Korotkoff sounds, the creation of the RS involved two experts identifying sections of the recorded sounds and annotating sections of noise contamination. The experts determined the systolic and diastolic pressure in 100 recorded Korotkoff sound recordings, using a simultaneous electrocardiograph as a reference signal. The recorded Korotkoff sounds were acquired from 25 healthy subjects (16 men and 9 women) with a total of four measurements per subject. Two of these measurements contained purposely induced noise artifact caused by subject movement. Morphological changes in the cuff pressure signal and the width of the Korotkoff pulse were extracted features which were believed to be correlated with the noise presence in the recorded Korotkoff sounds. Verification of reliable Korotkoff pulses was also performed using extracted features from the oscillometric waveform as recorded from the inflatable cuff. The time between an identified noise section and a verified Korotkoff pulse was the key feature used to determine the validity of possible systolic and diastolic pressures in noise contaminated Korotkoff sounds. The performance of the algorithm was assessed based on the ability to: verify if a signal was contaminated with any noise; the accuracy, sensitivity and specificity of this noise classification, and the systolic and diastolic pressure differences between the result obtained from the algorithm and the RS. 90% of the actual noise contaminated signals were correctly identified, and a sample-wise accuracy, sensitivity and specificity of 97.0%, 80.61% and 98.16%, respectively, were obtained from 100 pooled signals. The mean systolic and diastolic differences were 0.37 ± 3.31 and 3.10 ± 5.46 mmHg, respectively, when the artifact detection algorithm was utilized, with the algorithm correctly determined if the signal was clean enough to attempt an estimation of systolic or diastolic pressures in 93% of blood pressure measurements.

Signal quality measures for pulse oximetry through waveform morphology analysis.

Pulse oximetry has been extensively used to estimate oxygen saturation in blood, a vital physiological parameter commonly used when monitoring a subject's health status. However, accurate estimation of this parameter is difficult to achieve when the fundamental signal from which it is derived, the photoplethysmograph (PPG), is contaminated with noise artifact induced by movement of the subject or the measurement apparatus. This study presents a novel method for automatic rejection of artifact contaminated pulse oximetry waveforms, based on waveform morphology analysis. The performance of the proposed algorithm is compared to a manually annotated gold standard. The creation of the gold standard involved two experts identifying sections of the PPG signal containing good quality PPG pulses and/or noise, in 104 fingertip PPG signals, using a simultaneous electrocardiograph (ECG) signal as a reference signal. The fingertip PPG signals were each 1 min in duration and were acquired from 13 healthy subjects (10 males and 3 females). Each signal contained approximately 20 s of purposely induced artifact noise from a variety of activities involving subject movement. Some unique waveform morphology features were extracted from the PPG signals, which were believed to be correlated with signal quality. A simple decision-tree classifier was employed to arrive at a classification decision, at a pulse-by-pulse resolution, of whether a pulse was of acceptable quality for use or not. The performance of the algorithm was assessed using Cohen's kappa coefficient (κ), sensitivity, specificity and accuracy measures. A mean κ of 0.64 ± 0.22 was obtained, while the mean sensitivity, specificity and accuracy were 89 ± 10%, 77 ± 19% and 83 ± 11%, respectively. Furthermore, a heart rate estimate, extracted from uncontaminated sections of PPG, as identified by the algorithm, was compared with the heart rate derived from an uncontaminated simultaneous ECG signal. The mean error between both heart rate readings was 0.49 ± 0.66 beats per minute (BPM), in comparison to an error value observed without using the artifact detection algorithm of 7.23 ± 5.78 BPM. These results demonstrate that automated identification of signal artifact in the PPG signal through waveform morphology analysis is achievable. In addition, a clear improvement in the accuracy of the derived heart rate is also evident when such methods are employed.