LIV Well when Life Is Limited: Technology to Support Independence at Home.

There has been significant growth in technologies and services creating 'care at home' ecosystems for people with life-limiting conditions such as dementia. Dementia is one of the leading causes of disability and loss of independence that causes a heavy burden for families and caregivers. There is a clear need to support independent living of people living with dementia and their caregivers. Health technologies can help to foster supported living and social connection. The LIV app, developed by Miroma Project Factory and piloted in collaboration with CSIRO, was designed to achieve these aims. Here we describe the development and functionality of the app and present the preliminary findings from the pilot trial.

Detecting Depression-related Movement Changes in Older Adults using Smart Home Motion Sensors - A Feasibility Study.

Smart home sensor data is being increasingly used to identify health risks through passive tracking of specific behaviours and activity patterns. This study explored the feasibility of using motion sensor data to track changes in daytime movement patterns within the home, and their potential association with depression in older adults. This study analysed the motion sensor data collected during a one-year smart home trial, and explored their association with Geriatric Depression Scale (GDS) scores collected at three different time points during the trial (i.e., baseline, mid-trial, and end-trial). Our results showed that movement patterns are generally reduced when older adults are in a depressed state compared to when being in a not-depressed state. In particular, the reduced movement activity in depressed states was significant (p<.05) when the participant's GDS state changed between depressed and not-depressed for the first time during the three time points of the trial when GDS was collected.Clinical relevance- Our results establish the feasibility and potential use of motion sensor data from ambient sensors in a smart home for passive and remote assessment of older adults' depression status, that is comparable to their GDS scores, through changes in their in-home day-time movement patterns. Also since reduced movement activity may be a general indicator of potential health risks, this study provides preliminary evidence for using in-home movement activity monitoring as an general indicator of health risks.

The Significance and Limitations of Sensor-based Agitation Detection in People Living with Dementia.

Agitation, a commonly observed behaviour in people living with dementia (PLwD), is frequently interpreted as a response to physiological, environmental, or emotional stress. Agitation has the potential to pose health risks to both individuals and their caregivers, and can contribute to increased caregiver burden and stress. Early detection of agitation can facilitate with timely intervention, which has the potential to prevent escalation to other challenging behaviors. Wearable and ambient sensors are frequently used to monitor physiological and behavioral conditions and the collected signals can be engaged to detect the onset of an agitation episode. This paper delves into the current sensor-based methods for detecting agitation in PLwD, and reviews the strengths and limitations of existing works. Future directions to enable real-time agitation detection to empower caregivers are also deliberated, with a focus on their potential to reduce caregiver burden by facilitating early support, assistance and interventions to timely manage agitation episodes in PLwD.

Sensor-Based Assessment of Social Isolation and Loneliness in Older Adults: A Survey.

Social isolation (SI) and loneliness are 'invisible enemies'. They affect older people's health and quality of life and have significant impact on aged care resources. While in-person screening tools for SI and loneliness exist, staff shortages and psycho-social challenges fed by stereotypes are significant barriers to their implementation in routine care. Autonomous sensor-based approaches can be used to overcome these challenges by enabling unobtrusive and privacy-preserving assessments of SI and loneliness. This paper presents a comprehensive overview of sensor-based tools to assess social isolation and loneliness through a structured critical review of the relevant literature. The aim of this survey is to identify, categorise, and synthesise studies in which sensing technologies have been used to measure activity and behavioural markers of SI and loneliness in older adults. This survey identified a number of feasibility studies using ambient sensors for measuring SI and loneliness activity markers. Time spent out of home and time spent in different parts of the home were found to show strong associations with SI and loneliness scores derived from standard instruments. This survey found a lack of long-term, in-depth studies in this area with older populations. Specifically, research gaps on the use of wearable and smart phone sensors in this population were identified, including the need for co-design that is important for effective adoption and practical implementation of sensor-based SI and loneliness assessment in older adults.

A validation study of a ballistocardiograph sleep tracker against polysomnography.

