Communications Challenges in Health and Well-Being

Dataset imbalances pose a significant challenge to predictive modeling in both medical and financial domains, where conventional strategies, including resampling and algorithmic modifications, often fail to adequately address minority class underrepresentation. This study theoretically and practically investigates how the inherent nature of medical data affects the classification of minority classes. It employs ten machine and deep learning classifiers, ranging from ensemble learners to cost-sensitive algorithms, across comparably sized medical and financial datasets. Despite these efforts, none of the classifiers achieved effective classification of the minority class in the medical dataset, with sensitivity below 5.0% and area under the curve (AUC) below 57.0%. In contrast, the similar classifiers applied to the financial dataset demonstrated strong discriminative power, with overall accuracy exceeding 95.0%, sensitivity over 73.0%, and AUC above 96.0%. This disparity underscores the unpredictable variability inherent in the nature of medical data, as exemplified by the dispersed and homogeneous distribution of the minority class among other classes in principal component analysis (PCA) graphs. The application of the synthetic minority oversampling technique (SMOTE) introduced 62 synthetic patients based on merely 20 original cases, casting doubt on its clinical validity and the representation of real-world patient variability. Furthermore, post-SMOTE feature importance analysis, utilizing SHapley Additive exPlanations (SHAP) and tree-based methods, contradicted established cerebral stroke parameters, further questioning the clinical coherence of synthetic dataset augmentation. These findings call into question the clinical validity of the SMOTE technique and underscore the urgent need for advanced modeling techniques and algorithmic innovations for predicting minority-class outcomes in medical datasets without depending on resampling strategies. This approach underscores the importance of developing methods that are not only theoretically robust but also clinically relevant and applicable to real-world clinical scenarios. Consequently, this study underscores the importance of future research efforts to bridge the gap between theoretical advancements and the practical, clinical applications of models like SMOTE in healthcare. Full article

Early identification of stroke symptoms is essential. The rate of stroke identification by call-takers at emergency medical communication centres (EMCCs) varies, and patients who are found in a lying down position are often not identified as having an ongoing stroke. Objectives: this study aimed to explore signs and symptoms of stroke in patients who had fallen or were found in a lying position. Design: a retrospective exploratory qualitative study design was used. Method: a total of 29 emergency calls to EMCCs regarding patients discharged with a stroke diagnosis from a large teaching hospital in Stockholm, Sweden, in January–June 2011, were analysed using qualitative content analysis. Results: during the emergency calls, the callers described a sudden change in the patient’s health status including signs such as the patient’s loss of bodily control, the patient’s perception of a change in sensory perception, and the callers’ inability to communicate with the patient. Conclusions: The callers’ descriptions of stroke in a person found in a lying position are not always as described in assessment protocols describing the onset of a stroke. Instead, the symptom descriptions are much vaguer. Therefore, to increase identification of stroke during emergency calls, there is a need for an increased understanding of how callers describe stroke symptoms and communicate with the call-takers. Full article

The automotive industry and scientific community are making efforts to develop innovative solutions that would increase successful driver performance in preventing crashes caused by drivers’ health and concentration. High stress is one of the causes of impaired driver performance. This study investigates the ability to classify different stress levels based on capacitive electrocardiogram (cECG) recorded during driving by unobtrusive acquisition systems with different hardware implementations. The proposed machine-learning model extracted only four features, based on the detection of the R peak, which is the most reliably detected characteristic point even in inferior quality cECG. Another criterion for selecting the features is their low computational complexity, which enables real-time application. The proposed method was validated on three open data sets recorded during driving: electrocardiogram (ECG) recorded by electrodes with direct skin contact (high quality); cECG recorded without direct skin contact through clothes by electrodes built into a portable multi-modal cushion (middle quality); and cECG recorded through the clothes without direct skin contact by electrodes built into a car seat (lowest quality). The proposed model achieved a high accuracy of 100% for high-quality ECG, 96.67% for middle-quality cECG, and 98.08% for the lower-quality cECG. Full article

Background: Communication is an important clinical tool for the prevention and control of diseases, to advise and inform patients and the public, providing them with essential knowledge regarding healthcare and disease management. This study explored the experience of communication between healthcare professionals (HCPs) and people with long-term lung conditions, from the patient perspective. Methods: This qualitative study analyzed the experience of people with chronic lung disease, recruited via Asthma & Lung UK (A&LUK) and COPD research databases. A&LUK invited people who had expressed a desire to be involved in research associated with their condition via their Expert Patient Panel and associated patients’ groups. Two focus group interviews (12 participants) and one individual interview (1 participant) were conducted. Thematic analysis was used for data analysis. Results: Two main themes were identified and we named them ‘involving communication’ and ‘communication needs to be improved. ‘They included seven subthemes: community-led support increased the patients’ social interaction with peers; allied-HCP-led support increased patients’ satisfaction; disliking being repeatedly asked the same basic information; feeling communication was unengaging, lacking personal specifics and the use of medical terminology and jargon. Conclusions: The study has identified what most matters in the process of communication with HCPs in people with long-term respiratory diseases of their healthcare management. The findings of the study can be used to improve the patient–healthcare professional relationship and facilitate a better communication flow in long-term healthcare management. Full article

