Predictive performance and impact of algorithms in remote monitoring of chronic conditions: A systematic review and meta-analysis

Computational Decision Support for PE Diagnosis based on Ventilation Perfusion Ratio

AIM

The aim of this study is to investigate whether computer-aided, semi-automated 3D lung lobe quantification can support decision-making on PE diagnosis based on the ventilation-perfusion ratio in clinical practice.

METHODS

A study cohort of 100 patients (39 male, 61 female, age 64.8±15.8 years) underwent ventilation/perfusion single photon emission computed tomography (V/Q-SPECT/CT) to exclude acute PE on SPECT/CT OPTIMA NM/CT 640 (GE Healthcare). Two 3D lung lobe quantification software tools (Q. Lung: Xeleris 4.0, GE Healthcare and LLQ: Hermes Hybrid 3D Lung Lobar Quantification, Hermes Medical Solutions) were used to evaluate the numerical lobar ventilation/perfusion ratio (VQR) and lobar volume/perfusion ratio (VPR). A test of linearity and equivalence of the two 3D software tools was performed using Pearson, Spearman, quadratic weighted kappa and the mean squared deviation for VPR/VQR. An algorithm was developed that identified PE candidates using ROC analysis. The agreement between the PE findings of an experienced nuclear medicine expert and the calculated PE candidates was represented by the magnitude of the YOUDEN index (J) and the size of the area under the receiver operating curve (AUC).

RESULTS

Both 3D software tools showed good comparability. The YOUDEN index for QLUNG(VPR/VQR)/LLQ(VPR/VQR) was in the range from 0.2 to 0.5. The mean AUC averaged over all lung lobes for QLUNG(VPR) was 0.66, CI95%: ±14.0%, for QLUNG(VQR) 0.66, CI95%: ±13.3%, for LLQ(VPR) 0.64, CI95%: ±14.7% and for LLQ(VQR) 0.65, CI95%: ±13.1%.

CONCLUSION

This study reveals that 3D software tools are feasible for numerical PE detection. The clinical decision can be supported by using a numerical algorithm based on ROC analysis.

Feasibility of Observing Cerebrovascular Disease Phenotypes with Smartphone Monitoring: Study Design Considerations for Real-World Studies

Accelerated by the adoption of remote monitoring during the COVID-19 pandemic, interest in using digitally captured behavioral data to predict patient outcomes has grown; however, it is unclear how feasible digital phenotyping studies may be in patients with recent ischemic stroke or transient ischemic attack. In this perspective, we present participant feedback and relevant smartphone data metrics suggesting that digital phenotyping of post-stroke depression is feasible. Additionally, we proffer thoughtful considerations for designing feasible real-world study protocols tracking cerebrovascular dysfunction with smartphone sensors.

ECG-based data-driven solutions for diagnosis and prognosis of cardiovascular diseases: A systematic review

Cardiovascular diseases (CVD) are a leading cause of death globally, and result in significant morbidity and reduced quality of life. The electrocardiogram (ECG) plays a crucial role in CVD diagnosis, prognosis, and prevention; however, different challenges still remain, such as an increasing unmet demand for skilled cardiologists capable of accurately interpreting ECG. This leads to higher workload and potential diagnostic inaccuracies. Data-driven approaches, such as machine learning (ML) and deep learning (DL) have emerged to improve existing computer-assisted solutions and enhance physicians' ECG interpretation of the complex mechanisms underlying CVD. However, many ML and DL models used to detect ECG-based CVD suffer from a lack of explainability, bias, as well as ethical, legal, and societal implications (ELSI). Despite the critical importance of these Trustworthy Artificial Intelligence (AI) aspects, there is a lack of comprehensive literature reviews that examine the current trends in ECG-based solutions for CVD diagnosis or prognosis that use ML and DL models and address the Trustworthy AI requirements. This review aims to bridge this knowledge gap by providing a systematic review to undertake a holistic analysis across multiple dimensions of these data-driven models such as type of CVD addressed, dataset characteristics, data input modalities, ML and DL algorithms (with a focus on DL), and aspects of Trustworthy AI like explainability, bias and ethical considerations. Additionally, within the analyzed dimensions, various challenges are identified. To these, we provide concrete recommendations, equipping other researchers with valuable insights to understand the current state of the field comprehensively.

