Frontiers in digital health最新文献

Background: mHealth has increasingly been touted as having the potential to help Sub-Saharan Africa achieve their health-related sustainable development goals by reducing maternal mortality rates. Such interventions are implemented as one-way or two-way systems where maternal clients receive pregnancy related information via SMS. While such technologies often view the users (the maternal health client) as having agency to adopt, we know from pregnancy literature that the pregnancy experience in Africa and other developing countries is often more collective. In addition to the maternal health client, other members of the community have high stakes in the pregnancy, and this often affects maternal healthcare-seeking behavior.

Objective: The aim of this paper, therefore, is to understand the pathways through which these other members of the community affect mHealth use.

Methods: The study used a qualitative approach and a case study research design. We analyzed two mHealth cases from Kenya and Malawi. In the Kenyan case, maternal health clients had mobile phones to receive pregnancy-related messages, while in the Malawi case, maternal health clients did not have mobile phones. Data were collected through interviews and focus group discussions. The study used an inductive thematic analysis to analyze the data.

Results: The findings show that maternal stakeholders form a community of purpose (CoP) that plays a crucial role in the implementation, uptake, and use of mHealth. The CoP influences maternal health clients through a diverse range of mechanisms ranging from sensitization, bridging the digital literacy gap and legitimization of the intervention. The nature of influence is largely dependent on the contextual socio-cultural nuances.

Conclusion: Our results provide useful insights to mHealth implementers to know how best to leverage the CoP for better mHealth uptake and usage. For example, engaging healthcare providers could champion adoption and use, while engaging other family-related stakeholders will ensure better usage and compliance, encourage behavior change, and reduce mHealth attrition.

Introduction: Internet addiction refers to the excessive and uncontrolled utilization of the Internet, which disrupts one's daily activities. The current state of knowledge regarding internet addiction in Ethiopia is limited. Consequently, the objective of this study is to ascertain the combined prevalence of Internet addiction and its correlated factors among university students in Ethiopia.

Methods: To identify potential research findings, an extensive literature search was conducted using electronic databases such as PubMed/MEDLINE, Web of Science, and Google Scholar. The presence of heterogeneity between studies was assessed using Cochrane Q test statistics and I2 test statistics, while the effects of small studies were examined using Eggers statistical tests at a 5% significance level. Additionally, a sensitivity analysis was carried out. A random effects model was used to estimate the pooled prevalence and associated factors of Internet addiction among students. The primary focus of this research was to determine the prevalence of Internet addiction, while the secondary aim was to identify the factors associated with Internet addiction.

Results: To determine the overall prevalence of Internet addiction among university students in Ethiopia, a comprehensive analysis of 11 studies was conducted. The results of this study show that the pooled prevalence of Internet addiction was 43.42% (95% CI: 28.54, 58.31). The results also suggest that certain factors such as online gaming, depression, and current khat chewing are significantly associated with internet addiction among university students.

Conclusions: In Ethiopia, about one-third of university students suffer from internet addiction. The prevalence of Internet addiction among Ethiopian students is associated with online gaming, depression, and concurrent khat consumption. Therefore, we strongly recommend that health planners and policymakers prioritize monitoring and addressing Internet use and addiction in the Ethiopian context.

Introduction: Endometriosis (EMs) and adenomyosis (AD) are common gynecological diseases that impact women's health, and they share symptoms such as dysmenorrhea, chronic pain, and infertility, which adversely affect women's quality of life. Current diagnostic approaches for EMs and AD involve invasive surgical procedures, and thus, methods of noninvasive differentiation between EMs and AD are needed. This retrospective cohort study introduces a novel, noninvasive classification methodology employing a stacked ensemble machine learning (ML) model that utilizes peripheral blood and coagulation markers to distinguish between EMs and AD.

Methods: The study included a total of 558 patients (329 with EMs and 229 with AD), in whom key hematological and coagulation markers were analyzed to identify distinctive profiles. Feature selection was conducted through ML (logistic regression, support vector machine, and K-nearest neighbors) to determine significant hematological markers.

