A Descriptive Comparative Pilot Study: Association Between Use of a Self-monitoring Device and Sleep and Stress Outcomes in Pregnancy.

Pregnancy is a challenging time for maintaining quality sleep and managing stress. Digital self-monitoring technologies are popular because of assumed increased patient engagement leading to an impact on health outcomes. However, the actual association between wear time of such devices and improved sleep/stress outcomes remains untested. Here, a descriptive comparative pilot study of 20 pregnant women was conducted to examine associations between wear time (behavioral engagement) of self-monitoring devices and sleep/stress pregnancy outcomes. Women used a ring fitted to their finger to monitor sleep/stress data, with access to a self-monitoring program for an average of 9½ weeks. Based on wear time, participants were split into two engagement groups. Using a linear mixed-effects model, the high engagement group showed higher levels of stress and a negative trend in sleep duration and quality. The low engagement group showed positive changes in sleep duration, and quality and experienced below-normal sleep onset latency at the start of the pilot but trended toward normal levels. Engagement according to device wear time was not associated with improved outcomes. Further research should aim to understand how engagement with self-monitoring technologies impacts sleep/stress outcomes in pregnancy.

Accuracy Assessment of Oura Ring Nocturnal Heart Rate and Heart Rate Variability in Comparison With Electrocardiography in Time and Frequency Domains: Comprehensive Analysis.

BACKGROUND: Photoplethysmography is a noninvasive and low-cost method to remotely and continuously track vital signs. The Oura Ring is a compact photoplethysmography-based smart ring, which has recently drawn attention to remote health monitoring and wellness applications. The ring is used to acquire nocturnal heart rate (HR) and HR variability (HRV) parameters ubiquitously. However, these parameters are highly susceptible to motion artifacts and environmental noise. Therefore, a validity assessment of the parameters is required in everyday settings. OBJECTIVE: This study aims to evaluate the accuracy of HR and time domain and frequency domain HRV parameters collected by the Oura Ring against a medical grade chest electrocardiogram monitor. METHODS: We conducted overnight home-based monitoring using an Oura Ring and a Shimmer3 electrocardiogram device. The nocturnal HR and HRV parameters of 35 healthy individuals were collected and assessed. We evaluated the parameters within 2 tests, that is, values collected from 5-minute recordings (ie, short-term HRV analysis) and the average values per night sleep. A linear regression method, the Pearson correlation coefficient, and the Bland-Altman plot were used to compare the measurements of the 2 devices. RESULTS: Our findings showed low mean biases of the HR and HRV parameters collected by the Oura Ring in both the 5-minute and average-per-night tests. In the 5-minute test, the error variances of the parameters were different. The parameters provided by the Oura Ring dashboard (ie, HR and root mean square of successive differences [RMSSD]) showed relatively low error variance compared with the HRV parameters extracted from the normal interbeat interval signals. The Pearson correlation coefficient tests (P<.001) indicated that HR, RMSSD, average of normal heart beat intervals (AVNN), and percentage of successive normal beat-to-beat intervals that differ by more than 50 ms (pNN50) had high positive correlations with the baseline values; SD of normal beat-to-beat intervals (SDNN) and high frequency (HF) had moderate positive correlations, and low frequency (LF) and LF:HF ratio had low positive correlations. The HR, RMSSD, AVNN, and pNN50 had narrow 95% CIs; however, SDNN, LF, HF, and LF:HF ratio had relatively wider 95% CIs. In contrast, the average-per-night test showed that the HR, RMSSD, SDNN, AVNN, pNN50, LF, and HF had high positive relationships (P<.001), and the LF:HF ratio had a moderate positive relationship (P<.001). The average-per-night test also indicated considerably lower error variances than the 5-minute test for the parameters. CONCLUSIONS: The Oura Ring could accurately measure nocturnal HR and RMSSD in both the 5-minute and average-per-night tests. It provided acceptable nocturnal AVNN, pNN50, HF, and SDNN accuracy in the average-per-night test but not in the 5-minute test. In contrast, the LF and LF:HF ratio of the ring had high error rates in both tests.

Robust PPG Peak Detection Using Dilated Convolutional Neural Networks.