From the beginning of their clinical training, medical students demonstrate difficulties when incorporating patient perspectives. This study aimed to assess if students, after an instructional programme, increased their sensitivity towards patients’ needs and carried out bidirectional conversations. An observational study involving 109 medical students prior to their clerkships was designed. They attended a five-step training programme designed to encourage the use of communication skills (CSs) to obtain patients’ perspectives. The course developed experiential and reflective educational strategies. The students improved their use of CSs throughout three sessions, and the overall score for these patient consultations went up in the opinions of both the external observer (EO) (5; 6.6; 7.5) and the simulated patients SPs (5.3; 6.6; 7.8). Most of the students (83.9%) considered that the CSs addressed were useful for clinical practice, particularly the interviews and the feedback received by the SP and the lecturer. The programme seems to help the students use CSs that facilitate a more bidirectional conversation in a simulated learning environment. It is feasible to integrate these skills into a broader training programme. More research is needed to assess whether the results are applicable to students in real settings and whether they influence additional outcomes. Full article

Background: In Australia, despite the success of tobacco control policy interventions, policymakers remain resistant to policy-based approaches to diet, alcohol, physical inactivity and obesity, concerned about community perceptions of such interventions as “nanny-statist”. We examined how people’s general positions on government intervention related to their positions on different preventive policy options. Methods: Data were from a 2018 nationally representative cross-sectional telephone survey of 2601 Australian adults. Survey questions related to endorsement of different conceptualisations of government intervention (nanny state, paternalistic, shared responsibility and communitarian) and support for specific health interventions, using forced-choice questions about preferences for individual/treatment measures versus population/preventive health measures. We analysed associations between scores on different conceptualisations of government intervention and support of different policy options for tobacco and diet, and preferences for prevention over treatment. Results: The Nanny State Scale showed an inverse relationship with support for tobacco- and diet-related interventions, and alternative conceptualisations (paternalistic, shared responsibility and communitarian) showed a positive relationship. Effect sizes in all cases were small. Those aged 55+ demonstrated greater support for policy action on tobacco and diet, and greater preference for systemic rather than individual-level interventions. Conclusion: General disposition towards government intervention, although correlated with support for specific policy actions, is not deterministic. Full article

Background: Non-invasive tracking of beat-to-beat pulse transit time (PTT) via piezoelectric/piezocapacitive sensors (PES/PCS) may expand perioperative hemodynamic monitoring. This study evaluated the ability for PTT via PES/PCS to correlate with systolic, diastolic, and mean invasive blood pressure (SBP_IBP, DBP_IBP, and MAP_IBP, respectively) and to detect SBP_IBP fluctuations. Methods: PES/PCS and IBP measurements were performed in 20 patients undergoing abdominal, urological, and cardiac surgery. A Pearson’s correlation analysis (r) between 1/PTT and IBP was performed. The predictive ability of 1/PTT with changes in SBP_IBP was determined by area under the curve (reported as AUC, sensitivity, specificity). Results: Significant correlations between 1/PTT and SBP_IBP were found for PES (r = 0.64) and PCS (r = 0.55) (p < 0.01), as well as MAP_IBP/DBP_IBP for PES (r = 0.6/0.55) and PCS (r = 0.5/0.45) (p < 0.05). A 7% decrease in 1/PTT_PES predicted a 30% SBP_IBP decrease (0.82, 0.76, 0.76), while a 5.6% increase predicted a 30% SBP_IBP increase (0.75, 0.7, 0.68). A 6.6% decrease in 1/PTT_PCS detected a 30% SBP_IBP decrease (0.81, 0.72, 0.8), while a 4.8% 1/PTT_PCS increase detected a 30% SBP_IBP increase (0.73, 0.64, 0.68). Conclusions: Non-invasive beat-to-beat PTT via PES/PCS demonstrated significant correlations with IBP and detected significant changes in SBP_IBP. Thus, PES/PCS as a novel sensor technology may augment intraoperative hemodynamic monitoring during major surgery. Full article