Results: Red cell distribution width, mean corpuscular hemoglobin concentration, activated partial thromboplastin time, international normalized ratio, and antithrombin III were proved to be the key distinguishing indexes for disease differentiation. Among all the ML classification models developed, the stacked ensemble model demonstrated superior performance (area under the curve = 0.803, 95% credibility interval = 0.701-0.904). Our findings demonstrate the effectiveness of the stacked ensemble ML model for classifying EMs and AD.

Discussion: Integrating biomarkers into this multi-algorithm framework offers a novel approach to noninvasive diagnosis. These results advocate for the application of stacked ensemble ML utilizing cost-effective and readily available peripheral blood and coagulation indicators for the early, rapid, and noninvasive differential diagnosis of EMs and AD, offering a potentially transformative approach for clinical decision-making and personalized treatment strategies.

Background: Consumer facing wearable devices capture significant amounts of biometric data. The primary aim of this study is to determine the accuracy of consumer-facing wearable technology for continuous monitoring compared to standard anesthesia monitoring during endoscopic procedures. Secondary aims were to assess patient and provider perceptions of these devices in clinical settings.

Methods: Patients undergoing endoscopy with anesthesia support from June 2021 to June 2022 were provided a smartwatch (Apple Watch Series 7, Apple Inc., Cupertino, CA) and accessories including continuous ECG monitor and pulse oximeter (Qardio Inc., San Francisco, CA) for the duration of their procedure. Vital sign data from the wearable devices was compared to in-room anesthesia monitors. Concordance with anesthesia monitoring was assessed with interclass correlation coefficients (ICC). Surveys were then distributed to patients and clinicians to assess patient and provider preferences regarding the use of the wearable devices during procedures.

Results: 292 unique procedures were enrolled with a median anesthesia duration of 34 min (IQR 25-47). High fidelity readings were successfully recorded with wearable devices for heart rate in 279 (95.5%) cases, oxygen in 203 (69.5%), and respiratory rate in 154 (52.7%). ICCs for watch and accessories were 0.54 (95% CI 0.46-0.62) for tachycardia, 0.03 (95% CI 0-0.14) for bradycardia, and 0.33 (0.22-0.43) for oxygen desaturation. Patients generally felt the devices were more accurate (56.3% vs. 20.0% agree, p < 0.001) and more permissible (53.9% vs. 33.3% agree, p < 0.001) to wear during a procedure than providers.

Conclusion: Smartwatches perform poorly for continuous data collection compared to gold standard anesthesia monitoring. Refinement in software development is required if these devices are to be used for continuous, intensive vital sign monitoring.

Background: Artificial intelligence (AI) is transforming healthcare, yet little is known about Chinese oncologists' attitudes towards AI. This study investigated oncologists' knowledge, perceptions, and acceptance of AI in China.

Methods: A cross-sectional online survey was conducted among 228 oncologists across China. The survey examined demographics, AI exposure, knowledge and attitudes using 5-point Likert scales, and factors influencing AI adoption. Data were analyzed using descriptive statistics and chi-square tests.

Results: Respondents showed moderate understanding of AI concepts (mean 3.39/5), with higher knowledge among younger oncologists. Only 12.8% used ChatGPT. Most (74.13%) agreed AI is beneficial and could innovate healthcare, 52.19% respondents expressed trust in AI technology. Acceptance was cautiously optimistic (mean 3.57/5). Younger respondents (∼30) show significantly higher trust (p = 0.004) and acceptance (p = 0.009) of AI compared to older respondents, while trust is significantly higher among those with master's or doctorate vs. bachelor's degrees (p = 0.032), and acceptance is higher for those with prior IT experience (p = 0.035).Key drivers for AI adoption were improving efficiency (85.09%), quality (85.53%), reducing errors (84.65%), and enabling new approaches (73.25%).