Accurate peak determination from noise-corrupted photoplethysmogram (PPG) signal is the basis for further analysis of physiological quantities such as heart rate. Conventional methods are designed for noise-free PPG signals and are insufficient for PPG signals with low signal-to-noise ratio (SNR). This paper focuses on enhancing PPG noise-resiliency and proposes a robust peak detection algorithm for PPG signals distorted due to noise and motion artifact. Our algorithm is based on convolutional neural networks (CNNs) with dilated convolutions. We train and evaluate the proposed method using a dataset collected via smartwatches under free-living conditions in a home-based health monitoring application. A data generator is also developed to produce noisy PPG data used for model training and evaluation. The method performance is compared against other state-of-the-art methods and is tested with SNRs ranging from 0 to 45 dB. Our method outperforms the existing adaptive threshold, transform-based, and machine learning methods. The proposed method shows overall precision, recall, and F1-score of 82%, 80%, and 81% in all the SNR ranges. In contrast, the best results obtained by the existing methods are 78%, 80%, and 79%. The proposed method proves to be accurate for detecting PPG peaks even in the presence of noise.

The other COVID-19 survivors: Timing, duration, and health impact of post-acute sequelae of SARS-CoV-2 infection.

AIMS AND OBJECTIVES: To determine the frequency, timing, and duration of post-acute sequelae of SARS-CoV-2 infection (PASC) and their impact on health and function. BACKGROUND: Post-acute sequelae of SARS-CoV-2 infection is an emerging major public health problem that is poorly understood and has no current treatment or cure. PASC is a new syndrome that has yet to be fully clinically characterised. DESIGN: Descriptive cross-sectional survey (n = 5163) was conducted from online COVID-19 survivor support groups who reported symptoms for more than 21 days following SARS-CoV-2 infection. METHODS: Participants reported background demographics and the date and method of their covid diagnosis, as well as all symptoms experienced since onset of covid in terms of the symptom start date, duration, and Likert scales measuring three symptom-specific health impacts: pain and discomfort, work impairment, and social impairment. Descriptive statistics and measures of central tendencies were computed for participant demographics and symptom data. RESULTS: Participants reported experiencing a mean of 21 symptoms (range 1-93); fatigue (79.0%), headache (55.3%), shortness of breath (55.3%) and difficulty concentrating (53.6%) were the most common. Symptoms often remitted and relapsed for extended periods of time (duration M = 112 days), longest lasting symptoms included the inability to exercise (M = 106.5 days), fatigue (M = 101.7 days) and difficulty concentrating, associated with memory impairment (M = 101.1 days). Participants reported extreme pressure at the base of the head, syncope, sharp or sudden chest pain, and "brain pressure" among the most distressing and impacting daily life. CONCLUSIONS: Post-acute sequelae of SARS-CoV-2 infection can be characterised by a wide range of symptoms, many of which cause moderate-to-severe distress and can hinder survivors' overall well-being. RELEVANCE TO CLINICAL PRACTICE: This study advances our understanding of the symptoms of PASC and their health impacts.

Comparison of Oura Smart Ring Against ActiGraph Accelerometer for Measurement of Physical Activity and Sedentary Time in a Free-Living Context.

Smart rings, such as the Oura ring, might have potential in health monitoring. To be able to identify optimal devices for healthcare settings, validity studies are needed. The aim of this study was to compare the Oura smart ring estimates of steps and sedentary time with data from the ActiGraph accelerometer in a free-living context. A cross-sectional observational study design was used. A convenience sample of healthy adults (n = 42) participated in the study and wore an Oura smart ring and an ActiGraph accelerometer on the non-dominant hand continuously for 1 week. The participants completed a background questionnaire and filled out a daily log about their sleeping times and times when they did not wear the devices. The median age of the participants (n = 42) was 32 years (range, 18-46 years). In total, 191 (61% of the potential) days were compared. The Oura ring overestimated the step counts compared with the ActiGraph. The mean difference was 1416 steps (95% confidence interval, 739-2093 steps). Daily sedentary time was also overestimated by the ring; the mean difference was 17 minutes (95% confidence interval, -2 to 37 minutes). The use of the ring in nursing interventions needs to be considered.

Antihistamines for Postacute Sequelae of SARS-CoV-2 Infection.

