Background: Despite a recent rise in adoption, telemedicine consultations retention remains challenging, and aspects around the associated experiences and outcomes remain unclear. The need to further investigate these aspects was a motivating factor for conducting this scoping review. Objective: With a focus on synchronous telemedicine consultations between patients with nonmalignant chronic illnesses and health care professionals (HCPs), this scoping review aimed to gain insights into (1) the available evidence on telemedicine consultations to improve health outcomes for patients, (2) the associated behaviors and attitudes of patients and HCPs, and (3) how supplemental technology can assist in remote consultations. Methods: PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guided the scoping review process. Inclusion criteria were (1) involving adults with nonmalignant, noncommunicable chronic conditions as the study population; (2) focusing on health outcomes and experiences of and attitudes toward synchronous telemedicine consultations between patients and HCPs; and (3) conducting empirical research. A search strategy was applied to PubMed (including MEDLINE), CINAHL Complete, APA PsycNet, Web of Science, IEEE, and ACM Digital. Screening of articles and data extraction from included articles were performed in parallel and independently by 2 researchers, who corroborated their findings and resolved any conflicts. Results: Overall, 4167 unique articles were identified from the databases searched. Following multilayer filtration, 19 (0.46%) studies fulfilled the inclusion criteria for data extraction. They investigated 6 nonmalignant chronic conditions, namely chronic obstructive pulmonary disease, diabetes, chronic kidney disease, ulcerative colitis, hypertension, and congestive heart failure, and the telemedicine consultation modality varied in each case. Most observed positive health outcomes for patients with chronic conditions using telemedicine consultations. Patients generally favored the modality’s convenience, but concerns were highlighted around cost, practical logistics, and thoroughness of clinical examinations. The majority of HCPs were also in favor of the technology, but a minority experienced reduced job satisfaction. Supplemental technological assistance was identified in relation to technical considerations, improved remote workflow, and training in remote care use. Conclusions: For patients with noncommunicable chronic conditions, telemedicine consultations are generally associated with positive health outcomes that are either directly or indirectly related to their ailment, but sustained improvements remain unclear. These modalities also indicate the potential to empower such patients to better manage their condition. HCPs and patients tend to be satisfied with remote care experience, and most are receptive to the modality as an option. Assistance from supplemental technologies mostly resides in addressing technical issues, and additional modules could be integrated to address challenges relevant to patients and HCPs. However, positive outcomes and attitudes toward the modality might not apply to all cases, indicating that telemedicine consultations are more appropriate as options rather than replacements of in-person visits.

Background: Concussion in sport is an ongoing global concern. The head injury assessment (HIA) by the field of play is acknowledged as the first step in recognising and identifying concussion. While previous systematic literature reviews have evaluated the sensitivity of side-line screening tools and assessment protocols, no systematic review has evaluated the research designs and assessments used in a field setting. This systematic review investigated existing screening and diagnostic tools used in research as part of the HIA protocol to identify concussion that are currently used in professional, semi-professional and amateur (club) sports settings.Methods: A systematic searching of relevant databases was undertaken for peer-reviewed literature between 2015 and 2020.Results: Twenty-six studies met the inclusion criteria. Studies were of moderate to good quality, reporting a variety of designs. The majority of studies were undertaken in professional/elite environments with medical doctors and allied health practitioners (e.g., physical therapists) involved in 88% of concussion assessments. While gender was reported in 24 of the 26 studies, the majority of participants were male (77%). There was also a variety of concussion assessments (n = 20) with the sports concussion assessment tool (SCAT) used in less than half of the included studies.Conclusion: The majority of studies investigating concussion HIAs are focused on professional/elite sport. With concussion an issue at all levels of sport, future research should be directed at non-elite sport. Further, for research purposes, the SCAT assessment should also be used more widely to allow for consistency across studies.

The real-time forecasting of critical physiological indicators in intensive care units (ICUs) is essential for early intervention and clinical decision support. This study introduces a novel framework, StreamHealth Multi-Horizon AI, which has been designed to perform multivariate, multi-horizon time-series forecasting for vital signs, specifically for a person's blood oxygen saturation level (SpO2) and respiratory rate (RR). The framework leverages advanced attention-based models, with a particular emphasis on the Temporal Fusion Transformer (TFT) and Temporal Convolutional Network (TCN), and we benchmark its performance against classical deep learning architectures, including LSTM, GRU, Bi-LSTM, Bi-GRU, CNN, and Sequence-to-Sequence (Seq2Seq) models with and without attention mechanisms. Both univariate and multivariate forecasting tasks are explored across multiple prediction horizons (i.e., 7, 15 and 25 min), using physiological time-series data from the MIMIC-III database. The proposed system incorporates a cascaded fine-tuning strategy, wherein the TFT model is sequentially fine-tuned on individual patients' data, significantly enhancing the model's generalizability to unseen patient profiles. Empirical results demonstrate that the TFT model consistently outperforms baseline models across all forecasting settings, achieving lower RMSE and MAE values, and exhibiting superior capacity for capturing long-sequence dependencies and temporal feature dynamics. To validate its applicability in real-time clinical environments, the framework integrates a simulated streaming infrastructure using Apache Kafka and Apache Flink, enabling continuous data ingestion, forecasting, and visualization of vital signs. This end-to-end deployment underscores the system's potential for ICU monitoring, allowing clinicians to anticipate patient deterioration proactively. In summary, we introduce a comprehensive framework that uniquely integrates TFT with cascaded fine-tuning for multivariate, multi-horizon forecasting of critical ICU indicators. Additionally, we demonstrate a simulation for a real-time deployment pipeline using Kafka and Flink, enabling robust and generalizable ICU monitoring in clinical settings. As a result, this work has contributed a robust and clinically relevant AI solution for real-time healthcare monitoring.

