Generalizability assessment of AI models across hospitals in a low-middle and high income country.

The integration of artificial intelligence (AI) into healthcare systems within low-middle income countries (LMICs) has emerged as a central focus for various initiatives aiming to improve healthcare access and delivery quality. In contrast to high-income countries (HICs), which often possess the resources and infrastructure to adopt innovative healthcare technologies, LMICs confront resource limitations such as insufficient funding, outdated infrastructure, limited digital data, and a shortage of technical expertise. Consequently, many algorithms initially trained on data from non-LMIC settings are now being employed in LMIC contexts. However, the effectiveness of these systems in LMICs can be compromised when the unique local contexts and requirements are not adequately considered. In this study, we evaluate the feasibility of utilizing models developed in the United Kingdom (a HIC) within hospitals in Vietnam (a LMIC). Consequently, we present and discuss practical methodologies aimed at improving model performance, emphasizing the critical importance of tailoring solutions to the distinct healthcare systems found in LMICs. Our findings emphasize the necessity for collaborative initiatives and solutions that are sensitive to the local context in order to effectively tackle the healthcare challenges that are unique to these regions.

Emerging Monkeypox Virus Sublineage C.1 Causing Community Transmission, Vietnam, 2023.

We studied a community cluster of 25 mpox cases in Vietnam caused by emerging monkeypox virus sublineage C.1 and imported into Vietnam through 2 independent events; 1 major cluster carried a novel APOBEC3-like mutation. Three patients died; all had advanced HIV co-infection. Viral evolution and its potential consequences should be closely monitored.

Mitigating machine learning bias between high income and low-middle income countries for enhanced model fairness and generalizability.

Collaborative efforts in artificial intelligence (AI) are increasingly common between high-income countries (HICs) and low- to middle-income countries (LMICs). Given the resource limitations often encountered by LMICs, collaboration becomes crucial for pooling resources, expertise, and knowledge. Despite the apparent advantages, ensuring the fairness and equity of these collaborative models is essential, especially considering the distinct differences between LMIC and HIC hospitals. In this study, we show that collaborative AI approaches can lead to divergent performance outcomes across HIC and LMIC settings, particularly in the presence of data imbalances. Through a real-world COVID-19 screening case study, we demonstrate that implementing algorithmic-level bias mitigation methods significantly improves outcome fairness between HIC and LMIC sites while maintaining high diagnostic sensitivity. We compare our results against previous benchmarks, utilizing datasets from four independent United Kingdom Hospitals and one Vietnamese hospital, representing HIC and LMIC settings, respectively.

Understanding maternal sepsis risk factors and bacterial etiology: A case control study protocol.

INTRODUCTION: Maternal disorders are the third leading cause of sepsis globally, accounting for 5.7 million (12%) cases in 2017. There are increasing concerns about the emergence of antimicrobial resistance (AMR) in bacteria commonly causing maternal sepsis. Our aim is to describe the protocol for a clinical and microbiology laboratory study to understand risk factors for and the bacterial etiology of maternal sepsis in a tertiary Obstetrics and Gynaecology Hospital. METHODS: This case-control study aims to recruit 100 cases and 200 controls at Tu Du Hospital in Ho Chi Minh City, Vietnam, which had approximately 55,000 births in 2022. Women aged ≥ 18 years and ≥ 28 weeks gestation having a singleton birth will be eligible for inclusion as cases or controls, unless they have an uncomplicated localised or chronic infection, or an infection with SARS-CoV-2. Cases will include pregnant or recently pregnant women with sepsis recognised between the onset of labour and/or time of delivery/cessation of pregnancy for up to 42 days post-partum. Sepsis will be defined as suspected or confirmed infection with an obstetrically modified Sequential Organ Failure Assessment score of ≥ 2, treatment with intravenous antimicrobials and requested cultures of any bodily fluid. Controls will be matched by age, location, parity, mode of delivery and gestational age. Primary and secondary outcomes are risk factors associated with the development of maternal sepsis, the frequency of adverse outcomes due to maternal sepsis, bacterial etiology and AMR profiles of cases and controls. DISCUSSION: This study will improve understanding of the epidemiology and clinical implications of maternal sepsis management including the presence of AMR in women giving birth in Vietnam. It will help us to determine whether women in this setting are receiving optimal care and to identify opportunities for improvement.

Clinical evaluation of AI-assisted muscle ultrasound for monitoring muscle wasting in ICU patients.