Conclusions: Chinese oncologists are open to healthcare AI but remain prudently optimistic given limitations. Targeted education, especially for older oncologists, can facilitate AI implementation. AI is largely welcomed for its potential to augment human roles in enhancing efficiency, quality, safety, and innovations in oncology practice.

Background: In South Africa, between 1966 and 2014, there were three kidney transplant eras defined by evolving access to certain immunosuppressive therapies defined as Pre-CYA (before availability of cyclosporine), CYA (when cyclosporine became available), and New-Gen (availability of tacrolimus and mycophenolic acid). As such, factors influencing kidney graft failure may vary across these eras. Therefore, evaluating the consistency and reproducibility of models developed to study these variations using machine learning (ML) algorithms could enhance our understanding of post-transplant graft survival dynamics across these three eras.

Methods: This study explored the effectiveness of nine ML algorithms in predicting 10-year graft survival across the three eras. We developed and internally validated these algorithms using data spanning the specified eras. The predictive performance of these algorithms was assessed using the area under the curve (AUC) of the receiver operating characteristics curve (ROC), supported by other evaluation metrics. We employed local interpretable model-agnostic explanations to provide detailed interpretations of individual model predictions and used permutation importance to assess global feature importance across each era.

Results: Overall, the proportion of graft failure decreased from 41.5% in the Pre-CYA era to 15.1% in the New-Gen era. Our best-performing model across the three eras demonstrated high predictive accuracy. Notably, the ensemble models, particularly the Extra Trees model, emerged as standout performers, consistently achieving high AUC scores of 0.95, 0.95, and 0.97 across the eras. This indicates that the models achieved high consistency and reproducibility in predicting graft survival outcomes. Among the features evaluated, recipient age and donor age were the only features consistently influencing graft failure throughout these eras, while features such as glomerular filtration rate and recipient ethnicity showed high importance in specific eras, resulting in relatively poor historical transportability of the best model.

Conclusions: Our study emphasises the significance of analysing post-kidney transplant outcomes and identifying era-specific factors mitigating graft failure. The proposed framework can serve as a foundation for future research and assist physicians in identifying patients at risk of graft failure.

Although the rapid growth in the efficiency of medical imaging is undeniable, the expansion of health information technology (HIT) into medical imaging has not been as seamless or well-integrated as it was thought to be. The socio-technical complexities in medical imaging associated with HIT systems can cause risks to patient harm and inconvenience, both individually and collectively, often in new, unforeseen, and unexpected ways. This study reflects a retrospectively collected single incident report related to medical imaging HIT systems, aiming to develop a set of preventive and corrective strategies. A combination of multiple deductive approaches (existing frameworks), i.e., HIT Classification Systems and 18-step medical imaging process workflow and inductive method (content analysis), were used to analyze the incident. The incident was identified as a "system configuration"-related software issue, contributed by system upgrade- changes in hardware and software. The incident was determined to occur during steps 10-12, i.e., "study selection and retrieval," "calling up of patient's referral," and "image review and interpretation," causing severe disruptions in the clinical workflow for several weeks. We propose 16 preventive and corrective strategies grouped under four key areas based on the socio-technical aspects associated with HIT systems. The key areas are (i) preparation and integration for upgraded systems, (ii) training for medical imaging specialists, (iii) contingency planning/immediate backup system, and (iv) system design and configuration. These strategies are expected to help healthcare staff, analysts, reporters, researchers, and relevant stakeholders improve care delivery and patient safety in medical imaging in the context of any system upgrades.

Aim: A clinical study was performed to test the usability of a smartphone eye-scanning app at a needle exchange facility to detect drug use to support therapy.

Methods: The study recruited 24 subjects who visited the facility one to three times, making a total of 40 visits. During each visit the subjects underwent testing for non-convergence (NC), nystagmus (NY), and pupillary light reflex (PLR) using a smartphone-based eHealth system. The collected eye data were transformed into key features that represent eye characteristics. During each visit, a time-line follow-back interview on recent drug use and a usability questionnaire were completed.