Non-Invasive Wearable Devices for Monitoring Vital Signs in Patients with Type 2 Diabetes Mellitus: A Systematic Review

Type 2 diabetes mellitus (T2D) poses a significant global health challenge and demands effective self-management strategies, including continuous blood glucose monitoring (CGM) and lifestyle adaptations. While CGM offers real-time glucose level assessment, the quest for minimizing trauma and enhancing convenience has spurred the need to explore non-invasive alternatives for monitoring vital signs in patients with T2D. Objective: This systematic review is the first that explores the current literature and critically evaluates the use and reporting of non-invasive wearable devices for monitoring vital signs in patients with T2D. Methods: Employing the PRISMA and PICOS guidelines, we conducted a comprehensive search to incorporate evidence from relevant studies, focusing on randomized controlled trials (RCTs), systematic reviews, and meta-analyses published since 2017. Of the 437 publications identified, seven were selected based on predetermined criteria. Results: The seven studies included in this review used various sensing technologies, such as heart rate monitors, accelerometers, and other wearable devices. Primary health outcomes included blood pressure measurements, heart rate, body fat percentage, and cardiorespiratory endurance. Non-invasive wearable devices demonstrated potential for aiding T2D management, albeit with variations in efficacy across studies. Conclusions: Based on the low number of studies with higher evidence levels (i.e., RCTs) that we were able to find and the significant differences in design between these studies, we conclude that further evidence is required to validate the application, efficacy, and real-world impact of these wearable devices. Emphasizing transparency in bias reporting and conducting in-depth research is crucial for fully understanding the implications and benefits of wearable devices in T2D management.

Remote monitoring for long-term physical health conditions: an evidence and gap map

Background

Remote monitoring involves the measurement of an aspect of a patient's health without that person being seen face to face. It could benefit the individual and aid the efficient provision of health services. However, remote monitoring can be used to monitor different aspects of health in different ways. This evidence map allows users to find evidence on different forms of remote monitoring for different conditions easily to support the commissioning and implementation of interventions.

Objectives

The aim of this map was to provide an overview of the volume, diversity and nature of recent systematic reviews on the effectiveness, acceptability and implementation of remote monitoring for adults with long-term physical health conditions.

Data sources

We searched MEDLINE, nine further databases and Epistemonikos for systematic reviews published between 2018 and March 2022, PROSPERO for continuing reviews, and completed citation chasing on included studies.

Review methods

(Study selection and Study appraisal): Included systematic reviews focused on adult populations with a long-term physical health condition and reported on the effectiveness, acceptability or implementation of remote monitoring. All forms of remote monitoring where data were passed to a healthcare professional as part of the intervention were included. Data were extracted on the characteristics of the remote monitoring intervention and outcomes assessed in the review. AMSTAR 2 was used to assess quality. Results were presented in an interactive evidence and gap map and summarised narratively. Stakeholder and public and patient involvement groups provided feedback throughout the project.

Results

We included 72 systematic reviews. Of these, 61 focus on the effectiveness of remote monitoring and 24 on its acceptability and/or implementation, with some reviews reporting on both. The majority contained studies from North America and Europe (38 included studies from the United Kingdom). Patients with cardiovascular disease, diabetes and respiratory conditions were the most studied populations. Data were collected predominantly using common devices such as blood pressure monitors and transmitted via applications, websites, e-mail or patient portals, feedback provided via telephone call and by nurses. In terms of outcomes, most reviews focused on physical health, mental health and well-being, health service use, acceptability or implementation. Few reviews reported on less common conditions or on the views of carers or healthcare professionals. Most reviews were of low or critically low quality.

Limitations

Many terms are used to describe remote monitoring; we searched as widely as possible but may have missed some relevant reviews. Poor reporting of remote monitoring interventions may mean some included reviews contain interventions that do not meet our definition, while relevant reviews might have been excluded. This also made the interpretation of results difficult.