Postacute sequelae of SARS-CoV2 (PASC) infection is an emerging global health crisis, variably affecting millions worldwide. PASC has no established treatment. We describe 2 cases of PASC in response to opportune administration of over-the-counter antihistamines, with significant improvement in symptoms and ability to perform activities of daily living. Future studies are warranted to understand the potential role of histamine in the pathogenesis of PASC and explore the clinical benefits of antihistamines in the treatment of PASC.

Identifying target behaviors for weight management interventions for women who are overweight during pregnancy and the postpartum period: a qualitative study informed by the Behaviour Change Wheel.

BACKGROUND: Maternal overweight is increasing, and it is associated with several risk factors for both the mother and child. Healthy lifestyle behaviors adopted during pregnancy are likely to impact women's health positively after pregnancy. The study's aim was to identify and describe weight management behaviors in terms of the Capability, Opportunity and Motivation Behaviour (COM-B) -model to target weight management interventions from both the perspectives of women who are overweight and maternity care professionals. METHODS: This qualitative, descriptive study was conducted between 2019 and 2020. Individual interviews with pregnant and postpartum women who were overweight (n = 11) and focus group interviews with public health nurses (n = 5) were undertaken in two public maternity clinics in Southwest Finland. The data were analyzed using deductive content analysis consistent with the COM-B model. RESULTS: In the capability category, the women and the public health nurses thought that there was a need to find consistent ways to approach overweight, as it had often become a feature of the women's identities. The use of health technology was considered to be an element of antenatal care that could be used to approach the subject of weight and weight management. Smart wearables could also support an evaluation of the women's lifestyles. The opportunity category highlighted the lack of resources for support during perinatal care, especially after birth. Both groups felt that support from the family was the most important facilitating factor besides motivation. The women also expressed a conflict between pregnancy as an excuse to engage in unhealthy habits and pregnancy as a motivational period for a change of lifestyle. Furthermore, the women wanted to be offered a more robust stance on weight management and discreet counseling. CONCLUSIONS: Our findings offer a theoretical basis on which future research can define intervention and implementation strategies. Such interventions may offer clear advice and non-judgmental support during pregnancy and after delivery by targeting women's capabilities, opportunities, and motivation. Health technology could be a valuable component of intervention, as well as an implementation strategy, as they provide ways during maternity care to approach this topic and support women.

Pain Recognition With Electrocardiographic Features in Postoperative Patients: Method Validation Study.

BACKGROUND: There is a strong demand for an accurate and objective means of assessing acute pain among hospitalized patients to help clinicians provide pain medications at a proper dosage and in a timely manner. Heart rate variability (HRV) comprises changes in the time intervals between consecutive heartbeats, which can be measured through acquisition and interpretation of electrocardiography (ECG) captured from bedside monitors or wearable devices. As increased sympathetic activity affects the HRV, an index of autonomic regulation of heart rate, ultra-short-term HRV analysis can provide a reliable source of information for acute pain monitoring. In this study, widely used HRV time and frequency domain measurements are used in acute pain assessments among postoperative patients. The existing approaches have only focused on stimulated pain in healthy subjects, whereas, to the best of our knowledge, there is no work in the literature building models using real pain data and on postoperative patients. OBJECTIVE: The objective of our study was to develop and evaluate an automatic and adaptable pain assessment algorithm based on ECG features for assessing acute pain in postoperative patients likely experiencing mild to moderate pain. METHODS: The study used a prospective observational design. The sample consisted of 25 patient participants aged 18 to 65 years. In part 1 of the study, a transcutaneous electrical nerve stimulation unit was employed to obtain baseline discomfort thresholds for the patients. In part 2, a multichannel biosignal acquisition device was used as patients were engaging in non-noxious activities. At all times, pain intensity was measured using patient self-reports based on the Numerical Rating Scale. A weak supervision framework was inherited for rapid training data creation. The collected labels were then transformed from 11 intensity levels to 5 intensity levels. Prediction models were developed using 5 different machine learning methods. Mean prediction accuracy was calculated using leave-one-out cross-validation. We compared the performance of these models with the results from a previously published research study. RESULTS: Five different machine learning algorithms were applied to perform a binary classification of baseline (BL) versus 4 distinct pain levels (PL1 through PL4). The highest validation accuracy using 3 time domain HRV features from a BioVid research paper for baseline versus any other pain level was achieved by support vector machine (SVM) with 62.72% (BL vs PL4) to 84.14% (BL vs PL2). Similar results were achieved for the top 8 features based on the Gini index using the SVM method, with an accuracy ranging from 63.86% (BL vs PL4) to 84.79% (BL vs PL2). CONCLUSIONS: We propose a novel pain assessment method for postoperative patients using ECG signal. Weak supervision applied for labeling and feature extraction improves the robustness of the approach. Our results show the viability of using a machine learning algorithm to accurately and objectively assess acute pain among hospitalized patients. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/17783.