Background: The COVID-19 pandemic represents the most unprecedented global challenge in recent times. As the global community attempts to manage the pandemic in the long term, it is pivotal to understand what factors drive prevalence rates and to predict the future trajectory of the virus. Objective: This study had 2 objectives. First, it tested the statistical relationship between socioeconomic status and COVID-19 prevalence. Second, it used machine learning techniques to predict cumulative COVID-19 cases in a multicountry sample of 182 countries. Taken together, these objectives will shed light on socioeconomic status as a global risk factor of the COVID-19 pandemic. Methods: This research used exploratory data analysis and supervised machine learning methods. Exploratory analysis included variable distribution, variable correlations, and outlier detection. Following this, the following 3 supervised regression techniques were applied: linear regression, random forest, and adaptive boosting (AdaBoost). Results were evaluated using k-fold cross-validation and subsequently compared to analyze algorithmic suitability. The analysis involved 2 models. First, the algorithms were trained to predict 2021 COVID-19 prevalence using only 2020 reported case data. Following this, socioeconomic indicators were added as features and the algorithms were trained again. The Human Development Index (HDI) metrics of life expectancy, mean years of schooling, expected years of schooling, and gross national income were used to approximate socioeconomic status. Results: All variables correlated positively with the 2021 COVID-19 prevalence, with R2 values ranging from 0.55 to 0.85. Using socioeconomic indicators, COVID-19 prevalence was predicted with a reasonable degree of accuracy. Using 2020 reported case rates as a lone predictor to predict 2021 prevalence rates, the average predictive accuracy of the algorithms was low (R2=0.543). When socioeconomic indicators were added alongside 2020 prevalence rates as features, the average predictive performance improved considerably (R2=0.721) and all error statistics decreased. Thus, adding socioeconomic indicators alongside 2020 reported case data optimized the prediction of COVID-19 prevalence to a considerable degree. Linear regression was the strongest learner with R2=0.693 on the first model and R2=0.763 on the second model, followed by random forest (0.481 and 0.722) and AdaBoost (0.454 and 0.679). Following this, the second model was retrained using a selection of additional COVID-19 risk factors (population density, median age, and vaccination uptake) instead of the HDI metrics. However, average accuracy dropped to 0.649, which highlights the value of socioeconomic status as a predictor of COVID-19 cases in the chosen sample. Conclusions: The results show that socioeconomic status is an important variable to consider in future epidemiological modeling, and highlights the reality of the COVID-19 pandemic as a social phenomenon and a health care phenomenon. This paper also puts forward new considerations about the application of statistical and machine learning techniques to understand and combat the COVID-19 pandemic.

The aim of this study was to determine if the quantity of running load performed in pre-season affects the incidence of injury in elite Gaelic footballers. It was hypothesized that a greater quantity of running loads completed will reduce the incidence rate of injury. A cohort of elite male Gaelic football players (n = 25) was prospectively monitored throughout one full season. This longitudinal study involved the collection of GPS data and injury data. Participants were retrospectively divided into two groups and assigned to a high (HTL, completed >50% of pre-season sessions, n = 13) or low (LTL, completed 6.4 m/s) and high-intensity running meters (high-speed running meters and sprint meters combined) were collected during all sessions. A one-way analysis of variance (ANOVA) was completed to understand the variation of external training load data across the different phases of the season. A series of repeated measures of ANOVA were subsequently completed to understand the variation of external training load data across seasonal phases within the training groups. Although the LTL group had a higher incidence rate of non-contact injuries (large effect size) per 1000 h of exposure in each phase of the season, statistical analysis revealed that there was no significant difference (F = 4.32, p = 0.173, partial η² = 0.684, large) between the HTL (14.9 ± 4.17/1000 h) and the LTL (24.5 ± 7.36/1000 h) groups. Additionally, the HTL group was able to sustain greater running loads in the competitive phases of the season compared to the LTL group, total distance (F = 8.10, p < 0.001, partial η² = 0.299, large), high-speed running distance (F = 8.74, p < 0.001, partial η² = 0.304, large) and high-intensity running distance (F = 7.63, p < 0.001, partial η² = 0.276, large). Furthermore, players who complete a greater proportion of running loads in pre-season may alter their body composition levels to more optimal levels, which in turn may reduce the risk of injury while also helping increase or maintain performance-related fitness markers such as aerobic fitness.