Muscle ultrasound has been shown to be a valid and safe imaging modality to assess muscle wasting in critically ill patients in the intensive care unit (ICU). This typically involves manual delineation to measure the rectus femoris cross-sectional area (RFCSA), which is a subjective, time-consuming, and laborious task that requires significant expertise. We aimed to develop and evaluate an AI tool that performs automated recognition and measurement of RFCSA to support non-expert operators in measurement of the RFCSA using muscle ultrasound. Twenty patients were recruited between Feb 2023 and July 2023 and were randomized sequentially to operators using AI (n = 10) or non-AI (n = 10). Muscle loss during ICU stay was similar for both methods: 26 ± 15% for AI and 23 ± 11% for the non-AI, respectively (p = 0.13). In total 59 ultrasound examinations were carried out (30 without AI and 29 with AI). When assisted by our AI tool, the operators showed less variability between measurements with higher intraclass correlation coefficients (ICCs 0.999 95% CI 0.998-0.999 vs. 0.982 95% CI 0.962-0.993) and lower Bland Altman limits of agreement (± 1.9% vs. ± 6.6%) compared to not using the AI tool. The time spent on scans reduced significantly from a median of 19.6 min (IQR 16.9-21.7) to 9.4 min (IQR 7.2-11.7) compared to when using the AI tool (p < 0.001). AI-assisted muscle ultrasound removes the need for manual tracing, increases reproducibility and saves time. This system may aid monitoring muscle size in ICU patients assisting rehabilitation programmes.

Towards a machine-learning assisted non-invasive classification of dengue severity using wearable PPG data: a prospective clinical study.

BACKGROUND: Dengue epidemics impose considerable strain on healthcare resources. Real-time continuous and non-invasive monitoring of patients admitted to the hospital could lead to improved care and outcomes. We evaluated the performance of a commercially available wearable (SmartCare) utilising photoplethysmography (PPG) to stratify clinical risk for a cohort of hospitalised patients with dengue in Vietnam. METHODS: We performed a prospective observational study for adult and paediatric patients with a clinical diagnosis of dengue at the Hospital for Tropical Disease, Ho Chi Minh City, Vietnam. Patients underwent PPG monitoring early during admission alongside standard clinical care. PPG waveforms were analysed using machine learning models. Adult patients were classified between 3 severity classes: i) uncomplicated (ward-based), ii) moderate-severe (emergency department-based), and iii) severe (ICU-based). Data from paediatric patients were split into 2 classes: i) severe (during ICU stay) and ii) follow-up (14-21 days after the illness onset). Model performances were evaluated using standard classification metrics and 5-fold stratified cross-validation. FINDINGS: We included PPG and clinical data from 132 adults and 15 paediatric patients with a median age of 28 (IQR, 21-35) and 12 (IQR, 9-13) years respectively. 1781 h of PPG data were available for analysis. The best performing convolutional neural network models (CNN) achieved a precision of 0.785 and recall of 0.771 in classifying adult patients according to severity class and a precision of 0.891 and recall of 0.891 in classifying between disease and post-disease state in paediatric patients. INTERPRETATION: We demonstrate that the use of a low-cost wearable provided clinically actionable data to differentiate between patients with dengue of varying severity. Continuous monitoring and connectivity to early warning systems could significantly benefit clinical care in dengue, particularly within an endemic setting. Work is currently underway to implement these models for dynamic risk predictions and assist in individualised patient care. FUNDING: EPSRC Centre for Doctoral Training in High-Performance Embedded and Distributed Systems (HiPEDS) (Grant: EP/L016796/1) and the Wellcome Trust (Grants: 215010/Z/18/Z and 215688/Z/19/Z).

Endothelial and inflammatory pathophysiology in dengue shock: New insights from a prospective cohort study in Vietnam.

