Search Results (189)

Background: Artificial intelligence (AI) and machine learning (ML) have been reshaping maternal, fetal, neonatal, and reproductive healthcare by enhancing risk prediction, diagnostic accuracy, and operational efficiency across the perinatal continuum. However, no comprehensive synthesis has yet been published. Objective: To conduct a scoping review of reviews of AI/ML applications spanning reproductive, prenatal, postpartum, neonatal, and early child-development care. Methods: We searched PubMed, Embase, the Cochrane Library, Web of Science, and Scopus through April 2025. Two reviewers independently screened records, extracted data, and assessed methodological quality using AMSTAR 2 for systematic reviews, ROBIS for bias assessment, SANRA for narrative reviews, and JBI guidance for scoping reviews. Results: Thirty-nine reviews met our inclusion criteria. In preconception and fertility treatment, convolutional neural network-based platforms can identify viable embryos and key sperm parameters with over 90 percent accuracy, and machine-learning models can personalize follicle-stimulating hormone regimens to boost mature oocyte yield while reducing overall medication use. Digital sexual-health chatbots have enhanced patient education, pre-exposure prophylaxis adherence, and safer sexual behaviors, although data-privacy safeguards and bias mitigation remain priorities. During pregnancy, advanced deep-learning models can segment fetal anatomy on ultrasound images with more than 90 percent overlap compared to expert annotations and can detect anomalies with sensitivity exceeding 93 percent. Predictive biometric tools can estimate gestational age within one week with accuracy and fetal weight within approximately 190 g. In the postpartum period, AI-driven decision-support systems and conversational agents can facilitate early screening for depression and can guide follow-up care. Wearable sensors enable remote monitoring of maternal blood pressure and heart rate to support timely clinical intervention. Within neonatal care, the Heart Rate Observation (HeRO) system has reduced mortality among very low-birth-weight infants by roughly 20 percent, and additional AI models can predict neonatal sepsis, retinopathy of prematurity, and necrotizing enterocolitis with area-under-the-curve values above 0.80. From an operational standpoint, automated ultrasound workflows deliver biometric measurements at about 14 milliseconds per frame, and dynamic scheduling in IVF laboratories lowers staff workload and per-cycle costs. Home-monitoring platforms for pregnant women are associated with 7–11 percent reductions in maternal mortality and preeclampsia incidence. Despite these advances, most evidence derives from retrospective, single-center studies with limited external validation. Low-resource settings, especially in Sub-Saharan Africa, remain under-represented, and few AI solutions are fully embedded in electronic health records. Conclusions: AI holds transformative promise for perinatal care but will require prospective multicenter validation, equity-centered design, robust governance, transparent fairness audits, and seamless electronic health record integration to translate these innovations into routine practice and improve maternal and neonatal outcomes. Full article

Background/Objectives: Speech intelligibility (SI) is a key indicator of spoken language development, especially for children with hearing loss, as it directly impacts communication and social engagement. However, due to logistical and methodological challenges, SI assessment is often underutilised in clinical practice. This study aimed to evaluate the accuracy and consistency of an artificial intelligence (AI)-based transcription model in assessing SI in young children with cochlear implants (CI), hearing aids (HA), or normal hearing (NH), in comparison to naïve human listeners. Methods: A total of 580 speech samples from 58 five-year-old children were transcribed by three naïve listeners and the AI model. Word-level transcription accuracy was evaluated using Bland–Altman plots, intraclass correlation coefficients (ICCs), and word error rate (WER) metrics. Performance was compared across the CI, HA, and NH groups. Results: The AI model demonstrated high consistency with naïve listeners across all groups. Bland–Altman analyses revealed minimal bias, with fewer than 6% of sentences falling outside the 95% limits of agreement. ICC values exceeded 0.9 in all groups, with particularly strong agreement in the NH and CI groups (ICCs > 0.95). WER results further confirmed this alignment and indicated that children with CIs showed better SI performance than those using HAs. Conclusions: The AI-based method offers a reliable and objective solution for SI assessment in young children. Its agreement with human performance supports its integration into clinical and home environments for early intervention and ongoing monitoring of speech development in children with hearing loss. Full article