Long-Term IoT-Based Maternal Monitoring: System Design and Evaluation.

Pregnancy is a unique time when many mothers gain awareness of their lifestyle and its impacts on the fetus. High-quality care during pregnancy is needed to identify possible complications early and ensure the mother's and her unborn baby's health and well-being. Different studies have thus far proposed maternal health monitoring systems. However, they are designed for a specific health problem or are limited to questionnaires and short-term data collection methods. Moreover, the requirements and challenges have not been evaluated in long-term studies. Maternal health necessitates a comprehensive framework enabling continuous monitoring of pregnant women. In this paper, we present an Internet-of-Things (IoT)-based system to provide ubiquitous maternal health monitoring during pregnancy and postpartum. The system consists of various data collectors to track the mother's condition, including stress, sleep, and physical activity. We carried out the full system implementation and conducted a real human subject study on pregnant women in Southwestern Finland. We then evaluated the system's feasibility, energy efficiency, and data reliability. Our results show that the implemented system is feasible in terms of system usage during nine months. We also indicate the smartwatch, used in our study, has acceptable energy efficiency in long-term monitoring and is able to collect reliable photoplethysmography data. Finally, we discuss the integration of the presented system with the current healthcare system.

An Efficient and Robust Deep Learning Method with 1-D Octave Convolution to Extract Fetal Electrocardiogram.

The invasive method of fetal electrocardiogram (fECG) monitoring is widely used with electrodes directly attached to the fetal scalp. There are potential risks such as infection and, thus, it is usually carried out during labor in rare cases. Recent advances in electronics and technologies have enabled fECG monitoring from the early stages of pregnancy through fECG extraction from the combined fetal/maternal ECG (f/mECG) signal recorded non-invasively in the abdominal area of the mother. However, cumbersome algorithms that require the reference maternal ECG as well as heavy feature crafting makes out-of-clinics fECG monitoring in daily life not yet feasible. To address these challenges, we proposed a pure end-to-end deep learning model to detect fetal QRS complexes (i.e., the main spikes observed on a fetal ECG waveform). Additionally, the model has the residual network (ResNet) architecture that adopts the novel 1-D octave convolution (OctConv) for learning multiple temporal frequency features, which in turn reduce memory and computational cost. Importantly, the model is capable of highlighting the contribution of regions that are more prominent for the detection. To evaluate our approach, data from the PhysioNet 2013 Challenge with labeled QRS complex annotations were used in the original form, and the data were then modified with Gaussian and motion noise, mimicking real-world scenarios. The model can achieve a F1 score of 91.1% while being able to save more than 50% computing cost with less than 2% performance degradation, demonstrating the effectiveness of our method.

Investigation of Machine Learning Approaches for Traumatic Brain Injury Classification via EEG Assessment in Mice.

Due to the difficulties and complications in the quantitative assessment of traumatic brain injury (TBI) and its increasing relevance in today's world, robust detection of TBI has become more significant than ever. In this work, we investigate several machine learning approaches to assess their performance in classifying electroencephalogram (EEG) data of TBI in a mouse model. Algorithms such as decision trees (DT), random forest (RF), neural network (NN), support vector machine (SVM), K-nearest neighbors (KNN) and convolutional neural network (CNN) were analyzed based on their performance to classify mild TBI (mTBI) data from those of the control group in wake stages for different epoch lengths. Average power in different frequency sub-bands and alpha:theta power ratio in EEG were used as input features for machine learning approaches. Results in this mouse model were promising, suggesting similar approaches may be applicable to detect TBI in humans in practical scenarios.

Acute pain intensity monitoring with the classification of multiple physiological parameters.