Dengue shock (DS) is the most severe complication of dengue infection; endothelial hyperpermeability leads to profound plasma leakage, hypovolaemia and extravascular fluid accumulation. At present, the only treatment is supportive with intravenous fluid, but targeted endothelial stabilising therapies and host immune modulators are needed. With the aim of prioritising potential therapeutics, we conducted a prospective observational study of adults (≥16 years) with DS in Vietnam from 2019-2022, comparing the pathophysiology underlying circulatory failure with patients with septic shock (SS), and investigating the association of biomarkers with clinical severity (SOFA score, ICU admission, mortality) and pulmonary vascular leak (daily lung ultrasound for interstitial and pleural fluid). Plasma was collected at enrolment, 48 hours later and hospital discharge. We measured biomarkers of inflammation (IL-6, ferritin), endothelial activation (Ang-1, Ang-2, sTie-2, VCAM-1) and endothelial glycocalyx breakdown (hyaluronan, heparan sulfate, endocan, syndecan-1). We enrolled 135 patients with DS (median age 26, median SOFA score 7, 34 required ICU admission, 5 deaths), together with 37 patients with SS and 25 healthy controls. Within the DS group, IL-6 and ferritin were associated with admission SOFA score (IL-6: ßeta0.70, p<0.001 & ferritin: ßeta0.45, p<0.001), ICU admission (IL-6: OR 2.6, p<0.001 & ferritin: OR 1.55, p<0.001) and mortality (IL-6: OR 4.49, p = 0.005 & ferritin: OR 13.8, p = 0.02); both biomarkers discriminated survivors and non-survivors at 48 hours and all patients who died from DS had pre-mortem ferritin ≥100,000ng/ml. IL-6 most strongly correlated with severity of pulmonary vascular leakage (R = 0.41, p<0.001). Ang-2 correlated with pulmonary vascular leak (R = 0.33, p<0.001) and associated with SOFA score (ß 0.81, p<0.001) and mortality (OR 8.06, p = 0.002). Ang-1 was associated with ICU admission (OR 1.6, p = 0.005) and mortality (OR 3.62, p = 0.006). All 4 glycocalyx biomarkers were positively associated with SOFA score, but only syndecan-1 was associated with ICU admission (OR 2.02, p<0.001) and mortality (OR 6.51, p<0.001). This study highlights the central role of hyperinflammation in determining outcomes from DS; the data suggest that anti-IL-1 and anti-IL-6 immune modulators and Tie2 agonists may be considered as candidates for therapeutic trials in severe dengue.

Development and validation of multiplex real-time PCR for simultaneous detection of six bacterial pathogens causing lower respiratory tract infections and antimicrobial resistance genes.

BACKGROUND: Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, Escherichia coli, Streptococcus pneumoniae and Staphylococcus aureus are major bacterial causes of lower respiratory tract infections (LRTIs) globally, leading to substantial morbidity and mortality. The rapid increase of antimicrobial resistance (AMR) in these pathogens poses significant challenges for their effective antibiotic therapy. In low-resourced settings, patients with LRTIs are prescribed antibiotics empirically while awaiting several days for culture results. Rapid pathogen and AMR gene detection could prompt optimal antibiotic use and improve outcomes. METHODS: Here, we developed multiplex quantitative real-time PCR using EvaGreen dye and melting curve analysis to rapidly identify six major pathogens and fourteen AMR genes directly from respiratory samples. The reproducibility, linearity, limit of detection (LOD) of real-time PCR assays for pathogen detection were evaluated using DNA control mixes and spiked tracheal aspirate. The performance of RT-PCR assays was subsequently compared with the gold standard, conventional culture on 50 tracheal aspirate and sputum specimens of ICU patients. RESULTS: The sensitivity of RT-PCR assays was 100% for K. pneumoniae, A. baumannii, P. aeruginosa, E. coli and 63.6% for S. aureus and the specificity ranged from 87.5% to 97.6%. The kappa correlation values of all pathogens between the two methods varied from 0.63 to 0.95. The limit of detection of target bacteria was 1600 CFU/ml. The quantitative results from the PCR assays demonstrated 100% concordance with quantitative culture of tracheal aspirates. Compared to culture, PCR assays exhibited higher sensitivity in detecting mixed infections and S. pneumoniae. There was a high level of concordance between the detection of AMR gene and AMR phenotype in single infections. CONCLUSIONS: Our multiplex quantitative RT-PCR assays are fast and simple, but sensitive and specific in detecting six bacterial pathogens of LRTIs and their antimicrobial resistance genes and should be further evaluated for clinical utility.

ACORN (A Clinically-Oriented Antimicrobial Resistance Surveillance Network) II: protocol for case based antimicrobial resistance surveillance.