Introduction: Home-based cardiac telerehabilitation (HBCTR) is a multidisciplinary intervention aimed at optimizing functional, psychological, and social recovery in patients undergoing percutaneous coronary intervention (PCI). This rehabilitation model serves as an effective alternative to traditional center-based rehabilitation, providing a cost-effective and clinically advantageous approach. Methods: Following PRISMA guidelines, we conducted a systematic literature search across multiple databases (PubMed, CINAHL, Cochrane, Scopus, Web of Science). We included randomized controlled trials (RCTs), cohort, and observational studies assessing telerehabilitation in post-PCI patients. Primary outcomes focused on health-related quality of life (HRQoL) and adherence, while secondary outcomes included functional capacity (6 min walk test, VO₂max), cardiovascular risk factor control, and psychological well-being. Risk of bias was assessed using the Cochrane RoB 2.0 and ROBINS-I tools. Results: A total of 3575 articles were identified after removing duplicates, of which 877 were selected based on title and abstract, and 17 met the inclusion criteria, with strong RCT representation ensuring robust evidence synthesis. HBCTR was associated with significant improvements in exercise capacity, with increases in VO₂max ranging from +1.6 to +3.5 mL/kg/min and in 6 min walk distance from +34.7 to +116.6 m. HRQoL scores improved significantly, with physical and mental component scores increasing by +6.75 to +14.18 and +4.27 to +11.39 points, respectively. Adherence to telerehabilitation programs was consistently high, often exceeding 80%, and some studies reported reductions in hospital readmissions of up to 40%. Wearable devices and smartphone applications facilitated self-monitoring, enhancing adherence and reducing readmissions. Several studies also highlighted improvements in anxiety and depression scores ranging from 10% to 35%. Conclusions: HBCTR is a promising strategy for rehabilitation and quality-of-life improvement after PCI. It offers a patient-centered solution that leverages technology to enhance long-term outcomes. By integrating structured telerehabilitation programs, healthcare systems can expand accessibility, promote adherence, and improve equity in cardiovascular care. Full article

Chronic kidney disease (CKD) is associated with heart failure and neurological disorders. Therefore, point-of-care (POC) detection of CKD is essential, allowing disease monitoring from home and alleviating healthcare professionals’ workload. Lateral flow immunoassays (LFIAs) facilitate POC testing for a renal function biomarker, serum Cystatin C (CysC). LF devices were fabricated and optimised by varying the diluted sample volume, the nitrocellulose (NC) membrane, bed volume, AuNPs’ OD value and volume, and assay formats of partial or full LF systems. Notably, 310 samples were analysed to satisfy the minimum sample size for statistical calculations. This allowed for a comparison between the LFIAs’ results and the general Roche standard assay results from the Southern Health and Social Care Trust. Bland–Altman plots indicated the LFIAs measured 0.51 mg/L lower than the Roche assays. With the 95% confidence interval, the Roche method might be 0.24 mg/L below the LFIAs’ results or 1.27 mg/L above the LFIAs’ results. In summary, the developed non-fluorescent LFIAs could detect clinical CysC values in agreement with Roche assays. Even though the developed LFIA had an increased bias in low CysC concentration (below 2 mg/L) detection, the developed LFIA can still alert patients at the early stages of renal function impairment. Full article