STUDY OBJECTIVES: Consumer home sleep trackers provide a great opportunity for longitudinal objective sleep monitoring. Nonwearable sleep devices cause little to no disruption in the daily life routine and need little maintenance. However, their validity needs further investigation. This study aims to evaluate the accuracy of sleep outcomes of EMFIT Quantified Sleep (QS), an unobtrusive nonwearable sleep tracker based on ballistocardiography, against polysomnography. METHODS: 62 sleep-lab patients underwent a single clinical polysomnography with measures simultaneously collected through polysomnography and EMFIT QS. Resting heart rate, total sleep time, wake after sleep onset, sleep onset latency, and duration in sleep stages, collected from the 2 devices, were compared using paired t-tests and their agreement analyzed using Bland-Altman plots. Additionally, continuous heart rate and sleep stages in 30-seconds epochs were evaluated. RESULTS: EMFIT QS data loss occurred in 47% of participants. In the remaining 33 participants (15 women, with mean age of 53.7 ± 16.5 years), EMFIT QS overestimated total sleep time by 177.5 ± 119.4 minutes (p<0.001) and underestimated wake after sleep onset by 44.74 ± 68.81 minutes (P < .001). It accurately measured average resting heart rate and was able to distinguish sleep onset latency with some accuracy. However, the agreement between EMFIT QS and polysomnography on sleep-wake detection was low (kappa = 0.13, P < .001), EMFIT QS failed to distinguish sleep stages. CONCLUSIONS: A consensus between polysomnography and EMFIT QS was found in sleep onset latency and average heart rate. There was significant discrepancy and lack of consensus in other sleep outcomes. These findings indicated that further development is necessary before using EMFIT QS in clinical and research settings. CLINICAL TRIAL REGISTRATION: Registry: Australian New Zealand Clinical Trials Registry; Name: Sleep parameter validation of a consumer home sleep monitoring device, EMFIT Quantified Sleep (QS), against Polysomnography; URL: https://www.anzctr.org.au/ACTRN12621000600842.aspx; Identifier: ACTRN12621000600842. CITATION: Kholghi M, Szollosi I, Hollamby M, Bradford D, Zhang Q. A validation study of a ballistocardiograph sleep tracker against polysomnography. J Clin Sleep Med. 2022;18(4):1203-1210.

Association of Longitudinal Sleep and Next-day Indoor Mobility Measured via Passive Sensors among Community-dwelling Older Adults.

Previous studies have shown there is a relationship between sleep and mobility in older adults by collecting and analysing self-reported data from surveys and questionnaires, or by using objective measures from polysomnography or actigraphy. However, these methods have limitations for long-term monitoring, especially for community-dwelling adults. In this paper, we investigate the association between sleep and indoor mobility using longitudinal data collected over a period of about 12 months for older adults (65 years or older) living at home in Australia. The data was collected objectively and continuously using non-invasive and passive sensors. First, we explored whether sleep and indoor mobility are different across gender and age groups (70s, 80s, and 90s). Second, we investigate the association of sleep and next-day indoor mobility through a stepwise multivariate regression. We found that males and females have significant differences in mobility, time in bed, total time in sleep, number and duration of awakenings and sleep efficiency. Additionally, mobility and all sleep measures significantly vary across the three age groups, except for sleep onset latency between 80s and 90s. Our findings show that sleep efficiency and total sleep time are the key sleep measures affecting next-day mobility, while sleep onset latency has the least effect.Clinical relevance - Our study contributes to a better understanding of the sleep patterns of older adults and how they affect their physical functioning.

The Significance and Limitations of Monitoring Sleep during Pregnancy.

Sleep patterns often change during pregnancy and postpartum. However, if severe and persistent, these changes can depict a risk factor for significant health complications. It is thus essential to identify and understand changes in women's sleeping pattern over the course of pregnancy and postpartum, to offer an appropriate and timely intervention if necessary. In this paper, we discuss sleep disturbances during pregnancy and their association with pregnancy complications. We also review the state-of-the-art digital devices for real-time sleep assessment, and highlight their strengths and limitations.Clinical Relevance-This review highlights an importance of an individualized holistic pregnancy care program which engages both the healthcare professionals and the obstetric population, together with an educational module to increase the user awareness on the importance of sleep disturbances and their consequences during and after pregnancy.

Home-Based Sleep Sensor Measurements in an Older Australian Population: Before and during a Pandemic.

Older adults are susceptible to poor night-time sleep, characterized by short sleep duration and high sleep disruptions (i.e., more frequent and longer awakenings). This study aimed to longitudinally and objectively assess the changes in sleep patterns of older Australians during the 2020 pandemic lockdown. A non-invasive mattress-based device, known as the EMFIT QS, was used to continuously monitor sleep in 31 older adults with an average age of 84 years old before (November 2019-February 2020) and during (March-May 2020) the COVID-19, a disease caused by a form of coronavirus, lockdown. Total sleep time, sleep onset latency, wake after sleep onset, sleep efficiency, time to bed, and time out of bed were measured across these two periods. Overall, there was no significant change in total sleep time; however, women had a significant increase in total sleep time (36 min), with a more than 30-min earlier bedtime. There was also no increase in wake after sleep onset and sleep onset latency. Sleep efficiency remained stable across the pandemic time course between 84-85%. While this sample size is small, these data provide reassurance that objective sleep measurement did not deteriorate through the pandemic in older community-dwelling Australians.

A prospective cohort study of prodromal Alzheimer's disease: Prospective Imaging Study of Ageing: Genes, Brain and Behaviour (PISA).