Current acute pain intensity assessment tools are mainly based on self-reporting by patients, which is impractical for non-communicative, sedated or critically ill patients. In previous studies, various physiological signals have been observed qualitatively as a potential pain intensity index. On the basis of that, this study aims at developing a continuous pain monitoring method with the classification of multiple physiological parameters. Heart rate (HR), breath rate (BR), galvanic skin response (GSR) and facial surface electromyogram were collected from 30 healthy volunteers under thermal and electrical pain stimuli. The collected samples were labelled as no pain, mild pain or moderate/severe pain based on a self-reported visual analogue scale. The patterns of these three classes were first observed from the distribution of the 13 processed physiological parameters. Then, artificial neural network classifiers were trained, validated and tested with the physiological parameters. The average classification accuracy was 70.6%. The same method was applied to the medians of each class in each test and accuracy was improved to 83.3%. With facial electromyogram, the adaptivity of this method to a new subject was improved as the recognition accuracy of moderate/severe pain in leave-one-subject-out cross-validation was promoted from 74.9 ± 21.0 to 76.3 ± 18.1%. Among healthy volunteers, GSR, HR and BR were better correlated to pain intensity variations than facial muscle activities. The classification of multiple accessible physiological parameters can potentially provide a way to differentiate among no, mild and moderate/severe acute experimental pain.

Colloidal particle adsorption at liquid interfaces: capillary driven dynamics and thermally activated kinetics.

The adsorption of single colloidal microparticles (0.5-1 µm radius) at a water-oil interface has been recently studied experimentally using digital holographic microscopy [Kaz et al., Nat. Mater., 2012, 11, 138-142]. An initially fast adsorption dynamics driven by capillary forces is followed by an unexpectedly slow relaxation to equilibrium that is logarithmic in time and can span hours or days. The slow relaxation kinetics has been attributed to the presence of surface "defects" with nanoscale dimensions (1-5 nm) that induce multiple metastable configurations of the contact line perimeter. A kinetic model considering thermally activated transitions between such metastable configurations has been proposed [Colosqui et al., Phys. Rev. Lett., 2013, 111, 028302] to predict both the relaxation rate and the crossover point to the slow logarithmic regime. However, the adsorption dynamics observed experimentally before the crossover point has remained unstudied. In this work, we propose a Langevin model that is able to describe the entire adsorption process of single colloidal particles by considering metastable states produced by surface defects and thermal motion of the particle and liquid interface. Invoking the fluctuation dissipation theorem, we introduce a drag term that considers significant dissipative forces induced by thermal fluctuations of the liquid interface. Langevin dynamics simulations based on the proposed adsorption model yield close agreement with experimental observations for different microparticles, capturing the crossover from (fast) capillary driven dynamics to (slow) thermally activated kinetics.

Wetting Driven by Thermal Fluctuations on Terraced Nanostructures.

Theoretical analysis and fully atomistic molecular dynamics simulations reveal a Brownian ratchet mechanism by which thermal fluctuations drive the net displacement of immiscible liquids confined in channels or pores with micro- or nanoscale dimensions. The thermally driven displacement is induced by surface nanostructures with directional asymmetry and can occur against the direction of action of wetting or capillary forces. Mean displacement rates in molecular dynamics simulations are predicted via analytical solution of a Smoluchowski diffusion equation for the position probability density. The proposed physical mechanisms and derived analytical expressions can be applied to engineer surface nanostructures for controlling the dynamics of diverse wetting processes such as capillary filling, wicking, and imbibition in micro- or nanoscale systems.

Feasibility of continuous smart health monitoring in pregnant population: A mixed-method approach.