Background: Trauma-related injuries are among the most common reasons for paediatric hospital presentations and represent a substantial component of orthopaedic care. Their management poses unique challenges due to ongoing skeletal development in children. While most reported fractures occur at home or during sports, prior studies have primarily used data from urban European populations, limiting the relevance of their findings for rural and regional settings. Urban-centred research often informs public healthcare guidelines, treatment algorithms, and infrastructure planning, introducing a bias when findings are generalised outside of metropolitan populations. This study addresses that gap by analysing fracture data from two rural trauma centres in New South Wales, Australia. This study assesses paediatric fractures resulting from severe injury mechanisms in rural areas, identifying common fracture types, underlying mechanisms, and treatment approaches to highlight differences in demographics. These findings aim to cast a light on healthcare challenges that regional areas face and to improve the overall cultural safety of children who live and grow up outside of the metropolitan trauma networks. Methods: We analysed data from two major rural referral hospitals in New South Wales (NSW) for paediatric injuries presenting between 1 January 2018 and 31 December 2022. This study included 150 patients presenting with fractures following severe mechanisms of injury, triaged into Australasian Triage Scale (ATS) categories 1 and 2 upon initial presentation. Results: A total of 150 severe fractures were identified, primarily affecting the upper and lower limbs. Males presented more frequently than females, and children aged 10–14 years old were most commonly affected. High-energy trauma from motorcycle (dirt bike) accidents was the leading mechanism of injury among all patients, and accounted for >50% of injuries among 10–14-year-old patients. The most common fractures sustained in these events were upper limb fractures, notably of the clavicle (n = 26, 17.3%) and combined radius/ulna fractures (n = 26, 17.3%). Conclusions: Paediatric trauma in regional Australia presents a unique and under-reported challenge, with high-energy injuries frequently linked to unregulated underage dirt bike use. Unlike urban centres where low-energy mechanisms dominate, rural areas require targeted prevention strategies. While most cases were appropriately managed locally, some were transferred to tertiary centres. These findings lay the groundwork for multi-centre research, and support the need for region-specific policy reform in the form of improved formal injury surveillance, injury prevention initiatives, and the regulation of under-aged off-road vehicular usage. Full article

Platform markets are concentrating, even as their content suppliers fragment, yet this asymmetric evolution is poorly understood. Using panel data from 11–12 Korean home shopping firms (2015–2023), we employ Hansen threshold regression, instrumental variables, and panel fixed-effects models to examine its competitive impact. Our analysis of 104 firm-year observations reveals four key findings. First, platform concentration alone explains 94.4% of transmission fee variation, with fees rising from 41.15% to 68.72% as platform HHI increased from 1390 to 2154 while content HHI declined from 1797 to 1118. Second, we identify critical fee thresholds at 62.2% (p = 0.012) and 73% (p = 0.002) that divide markets into three distinct operating regimes. Third, the fee–profitability relationship reversed from negative (r = −0.145) to positive (r = 0.554), indicating fees’ evolution from cost burdens to selection mechanisms. Fourth, instrumental variable estimates (0.473) exceed OLS estimates (0.184) by 2.6 times, revealing severe selection bias. Simulations indicate a 60% fee cap would affect 25 firms (24%) while increasing total surplus by 15.1% and improving SME profitability by 2.9 percentage points. We propose the Asymmetry Ratio (Platform HHI/Content HHI) as a regulatory tool, with ratios exceeding 1.0 triggering enhanced scrutiny. Our findings demonstrate that asymmetric market evolution creates new sources of platform power requiring novel regulatory approaches. Full article

Despite the ideal of a unified Single Market, a powerful “home bias” pervades EU public procurement, hinting at unseen barriers that conventional analysis fails to capture. This study introduces an interpretable AI framework to investigate these dynamics, pairing a LightGBM model with SHapley Additive exPlanations (SHAP) to examine the vast Tenders Electronic Daily (TED) database (2018–2023). Concretely, we propose a fuzzy linguistic layer that translates SHAP’s complex quantitative outputs into intuitive, human-readable terms. Our model effectively distinguishes local from non-local awards (AUC ≈ 0.855), revealing that while high-value contracts expectedly attract broader competition, the most potent predictors are a country’s own history of local awards and structural factors like the buyer’s type and location. This points not to isolated incidents, but, rather, to deep-seated patterns shaping market fairness. Our combined XAI-Fuzzy approach offers a new instrument for transparent governance, enabling policymakers to diagnose market realities and forge a more genuinely open and equitable European public square. Full article