Results: Technical usability of the smartphone eye-scanning app was good for PLR and NC, where key features were generated in 82%-91% of the cases. For NY, only 60% succeeded due to cognitive problems to follow instructions. In most cases, subjects were under the influence of drugs when participating in the tests, with an average of 2.4 different drugs ingested within the last 24 h. The key features from PLR could distinguish use of opioids from central stimulants. The usability questionnaire results indicate that 23 of the 24 subjects could perform the eye-scanning by themselves after a short training, even when under severe influence of drugs. The caregiver assessed that 20 out of the 24 challenging subjects could potentially perform these tests in an indoors, home-like environment.

Conclusions: Smartphone-based eye-scanning is functional in a patient population with heavy drug use, also when under the influence of drugs. The use of central stimulants can be distinguished from the use of opioids.

The extraordinary growth of health technologies has determined an increasing interest in biobanks that represent a unique wealth for research, experimentation, and validation of new therapies. "Human" biobanks are repositories of various types of human biological samples. Through years the paradigm has shifted from spontaneous collections of biological material all over the world to institutional, organized, and well-structured forms. Imaging biobanks represent a novel field and are defined by European Society of Radiology as: "organized databases of medical images, and associated imaging biomarkers shared among multiple researchers, linked to other biorepositories". Modern radiology and nuclear medicine can provide multiple imaging biomarkers, that express the phenotype related to certain diseases, especially in oncology. Imaging biobanks, not a mere catalogue of bioimages associated to clinical data, involve advanced computer technologies to implement the emergent field of radiomics and radiogenomics. Since Europe hosts most of the biobanks, juridical and ethical framework, with a specific referral to Italy, is analyzed. Linking imaging biobanks to traditional ones appears to be a crucial step that needs to be driven by medical imaging community under clear juridical and ethical guidelines.

Background: Despite growing international attention, there remains an urgent need to develop mental health services within low and middle income countries. The Khmer Rouge period in Cambodia saw the destruction of all health services infrastructure in the 1970s. Consequently, Cambodia has struggled to rebuild both its economy and healthcare system, with the number of qualified mental health clinicians remaining disproportionately low. Resultantly, there is a pressing need to develop low-cost community based alternatives of mental healthcare.

Methods: Using a mixed methods design, researchers developed an 8-week peer-led intervention, known as a Friendship Group, for adults with physical disabilities using both face-to-face and online delivery methods. The Wilcoxon Signed-Rank test was used to assess changes in pre-post survey scores and qualitative data was collected in form of five focus groups post intervention.

Results: 41 participants were allocated across four Friendship groups - two were online and two face-to-face. Attrition rate was 22% post-intervention (n = 32). ITT analyses showed a statistically significant decrease in psychological distress scores [Z = -3.808, p < .001] from pre [Mdn = 20, IQR = 16.5-25.5] to post [Mdn = 16, IQR = 14-18.5] intervention. A Wilcoxon signed-ranks test also showed a statistically significant decrease in PTSD scores [Z = -2.239, p < .025] from pre [Mdn = 4, IQR = 3-5] to post [Mdn = 3, IQR = 2.75-4] intervention. There was also a statistically significant decrease in worry scores [Z = -3.904, p < .001] from pre [Mdn = 5, IQR = 3.5-6.5] to post [Mdn = 3, IQR = 3-4] intervention. There were no significant group differences between the face to face and online groups. A number of interconnected themes emerged from focus group data (n = 5), these included the mental health benefits of Friendship Groups as conceptualised through knowledge acquisition, skill development and peer support.

Conclusions: The Friendship group intervention delivered in both online and face-to-face formats appears feasible and acceptable within the Cambodian context. Initial data revealed positive findings in terms of reduction in psychological distress, worry and PTSD symptoms as well increased feeling as calm.