Conclusions and future work

The map provides an interactive, visual representation of evidence on the effectiveness of remote monitoring and its acceptability and successful implementation. This evidence could support the commissioning and delivery of remote monitoring interventions, while the limitations and gaps could inform further research and technological development. Future reviews should follow the guidelines for conducting and reporting systematic reviews and investigate the application of remote monitoring in less common conditions.

Review registration

A protocol was registered on the OSF registry (https://doi.org/10.17605/OSF.IO/6Q7P4).

Funding

This award was funded by the National Institute for Health and Care Research (NIHR) Health Services and Delivery Research programme (NIHR award ref: NIHR135450) as part of a series of evidence syntheses under award NIHR130538. For more information, visit https://fundingawards.nihr.ac.uk/award/NIHR135450 and https://fundingawards.nihr.ac.uk/award/NIHR130538. The report is published in full in Health and Social Care Delivery Research; Vol. 11, No. 22. See the NIHR Funding and Awards website for further project information.

A rapid mixed-methods evaluation of remote home monitoring models during the COVID-19 pandemic in England

Background

Remote home monitoring services were developed and implemented for patients with COVID-19 during the pandemic. Patients monitored blood oxygen saturation and other readings (e.g. temperature) at home and were escalated as necessary.

Objective

To evaluate effectiveness, costs, implementation, and staff and patient experiences (including disparities and mode) of COVID-19 remote home monitoring services in England during the COVID-19 pandemic (waves 1 and 2).

Methods

A rapid mixed-methods evaluation, conducted in two phases. Phase 1 (July-August 2020) comprised a rapid systematic review, implementation and economic analysis study (in eight sites). Phase 2 (January-June 2021) comprised a large-scale, multisite, mixed-methods study of effectiveness, costs, implementation and patient/staff experience, using national data sets, surveys (28 sites) and interviews (17 sites).

Results

Phase 1 Findings from the review and empirical study indicated that these services have been implemented worldwide and vary substantially. Empirical findings highlighted that communication, appropriate information and multiple modes of monitoring facilitated implementation; barriers included unclear referral processes, workforce availability and lack of administrative support. Phase 2 We received surveys from 292 staff (39% response rate) and 1069 patients/carers (18% response rate). We conducted interviews with 58 staff, 62 patients/carers and 5 national leads. Despite national roll-out, enrolment to services was lower than expected (average enrolment across 37 clinical commissioning groups judged to have completed data was 8.7%). There was large variability in implementation of services, influenced by patient (e.g. local population needs), workforce (e.g. workload), organisational (e.g. collaboration) and resource (e.g. software) factors. We found that for every 10% increase in enrolment to the programme, mortality was reduced by 2% (95% confidence interval: 4% reduction to 1% increase), admissions increased by 3% (-1% to 7%), in-hospital mortality fell by 3% (-8% to 3%) and lengths of stay increased by 1.8% (-1.2% to 4.9%). None of these results are statistically significant. We found slightly longer hospital lengths of stay associated with virtual ward services (adjusted incidence rate ratio 1.05, 95% confidence interval 1.01 to 1.09), and no statistically significant impact on subsequent COVID-19 readmissions (adjusted odds ratio 0.95, 95% confidence interval 0.89 to 1.02). Low patient enrolment rates and incomplete data may have affected chances of detecting possible impact. The mean running cost per patient varied for different types of service and mode; and was driven by the number and grade of staff. Staff, patients and carers generally reported positive experiences of services. Services were easy to deliver but staff needed additional training. Staff knowledge/confidence, NHS resources/workload, dynamics between multidisciplinary team members and patients' engagement with the service (e.g. using the oximeter to record and submit readings) influenced delivery. Patients and carers felt services and human contact received reassured them and were easy to engage with. Engagement was conditional on patient, support, resource and service factors. Many sites designed services to suit the needs of their local population. Despite adaptations, disparities were reported across some patient groups. For example, older adults and patients from ethnic minorities reported more difficulties engaging with the service. Tech-enabled models helped to manage large patient groups but did not completely replace phone calls.

Limitations

Limitations included data completeness, inability to link data on service use to outcomes at a patient level, low survey response rates and under-representation of some patient groups.