In this study, we develop a graphene-trenched silicon Schottky junction for humidity sensing. This novel structure comprises suspended graphene bridging etched trenches on a silicon substrate, creating both free-standing and substrate-contacting regions of graphene that enhance water adsorption sensing. Suspended graphene is intrinsically insensitive to water adsorption, making it difficult for adsorbed H₂O to effectively dope the graphene. In contrast, when graphene is supported on the silicon substrate, water molecules can effectively dope the graphene by modifying the silicon’s impurity bands and their hybridization with graphene. This humidity-induced doping leads to a significant modulation of the Schottky barrier at the graphene–silicon interface, which serves as the core sensing mechanism. We investigate the current–voltage (I–V) characteristics of these devices as a function of trench width and relative humidity. Our analysis shows that humidity influences key device parameters, including the Schottky barrier height, ideality factor, series resistance, and normalized sensitivity. Specifically, larger trench widths reduce the graphene density of states, an effect that is accounted for in our analysis of these parameters. The sensor operates under both forward and reverse bias, enabling tunable sensitivity, high selectivity, and low power consumption. These features make it promising for applications in industrial and home safety, environmental monitoring, and process control. Full article

Background/Objectives: Patients recovering from COVID-19 often experience persistent respiratory symptoms, necessitating pulmonary function monitoring. While clinical spirometry is the gold standard, home spirometry offers a remote alternative. This study evaluated the validity of an ultrasonic home-based spirometer for monitoring lung function in post-COVID-19 pneumonia patients over 12 weeks. Methods: This prospective study included 30 post-COVID pneumonia patients who underwent clinical spirometry at weeks 4, 8 and 12. Participants performed weekly home spirometry using the SpiroHome Personal^® device. Agreement between home and clinical spirometry was assessed using a Bland–Altman analysis, intraclass correlation coefficients (ICCs), and Pearson correlation coefficients. Pulmonary function changes over time were analyzed using repeated measures ANOVA. Results: Home spirometry showed strong agreement with clinical spirometry for forced vital capacity (FVC) and forced expiratory volume in the first second (FEV1), with ICC values exceeding 0.92. The Bland–Altman analysis demonstrated minimal bias, though limits of agreement exceeded the clinically accepted threshold of ±150 mL. FEV1/FVC ratios showed greater variability. Pulmonary function improved significantly over 12 weeks for both methods (p < 0.002). Patient adherence to home spirometry remained high, with a median of 18.50 sessions [IQR: 15.00–26.00] and an overall compliance rate of 98.33% ± 9.13%. Conclusions: Home spirometry provides reliable pulmonary function measurements, particularly for FVC and FEV1, supporting its role as a remote monitoring tool. Despite minor variability in FEV1/FVC, home spirometry enables frequent assessment of lung function recovery, potentially reducing hospital visits and improving patient management. Full article

Background: Activities of Daily Living (ADLs) are essential tasks performed at home and used in healthcare to monitor sedentary behavior, track rehabilitation therapy, and monitor chronic obstructive pulmonary disease. The Barthel Index, used by healthcare professionals, has limitations due to its subjectivity. Human activity recognition (HAR) is a more accurate method using Information and Communication Technologies (ICTs) to assess ADLs more accurately. This work aims to create a singular, adaptable, and heterogeneous ADL dataset that integrates information from various sources, ensuring a rich representation of different individuals and environments. Methods: A literature review was conducted in Scopus, the University of California Irvine (UCI) Machine Learning Repository, Google Dataset Search, and the University of Cauca Repository to obtain datasets related to ADLs. Inclusion criteria were defined, and a list of dataset characteristics was made to integrate multiple datasets. Twenty-nine datasets were identified, including data from various accelerometers, gyroscopes, inclinometers, and heart rate monitors. These datasets were classified and analyzed from the review. Tasks such as dataset selection, categorization, analysis, cleaning, normalization, and data integration were performed. Results: The resulting unified dataset contained 238,990 samples, 56 activities, and 52 columns. The integrated dataset features a wealth of information from diverse individuals and environments, improving its adaptability for various applications. Conclusions: In particular, it can be used in various data science projects related to ADL and HAR, and due to the integration of diverse data sources, it is potentially useful in addressing bias in and improving the generalizability of machine learning models. Full article