Background: Antimicrobial resistance surveillance is essential for empiric antibiotic prescribing, infection prevention and control policies and to drive novel antibiotic discovery. However, most existing surveillance systems are isolate-based without supporting patient-based clinical data, and not widely implemented especially in low- and middle-income countries (LMICs). Methods: A Clinically-Oriented Antimicrobial Resistance Surveillance Network (ACORN) II is a large-scale multicentre protocol which builds on the WHO Global Antimicrobial Resistance and Use Surveillance System to estimate syndromic and pathogen outcomes along with associated health economic costs. ACORN-healthcare associated infection (ACORN-HAI) is an extension study which focuses on healthcare-associated bloodstream infections and ventilator-associated pneumonia. Our main aim is to implement an efficient clinically-oriented antimicrobial resistance surveillance system, which can be incorporated as part of routine workflow in hospitals in LMICs. These surveillance systems include hospitalised patients of any age with clinically compatible acute community-acquired or healthcare-associated bacterial infection syndromes, and who were prescribed parenteral antibiotics. Diagnostic stewardship activities will be implemented to optimise microbiology culture specimen collection practices. Basic patient characteristics, clinician diagnosis, empiric treatment, infection severity and risk factors for HAI are recorded on enrolment and during 28-day follow-up. An R Shiny application can be used offline and online for merging clinical and microbiology data, and generating collated reports to inform local antibiotic stewardship and infection control policies. Discussion: ACORN II is a comprehensive antimicrobial resistance surveillance activity which advocates pragmatic implementation and prioritises improving local diagnostic and antibiotic prescribing practices through patient-centred data collection. These data can be rapidly communicated to local physicians and infection prevention and control teams. Relative ease of data collection promotes sustainability and maximises participation and scalability. With ACORN-HAI as an example, ACORN II has the capacity to accommodate extensions to investigate further specific questions of interest.

Magnitude, Patterns, and Associated Predictors of Cardiovascular Events in Tetanus: A 2-Year, Single-Center, Ambidirectional Cohort Study Involving 572 Patients.

Background: Cardiovascular events (CEs) remain the leading cause of death in patients with tetanus. We examined the incidence, patterns, and associated predictors of CEs among patients with tetanus in Vietnam. Methods: An ambidirectional cohort study was conducted on hospitalized adult patients with tetanus at the Hospital for Tropical Diseases between 2019 and 2020. Information on demographics, tetanus disease, CEs and outcomes were collected. Results: Among all 572 included patients, CEs accounted for 10.8% (95%CI 8.6-13.7%) and included Takotsubo cardiomyopathy (40.3%, 95%CI 29.0-52.8%), arrhythmia (19.4%, 95%CI 11.4-30.9%), sudden cardiac arrest (16.1%, 95%CI 9.0-27.2%), myocardial infarction (11.3%, 95%CI 5.6-21.5%), heart failure (6.5%, 95%CI 2.5-15.4%) and pulmonary embolism (6.5%, 95%CI 2.5-15.4%). CEs occurred from day 5 to 20 of illness. Among 62 CE patients, 21% (95%CI 12.7-32.6%) died and 61.3% (95%CI 48.9-72.4%) developed autonomic nervous system dysfunction (ANSD). Three-fourths (24/32) of patients with Takotsubo cardiomyopathy or myocardial infarction had ANSD. CEs were significantly associated with modified Ablett scores (AOR = 2.42, 95%CI 1.1-5.6, P = .04), underlying diseases (AOR = 2.7, 95%CI 1.1-6.8, P = .04) and overweight (AOR = 0.18, 95%CI .04-.8, P = .02). Conclusions: CEs are not rare and associated with high mortality. The most common CE is Takotsubo cardiomyopathy. CEs can occur at any stage of illness, with or without ANSD. To prevent mortality, it is pivotal to screen CEs in patients with tetanus, especially those with underlying diseases, high modified Ablett scores, and a normal or low BMI. More studies are needed to fully elucidate the impact of ANSD on the cardiovascular function and the CE associated mortality in tetanus.

2D-WinSpatt-Net: A Dual Spatial Self-Attention Vision Transformer Boosts Classification of Tetanus Severity for Patients Wearing ECG Sensors in Low- and Middle-Income Countries.