Future work

Further research should consider the long-term impact and cost-effectiveness of these services and the appropriateness of different models for different groups of patients.

Conclusions

We were not able to find quantitative evidence that COVID-19 remote home monitoring services have been effective. However, low enrolment rates, incomplete data and varied implementation reduced our chances of detecting any impact that may have existed. While services were viewed positively by staff and patients, barriers to implementation, delivery and engagement should be considered.

Study registration

This study is registered with the ISRCTN (14962466).

Funding

This project was funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme (RSET: 16/138/17; BRACE: 16/138/31) and NHSEI and will be published in full in Health and Social Care Delivery Research; Vol. 11, No. 13. See the NIHR Journals Library website for further project information. The views expressed in this publication are those of the authors and not necessarily those of the National Institute for Health and Care Research or the Department of Health and Social Care.

How do nurses work in chronic management in the age of artificial intelligence? development and future prospects

AI is undeniably revolutionizing medical research and patient care across diverse fields. Chronic disease nursing care, a pivotal aspect of clinical management, has significantly reaped the benefits of AI across numerous dimensions. Understanding the operational principles of artificial intelligence before implementation is crucial, avoiding indiscriminate replacement of all tasks with AI. Nurses serve as the primary force in symptom group research, expanding beyond diabetes to encompass various chronic diseases; their primary responsibility involves recording patients' daily symptoms and vital signs. However, a substantial portion of current AI research excludes nurses from the developmental phase, encompassing them solely in user and feedback populations. The comprehensive design of the symptom analysis and long-term management approach necessitates the guidance and oversight of nurses; however, their current insufficient involvement might stem from nursing staff's comparatively limited comprehension of AI and their ambiguous perception of their role's value in AI. Therefore, an imperative exploration of nurses' roles in symptom analysis and long-term management, leveraging the latest research in these areas, is vital to pinpoint breakthroughs in nurses' AI involvement in the future.

Patients' experiences of, and engagement with, remote home monitoring services for COVID-19 patients: A rapid mixed-methods study

INTRODUCTION

Remote home monitoring models were implemented during the COVID-19 pandemic to shorten hospital length of stay, reduce unnecessary hospital admission, readmission and infection and appropriately escalate care. Within these models, patients are asked to take and record readings and escalate care if advised. There is limited evidence on how patients and carers experience these services. This study aimed to evaluate patient experiences of, and engagement with, remote home monitoring models for COVID-19.

METHODS

A rapid mixed-methods study was carried out in England (conducted from March to June 2021). We remotely conducted a cross-sectional survey and semi-structured interviews with patients and carers. Interview findings were summarized using rapid assessment procedures sheets and data were grouped into themes (using thematic analysis). Survey data were analysed using descriptive statistics.

RESULTS

We received 1069 surveys (18% response rate) and conducted interviews with patients (n = 59) or their carers (n = 3). 'Care' relied on support from staff members and family/friends. Patients and carers reported positive experiences and felt that the service and human contact reassured them and was easy to engage with. Yet, some patients and carers identified problems with engagement (e.g., hesitancy to self-escalate care). Engagement was influenced by patient factors such as health and knowledge, support from family/friends and staff, availability and ease of use of informational and material resources (e.g., equipment) and service factors.

CONCLUSION

Remote home monitoring models place responsibility on patients to self-manage symptoms in partnership with staff; yet, many patients required support and preferred human contact (especially for identifying problems). Caring burden and experiences of those living alone and barriers to engagement should be considered when designing and implementing remote home monitoring services.

PATIENT OR PUBLIC CONTRIBUTION

The study team met with service users and public members of the evaluation teams throughout the project in a series of workshops. Workshops informed study design, data collection tools and data interpretation and were conducted to also discuss study dissemination. Public patient involvement (PPI) members helped to pilot patient surveys and interview guides with the research team. Some members of the public also piloted the patient survey. Members of the PPI group were given the opportunity to comment on the manuscript, and the manuscript was amended accordingly.