Background: Degenerative lumbar disease (DLD) affects older adults, causing lumbar degeneration and lower extremity dysfunction. The five-times sit-to-stand test (5xSTS) reveals kinematic changes associated with DLD. While marker-based motion capture systems detect these changes, their complexity limits clinical use. Markerless motion capture offers a portable alternative, yet its functional assessment applications in DLD remain underexplored. Thus, the aim of this study is to evaluate the reliability and validity of markerless motion capture for assessing functional tests in DLD patients. Methods: This study included 11 healthy individuals (mean age: 27.28 ± 6.92 years) and 10 with DLD (mean age: 70.00 ± 8.08 years). Participants performed the 5xSTS while being recorded by marker-based (VICON) and markerless (MediaPipe) systems using two synchronized cameras. Test–retest reliability was assessed over one week via the intraclass correlation coefficient (ICC). Concurrent validity and agreement between VICON and MediaPipe were evaluated via Pearson/Spearman correlation coefficients, systematic bias, and the root mean square error (RMSE). Movement time, joint excursions, and angular velocities were also analyzed and compared across two groups. Results: Both systems showed high test–retest reliability (ICC: 0.81–0.99) and strong correlations (r: 0.75–0.99). The highest RMSE was observed at the ankle in the anterior–posterior (A–P) direction in the DLD group (54.55 mm) and at the hip A–P axis in the control group (51.20 mm). The lowest RMSE was found at the knee medial–lateral (M–L) axis in the DLD group (7.88 mm) and at the ankle M–L axis in the control group (8.54 mm). Bias values ranged from 0.30 mm (hip vertical in control group) to +53.47 mm (ankle A–P in DLD group), with underestimation more common at the hip and overestimation at the ankle. The control group demonstrated a faster 5xSTS completion time (5.89 ± 0.69 s vs. 8.13 ± 1.96 s, p < 0.05), greater hip joint excursions during sit-to-stand (65.07 ± 25.94° vs. 38.13 ± 9.84°, p < 0.05) and stand-to-sit (62.56 ± 24.74° vs. 27.85 ± 11.45°, p < 0.05) tests, and higher angular velocities compared to the DLD group. Conclusion: MediaPipe markerless motion capture can quantify 3D kinematic changes in DLD patients during functional performance. It enables a clinical evaluation with minimal setup, offers potential for remote assessment, and accurately detects sagittal plane movement. The two-camera system provides 3D kinematic data comparable to multi-camera systems, suitable for home rehabilitation and assessment. Full article

Background/Objectives: Urinary incontinence (UI) is particularly prevalent in women of all ages, but especially in older women, due to risk factors that contribute to pelvic floor muscle weakness. Urinary incontinence can have a significant psychosocial impact and compromise the quality of life of affected women. It is reported that physical activity seems to be inversely associated with UI. The aim is to summarize the impact of different modalities of physical exercises in management of older women with UI. Methods: Searches were performed in the databases PubMed, Web of Science, EMBASE and Scopus. The searches were performed on 10 December 2024. Only randomized clinical trials were included. Ninety-three papers were initially screened, and five full papers reached all the inclusion criteria describing the effects of exercise on urinary incontinence. The studies included a total of 352 elderly participants aged 60 years or older. Of these participants, 268 lived in nursing homes and 84 lived in a local community. The intervention period varied among the selected studies from 6 weeks to 12 months, and in general, the pelvic floor training was used. Results: Improvements in the quality of life and in the physical performance of the older women with UI were reported, although the risk of bias was classified as “some concerns”. Conclusions: Moreover, it is revealed that different types of exercises are used to “work” the pelvic floor muscle and contribute, in general, to reducing the symptoms, and improving the quality of life of older women with UI due to the positive impact of the exercises. Full article