This prospective cohort study, "Prospective Imaging Study of Ageing: Genes, Brain and Behaviour" (PISA) seeks to characterise the phenotype and natural history of healthy adult Australians at high future risk of Alzheimer's disease (AD). In particular, we are recruiting midlife and older Australians with high and low genetic risk of dementia to discover biological markers of early neuropathology, identify modifiable risk factors, and establish the very earliest phenotypic and neuronal signs of disease onset. PISA utilises genetic prediction to recruit and enrich a prospective cohort and follow them longitudinally. Online surveys and cognitive testing are used to characterise an Australia-wide sample currently totalling over 3800 participants. Participants from a defined at-risk cohort and positive controls (clinical cohort of patients with mild cognitive impairment or early AD) are invited for onsite visits for detailed functional, structural and molecular neuroimaging, lifestyle monitoring, detailed neurocognitive testing, plus blood sample donation. This paper describes recruitment of the PISA cohort, study methodology and baseline demographics.

Clinical Document Classification Using Labeled and Unlabeled Data Across Hospitals.

Reviewing radiology reports in emergency departments is an essential but laborious task. Timely follow-up of patients with abnormal cases in their radiology reports may dramatically affect the patient's outcome, especially if they have been discharged with a different initial diagnosis. Machine learning approaches have been devised to expedite the process and detect the cases that demand instant follow up. However, these approaches require a large amount of labeled data to train reliable predictive models. Preparing such a large dataset, which needs to be manually annotated by health professionals, is costly and time-consuming. This paper investigates a semi-supervised transfer learning framework for radiology report classification across three hospitals. The main goal is to leverage both vastly available clinical unlabeled data and already learned knowledge in order to improve a learning model where limited labeled data is available. Our experimental findings show that (1) convolutional neural networks (CNNs), while being independent of any problem-specific feature engineering, achieve significantly higher effectiveness compared to conventional supervised learning approaches, (2) leveraging unlabeled data in training a CNN-based classifier reduces the dependency on labeled data by more than 50% to reach the same performance of a fully supervised CNN, and (3) transferring the knowledge gained from available labeled data in an external source hospital significantly improves the performance of a semi-supervised CNN model over their fully supervised counterparts in a target hospital.

Active learning reduces annotation time for clinical concept extraction.

OBJECTIVE: To investigate: (1) the annotation time savings by various active learning query strategies compared to supervised learning and a random sampling baseline, and (2) the benefits of active learning-assisted pre-annotations in accelerating the manual annotation process compared to de novo annotation. MATERIALS AND METHODS: There are 73 and 120 discharge summary reports provided by Beth Israel institute in the train and test sets of the concept extraction task in the i2b2/VA 2010 challenge, respectively. The 73 reports were used in user study experiments for manual annotation. First, all sequences within the 73 reports were manually annotated from scratch. Next, active learning models were built to generate pre-annotations for the sequences selected by a query strategy. The annotation/reviewing time per sequence was recorded. The 120 test reports were used to measure the effectiveness of the active learning models. RESULTS: When annotating from scratch, active learning reduced the annotation time up to 35% and 28% compared to a fully supervised approach and a random sampling baseline, respectively. Reviewing active learning-assisted pre-annotations resulted in 20% further reduction of the annotation time when compared to de novo annotation. DISCUSSION: The number of concepts that require manual annotation is a good indicator of the annotation time for various active learning approaches as demonstrated by high correlation between time rate and concept annotation rate. CONCLUSION: Active learning has a key role in reducing the time required to manually annotate domain concepts from clinical free text, either when annotating from scratch or reviewing active learning-assisted pre-annotations.

Active learning: a step towards automating medical concept extraction.

OBJECTIVE: This paper presents an automatic, active learning-based system for the extraction of medical concepts from clinical free-text reports. Specifically, (1) the contribution of active learning in reducing the annotation effort and (2) the robustness of incremental active learning framework across different selection criteria and data sets are determined. MATERIALS AND METHODS: The comparative performance of an active learning framework and a fully supervised approach were investigated to study how active learning reduces the annotation effort while achieving the same effectiveness as a supervised approach. Conditional random fields as the supervised method, and least confidence and information density as 2 selection criteria for active learning framework were used. The effect of incremental learning vs standard learning on the robustness of the models within the active learning framework with different selection criteria was also investigated. The following 2 clinical data sets were used for evaluation: the Informatics for Integrating Biology and the Bedside/Veteran Affairs (i2b2/VA) 2010 natural language processing challenge and the Shared Annotated Resources/Conference and Labs of the Evaluation Forum (ShARe/CLEF) 2013 eHealth Evaluation Lab. RESULTS: The annotation effort saved by active learning to achieve the same effectiveness as supervised learning is up to 77%, 57%, and 46% of the total number of sequences, tokens, and concepts, respectively. Compared with the random sampling baseline, the saving is at least doubled. CONCLUSION: Incremental active learning is a promising approach for building effective and robust medical concept extraction models while significantly reducing the burden of manual annotation.