The utilization of smart monitoring technology offers potential for enhancing health outcomes, yet its feasibility and acceptance among Hispanic pregnant individuals remain uncertain. This is particularly crucial to investigate within the context of apparently healthy individuals identified as low risk, who still face a 10% likelihood of complications. Given their frequent underrepresentation in healthcare services and relative lack of attention, improving the feasibility of remote monitoring in this population could yield significant benefits. To address this gap, our study aimed to adapt and evaluate the practicality of a smart monitoring platform among healthy Hispanic pregnant women during the second and third trimesters of pregnancy, as well as one week following childbirth, a period when complications often arise. This longitudinal study followed n = 16 participants for an average of 17 weeks. Participants were instructed to wear the Oura ring for objective data collection, including activity, sleep, and heart rate, and to complete survey questions through REDcap to assess mental health and lifestyle factors. The study framework utilized the RE-AIM approach, with acceptability and adherence as key components of the feasibility evaluation. Our findings revealed that completion rates for biweekly and monthly surveys remained consistently high until after childbirth (approximately 80%), while daily question completion remained above 80% until 38th week of gestation, declining thereafter. The wearing rate of the Oura ring remained consistently above 80% until the 35th gestational week, decreasing to around 31% postpartum. Participants cited barriers to wearing the ring during the postpartum period, including difficulties managing the newborn, forgetfulness, and concerns about scratching the baby's skin. The enrollment rate was 71.42%, with an attrition rate of 6.25%. Thematic analysis of one-on-one interviews identified three main themes: personal desire for health improvement, social acceptability and support, and conditions influencing device/platform efficiency. In conclusion, while adherence varied based on gestational week and survey frequency, the study demonstrated strong acceptability of the smart monitoring platform among the study population, indicated by the high enrollment rate. Qualitative insights underscored the significance of personal motivation, social support, and device/platform efficiency in enhancing patient engagement with digital health monitoring during pregnancy, offering valuable considerations for future healthcare interventions in this domain.

Objective monitoring of loneliness levels using smart devices: A multi-device approach for mental health applications.

Loneliness is linked to wide ranging physical and mental health problems, including increased rates of mortality. Understanding how loneliness manifests is important for targeted public health treatment and intervention. With advances in mobile sending and wearable technologies, it is possible to collect data on human phenomena in a continuous and uninterrupted way. In doing so, such approaches can be used to monitor physiological and behavioral aspects relevant to an individual's loneliness. In this study, we proposed a method for continuous detection of loneliness using fully objective data from smart devices and passive mobile sensing. We also investigated whether physiological and behavioral features differed in their importance in predicting loneliness across individuals. Finally, we examined how informative data from each device is for loneliness detection tasks. We assessed subjective feelings of loneliness while monitoring behavioral and physiological patterns in 30 college students over a 2-month period. We used smartphones to monitor behavioral patterns (e.g., location changes, type of notifications, in-coming and out-going calls/text messages) and smart watches and rings to monitor physiology and sleep patterns (e.g., heart-rate, heart-rate variability, sleep duration). Participants reported their loneliness feeling multiple times a day through a questionnaire app on their phone. Using the data collected from their devices, we trained a random forest machine learning based model to detect loneliness levels. We found support for loneliness prediction using a multi-device and fully-objective approach. Furthermore, behavioral data collected by smartphones generally were the most important features across all participants. The study provides promising results for using objective data to monitor mental health indicators, which could provide a continuous and uninterrupted source of information in mental healthcare applications.

The effect of SARS-COV-2 variant on non-respiratory features and mortality among vaccinated and non-fully vaccinated patients.

OBJECTIVE: To determine the effect of SARS-CoV-2 variants on non-respiratory features of COVID-19 in vaccinated and not fully vaccinated patients using a University of California database. METHODS: A longitudinal retrospective review of medical records (n = 63,454) from 1/1/2020-4/26/2022 using the UCCORDS database was performed to compare non-respiratory features, vaccination status, and mortality between variants. Chi-square tests were used to study the relationship between categorical variables using a contingency matrix. RESULTS: Fever was the most common feature across all variants. Fever was significantly higher in not fully vaccinated during the Delta and Omicron waves (p = 0.001; p = 0.001). Cardiac features were statistically higher in not fully vaccinated during Omicron; tachycardia was only a feature of not fully vaccinated during Delta and Omicron; diabetes and GI reflux were features of all variants regardless of vaccine status. Odds of death were significantly increased among those not fully vaccinated in the Delta and Omicron variants (Delta OR: 1.64, p = 0.052; Omicron OR: 1.96, p < 0.01). Vaccination was associated with a decrease in the frequency of non-respiratory features. CONCLUSIONS: Risk of non-respiratory features of COVID-19 is statistically higher in those not fully vaccinated across all variants. Risk of death and correlation with vaccination status varied.