A Powerful Paradigm for Cardiovascular Risk Stratification Using Multiclass, Multi-Label, and Ensemble-Based Machine Learning Paradigms: A Narrative Review

Background and Motivation: Cardiovascular disease (CVD) causes the highest mortality globally. With escalating healthcare costs, early non-invasive CVD risk assessment is vital. Conventional methods have shown poor performance compared to more recent and fast-evolving Artificial Intelligence (AI) methods. The proposed study reviews the three most recent paradigms for CVD risk assessment, namely multiclass, multi-label, and ensemble-based methods in (i) office-based and (ii) stress-test laboratories. Methods: A total of 265 CVD-based studies were selected using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) model. Due to its popularity and recent development, the study analyzed the above three paradigms using machine learning (ML) frameworks. We review comprehensively these three methods using attributes, such as architecture, applications, pro-and-cons, scientific validation, clinical evaluation, and AI risk-of-bias (RoB) in the CVD framework. These ML techniques were then extended under mobile and cloud-based infrastructure. Findings: Most popular biomarkers used were office-based, laboratory-based, image-based phenotypes, and medication usage. Surrogate carotid scanning for coronary artery risk prediction had shown promising results. Ground truth (GT) selection for AI-based training along with scientific and clinical validation is very important for CVD stratification to avoid RoB. It was observed that the most popular classification paradigm is multiclass followed by the ensemble, and multi-label. The use of deep learning techniques in CVD risk stratification is in a very early stage of development. Mobile and cloud-based AI technologies are more likely to be the future. Conclusions: AI-based methods for CVD risk assessment are most promising and successful. Choice of GT is most vital in AI-based models to prevent the RoB. The amalgamation of image-based strategies with conventional risk factors provides the highest stability when using the three CVD paradigms in non-cloud and cloud-based frameworks.

Use of mobile applications for heart failure self-management: a systematic review of experimental and qualitative studies (Preprint)

Background

Heart failure self-management is essential to avoid decompensation and readmissions. Mobile apps seem promising in supporting heart failure self-management, and there has been a rapid growth in publications in this area. However, to date, systematic reviews have mostly focused on remote monitoring interventions using nonapp types of mobile technologies to transmit data to health care providers, rarely focusing on supporting patient self-management of heart failure.

Objective

This study aims to systematically review the evidence on the effect of heart failure self-management apps on health outcomes, patient-reported outcomes, and patient experience.

Methods

Four databases (PubMed, Embase, CINAHL, and PsycINFO) were searched for studies examining interventions that comprised a mobile app targeting heart failure self-management and reported any health-related outcomes or patient-reported outcomes or perspectives published from 2008 to December 2021. The studies were independently screened. The risk of bias was appraised using Cochrane tools. We performed a narrative synthesis of the results. The protocol was registered on PROSPERO (International Prospective Register of Systematic Reviews; CRD42020158041).

Results

A total of 28 articles (randomized controlled trials [RCTs]: n=10, 36%), assessing 23 apps, and a total of 1397 participants were included. The most common app features were weight monitoring (19/23, 83%), symptom monitoring (18/23, 78%), and vital sign monitoring (15/23, 65%). Only 26% (6/23) of the apps provided all guideline-defined core components of heart failure self-management programs: education, symptom monitoring, medication support, and physical activity support. RCTs were small, involving altogether 717 participants, had ≤6 months of follow-up, and outcomes were predominantly self-reported. Approximately 20% (2/10) of RCTs reported a significant improvement in their primary outcomes: heart failure knowledge (P=.002) and self-care (P=.004). One of the RCTs found a significant reduction in readmissions (P=.02), and 20% (2/10) of RCTs reported higher unplanned clinic visits. Other experimental studies also found significant improvements in knowledge, self-care, and readmissions, among others. Less than half of the studies involved patients and clinicians in the design of apps. Engagement with the intervention was poorly reported, with only 11% (3/28) of studies quantifying app engagement metrics such as frequency of use over the study duration. The most desirable app features were automated self-monitoring and feedback, personalization, communication with clinicians, and data sharing and integration.

Conclusions

Mobile apps may improve heart failure self-management; however, more robust evaluation studies are needed to analyze key end points for heart failure. On the basis of the results of this review, we provide a road map for future studies in this area.