Older adults’ independent life is compromised due to various problems, such as memory impairments and decision-making difficulties. This work initially proposes an integrated conceptual model for assisted living systems capable of providing helping means for older adults with mild memory impairments and their caregivers. The proposed model has four main components: (1) an indoor location and heading measurement unit in the local fog layer, (2) an augmented reality (AR) application to make interactions with the user, (3) an IoT-based fuzzy decision-making system to handle the direct and environmental interactions with the user, and (4) a user interface for caregivers to monitor the situation in real time and send reminders once required. Then, a preliminary proof-of-concept implementation is performed to evaluate the suggested mode’s feasibility. Functional experiments are carried out based on various factual scenarios, which validate the effectiveness of the proposed approach. The accuracy and response time of the proposed proof-of-concept system are further examined. The results suggest that implementing such a system is feasible and has the potential to promote assisted living. The suggested system has the potential to promote scalable and customizable assisted living systems to reduce the challenges of independent living for older adults. Full article

This study aimed to evaluate whether psychological resilience is an independent factor of self-rated health (SRH) among patients with knee osteoarthritis (KOA). A cross-sectional study with convenience sampling was designed. Patients with doctor-diagnosed KOA were recruited from the orthopedic outpatient departments of a hospital in southern Taiwan. Psychological resilience was measured by the 10-item Connor–Davidson Resilience Scale (CD–RISC-10), and SRH was measured by three items, including the current SRH, the preceding year-related SRH, and age-related SRH. The three-item SRH scale was categorized as “high” and “low–moderate” groups by terciles. Covariates included KOA history, site of knee pain, joint-specific symptoms measured by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), comorbidity measured by Charlson Comorbidity Index, and demographic variables (i.e., age, sex, education attainment, and living arrangements). A multiple logistic regression was used to detect the independent variables with significant odds ratios that can predict “high” SRH among participants. Results: In total, 98 patients with KOA (66 women and 32 men) with a mean age (±SD) of 68.3 ± 8.5 years were enrolled and were analyzed. A total of 38.8% (n = 38) of participants were categorized as “high SRH”, while 61.2% (n = 60) were categorized as “low–moderate SRH”. Multiple logistic regression showed that CD–RISC-10 had an increased odds ratio (OR) for high SRH (OR [95% CI] = 1.061 [1.003–1.122]; p = 0.038), whereas bilateral pain (vs. unilateral pain), WOMAC stiffness, and WOMAC physical limitation showed a decreased OR for high SRH (0.268 [0.098–0.732], 0.670 [0.450–0.998], and 0.943 [0.891–0.997], respectively). Our findings provide evidence indicating that psychological resilience plays a significant positive role in the SRH in our study sample. Further research is required to extend the growing knowledge regarding the application of psychological resilience on KOA. Full article

Social media represents a powerful tool for disseminating verified health information on topics such as colostomy, and the roles of healthcare professionals and institutions to ensure the veracity of the information conveyed is increasingly relevant. The main objectives of this study were to analyze the roles of these healthcare professionals and institutions in the conversation about colostomy, without being framed in a specific health communication campaign, and to know the use of reliable information in the conversation. The study was carried out by analyzing Twitter messages containing the hashtag “colostomy” and “Chron” between the 1 January and the 30 April 2022. It was conducted using the NodeXL software, focusing on content analysis of tweets and users’ accounts. The results show that accounts with healthcare activity influence the impressions generated on the network (p = 0.018), finding that nurses are the most active healthcare professionals (22.24%) also having a significant effect on the overall network interactions (p = 0.022). In contrast, we found that institutions do not actively participate on the network. We emphasize the responsibility of institutions for health education and the need for professionals to improve communication skills on social networks, but also the need to improve communication skills on social media to support public health campaigns through these increasingly important channels. Full article

IoT-based insulin pumps are used to deliver precise quantities of insulin to diabetic patients to regulate blood glucose levels. Generally, these levels correspond to the dietary patterns observed at time intervals that vary between patients. However, any misrepresentation in insulin levels may lead to fatal consequences. As a result, most IoT-based insulin pumps are rejected due to the possibility of external threats, which include software and hardware attacks. However, IoT-based insulin pumps are extremely useful in real-time patient monitoring, and for controlled insulin delivery to the patient based on their current glucose level. We propose a blockchain-based method to protect against the above-mentioned attacks. The system creates a patient-specific private blockchain wherein the dosage information is added as a new block by obtaining the approval of the doctor, chief doctor, nurse, and caretaker of the patient who are authorized blockchain miners. Secondly, it securely transfers prescription data, such as dosage quantity and time of delivery, to the IoT insulin pump, which ensures the dosage information is not modified during transit before insulin administration to the patient. The proposed approach uses a state-behavior-based solution that detects anomalies in the behavior of the insulin pump via temporal data analysis and immutable ledger verification, which are designed to eliminate fatal dosages in case of anomalies. The system is designed to work within binary outcome conditions, i.e., it verifies and delivers dosage or halts. There is no middle ground that an attacker can exploit, resulting in accountability for the system. Full article