Background/Objectives: In 2010, approximately 60.5 million people were affected by glaucoma, making it the leading cause of permanent vision impairment globally. With the rise of telehealth tools and technological advancements in glaucoma care, this review aims to provide an up-to-date analysis regarding remote monitoring systems in glaucoma management. Methods: A systematic literature search (in compliance with PRISMA guidelines) was conducted across six databases (CINAHL, MEDLINE, PsycINFO, Web of Science, Scopus, and Cochrane Library) and one grey literature source (Google Scholar), covering the period from 2000 to 2024. Relevant studies meeting predefined inclusion criteria were identified and analyzed. Results: The search identified 21 eligible studies focusing on various glaucoma telemonitoring tools. Several studies demonstrated the potential for continuous remote intraocular pressure (IOP) monitoring and highlighted the effectiveness of home-based visual field-testing technologies (e.g., Melbourne Rapid Fields, Eyecatcher, and VF-Home), which showed results closely matching in-clinic tests. All 21 studies underwent risk of bias assessment with appropriate tools based on study design, and none showed a high overall risk of bias, indicating robust methodology. Conclusions: Glaucoma telemonitoring tools are feasible and cost-effective, helping to reduce patient travel and waiting times and improving patient satisfaction. However, periodic in-person examinations remain necessary to optimally monitor disease progression and adjust treatments. Future directions should focus on interdisciplinary collaboration and the development of advanced algorithms (including artificial intelligence) to further enhance patient outcomes in teleglaucoma care. Full article

Building on the work–home resources model, this study develops a mediated moderation model to examine how benevolent sexism within intimate relationships influences female employees’ thriving at work through job resources and self-efficacy. In this cross-sectional study, we surveyed 209 married female employees from six Chinese public hospitals and their spouses via online questionnaires between April and July 2023. To mitigate common method bias, we implemented a two-wave data collection approach with a one-month interval. We employed confirmatory factor analysis, Pearson’s correlation, and multiple regression to test the hypothesized relationships. The results support the proposed model, indicating that benevolent sexism within intimate relationships moderates the positive effect of job resources on thriving at work, such that this relationship weakens when benevolent sexism within intimate relationships is high. Furthermore, this moderating effect is mediated by self-efficacy, as benevolent sexism within intimate relationships undermines female employees’ self-efficacy, thereby limiting their ability to leverage job resources effectively. These findings highlight the need for human resource managers to consider the personal circumstances of female employees and implement targeted interventions to facilitate their optimal utilization of job resources, thereby enhancing their ability to thrive in both professional and domestic domains. Full article

Artificial Intelligence (AI) is rapidly transforming orthodontic care by providing personalized treatment plans that enhance precision and efficiency. This narrative review explores the current applications of AI in orthodontics, particularly its role in predicting tooth movement, fabricating custom aligners, optimizing treatment times, and offering real-time patient monitoring. AI’s ability to analyze large datasets of dental records, X-rays, and 3D scans allows for highly individualized treatment plans, improving both clinical outcomes and patient satisfaction. AI-driven aligners and braces are designed to apply optimal forces to teeth, reducing treatment time and discomfort. Additionally, AI-powered remote monitoring tools enable patients to check their progress from home, decreasing the need for in-person visits and making orthodontic care more accessible. The review also highlights future prospects, such as the integration of AI with robotics for performing orthodontic procedures, predictive orthodontics for early intervention, and the use of 3D printing technologies to fabricate orthodontic devices in real-time. While AI offers tremendous potential, challenges remain in areas such as data privacy, algorithmic bias, and the cost of adopting AI technologies. However, as AI continues to evolve, its capacity to revolutionize orthodontic care will likely lead to more streamlined, patient-centered, and effective treatments. This review underscores the transformative role of AI in modern orthodontics and its promising future in advancing dental care. Full article