Foundation metrics for evaluating effectiveness of healthcare conversations powered by generative AI.

Generative Artificial Intelligence is set to revolutionize healthcare delivery by transforming traditional patient care into a more personalized, efficient, and proactive process. Chatbots, serving as interactive conversational models, will probably drive this patient-centered transformation in healthcare. Through the provision of various services, including diagnosis, personalized lifestyle recommendations, dynamic scheduling of follow-ups, and mental health support, the objective is to substantially augment patient health outcomes, all the while mitigating the workload burden on healthcare providers. The life-critical nature of healthcare applications necessitates establishing a unified and comprehensive set of evaluation metrics for conversational models. Existing evaluation metrics proposed for various generic large language models (LLMs) demonstrate a lack of comprehension regarding medical and health concepts and their significance in promoting patients' well-being. Moreover, these metrics neglect pivotal user-centered aspects, including trust-building, ethics, personalization, empathy, user comprehension, and emotional support. The purpose of this paper is to explore state-of-the-art LLM-based evaluation metrics that are specifically applicable to the assessment of interactive conversational models in healthcare. Subsequently, we present a comprehensive set of evaluation metrics designed to thoroughly assess the performance of healthcare chatbots from an end-user perspective. These metrics encompass an evaluation of language processing abilities, impact on real-world clinical tasks, and effectiveness in user-interactive conversations. Finally, we engage in a discussion concerning the challenges associated with defining and implementing these metrics, with particular emphasis on confounding factors such as the target audience, evaluation methods, and prompt techniques involved in the evaluation process.

Can Sleep Quality Attributes be Predicted from Physical Activity in Everyday Settings?

Sleep is crucial for physical, mental, and emotional well-being. Physical activity and sleep are known to be interrelated; however, limited research has been performed to investigate their interactions in long-term. Conventional studies have presented sleep quality prediction, focusing on a single sleep quality aspect, such as sleep efficiency. In addition, the relationship between daily physical activity and sleep quality has yet to be explored, despite physical activities being utilized in previous studies for sleep quality prediction. In this paper, we develop an Extreme Gradient boosting method to predict sleep duration, sleep efficiency, and deep sleep based on users' daily activity information collected from wearable devices. Our model is trained and tested using data collected with an OURA ring from 34 pregnant mothers for six months under free-living conditions. Our finding shows an accuracy of 90.58%, 95.38%, and 91.45% for sleep duration, efficiency, and deep sleep, respectively. Moreover, we assess the contribution of each physical activity parameter to the prediction results using the Shapley Additive Explanations method. Our results indicate that sedentary time is the most influential parameter for sleep duration prediction, while the inactive time feature (e.g., resting or lying down) has a strong negative relationship with sleep efficiency, and the pregnancy week is the most critical parameter for deep sleep prediction.

Heart Rate Variability and Pregnancy Complications: Systematic Review.

BACKGROUND: The autonomic nervous system (ANS) is known as a critical regulatory system for pregnancy-induced adaptations. If it fails to function, life-threatening pregnancy complications could occur. Hence, understanding and monitoring the underlying mechanism of action for these complications are necessary. OBJECTIVE: We aimed to systematically review the literature concerned with the associations between heart rate variability (HRV), as an ANS biomarker, and pregnancy complications. METHODS: We performed a comprehensive search in the PubMed, Medline Completion, CINAHL Completion, Web of Science Core Collection Classic, Cochrane Library, and SCOPUS databases in February 2022 with no time span limitation. We included studies concerned with the association between any pregnancy complications and HRV, with or without a control group. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guideline was used for the review of the studies, and Covidence software was used for the study selection process. For data synthesis, we used the guideline by Popay et al. RESULTS: Finally, 12 studies with 6656 participants were included. Despite the methodological divergency that hindered a comprehensive comparison, our findings suggest that ANS is linked with some common pregnancy complications including fetal growth. However, existing studies do not support an association between ANS and gestational diabetes mellitus. Studies that linked pulmonary and central nervous system disorders with ANS function did not provide enough evidence to draw conclusions. CONCLUSIONS: This review highlights the importance of understanding and monitoring the underlying mechanism of ANS in pregnancy-induced adaptations and the need for further research with robust methodology in this area.