Tetanus is a life-threatening bacterial infection that is often prevalent in low- and middle-income countries (LMIC), Vietnam included. Tetanus affects the nervous system, leading to muscle stiffness and spasms. Moreover, severe tetanus is associated with autonomic nervous system (ANS) dysfunction. To ensure early detection and effective management of ANS dysfunction, patients require continuous monitoring of vital signs using bedside monitors. Wearable electrocardiogram (ECG) sensors offer a more cost-effective and user-friendly alternative to bedside monitors. Machine learning-based ECG analysis can be a valuable resource for classifying tetanus severity; however, using existing ECG signal analysis is excessively time-consuming. Due to the fixed-sized kernel filters used in traditional convolutional neural networks (CNNs), they are limited in their ability to capture global context information. In this work, we propose a 2D-WinSpatt-Net, which is a novel Vision Transformer that contains both local spatial window self-attention and global spatial self-attention mechanisms. The 2D-WinSpatt-Net boosts the classification of tetanus severity in intensive-care settings for LMIC using wearable ECG sensors. The time series imaging-continuous wavelet transforms-is transformed from a one-dimensional ECG signal and input to the proposed 2D-WinSpatt-Net. In the classification of tetanus severity levels, 2D-WinSpatt-Net surpasses state-of-the-art methods in terms of performance and accuracy. It achieves remarkable results with an F1 score of 0.88 ± 0.00, precision of 0.92 ± 0.02, recall of 0.85 ± 0.01, specificity of 0.96 ± 0.01, accuracy of 0.93 ± 0.02 and AUC of 0.90 ± 0.00.

Clinical benefit of AI-assisted lung ultrasound in a resource-limited intensive care unit.

BACKGROUND: Interpreting point-of-care lung ultrasound (LUS) images from intensive care unit (ICU) patients can be challenging, especially in low- and middle- income countries (LMICs) where there is limited training available. Despite recent advances in the use of Artificial Intelligence (AI) to automate many ultrasound imaging analysis tasks, no AI-enabled LUS solutions have been proven to be clinically useful in ICUs, and specifically in LMICs. Therefore, we developed an AI solution that assists LUS practitioners and assessed its usefulness in  a low resource ICU. METHODS: This was a three-phase prospective study. In the first phase, the performance of four different clinical user groups in interpreting LUS clips was assessed. In the second phase, the performance of 57 non-expert clinicians with and without the aid of a bespoke AI tool for LUS interpretation was assessed in retrospective offline clips. In the third phase, we conducted a prospective study in the ICU where 14 clinicians were asked to carry out LUS examinations in 7 patients with and without our AI tool and we interviewed the clinicians regarding the usability of the AI tool. RESULTS: The average accuracy of beginners' LUS interpretation was 68.7% [95% CI 66.8-70.7%] compared to 72.2% [95% CI 70.0-75.6%] in intermediate, and 73.4% [95% CI 62.2-87.8%] in advanced users. Experts had an average accuracy of 95.0% [95% CI 88.2-100.0%], which was significantly better than beginners, intermediate and advanced users (p < 0.001). When supported by our AI tool for interpreting retrospectively acquired clips, the non-expert clinicians improved their performance from an average of 68.9% [95% CI 65.6-73.9%] to 82.9% [95% CI 79.1-86.7%], (p < 0.001). In prospective real-time testing, non-expert clinicians improved their baseline performance from 68.1% [95% CI 57.9-78.2%] to 93.4% [95% CI 89.0-97.8%], (p < 0.001) when using our AI tool. The time-to-interpret clips improved from a median of 12.1 s (IQR 8.5-20.6) to 5.0 s (IQR 3.5-8.8), (p < 0.001) and clinicians' median confidence level improved from 3 out of 4 to 4 out of 4 when using our AI tool. CONCLUSIONS: AI-assisted LUS can help non-expert clinicians in an LMIC ICU improve their performance in interpreting LUS features more accurately, more quickly and more confidently.

Applying artificial intelligence and digital health technologies, Viet Nam.

Problem: Direct application of digital health technologies from high-income settings to low- and middle-income countries may be inappropriate due to challenges around data availability, implementation and regulation. Hence different approaches are needed. Approach: Within the Viet Nam ICU Translational Applications Laboratory project, since 2018 we have been developing a wearable device for individual patient monitoring and a clinical assessment tool to improve dengue disease management. Working closely with local staff at the Hospital for Tropical Diseases, Ho Chi Minh City, we developed and tested a prototype of the wearable device. We obtained perspectives on design and use of the sensor from patients. To develop the assessment tool, we used existing research data sets, mapped workflows and clinical priorities, interviewed stakeholders and held workshops with hospital staff. Local setting: In Viet Nam, a lower middle-income country, the health-care system is in the nascent stage of implementing digital health technologies. Relevant changes: Based on patient feedback, we are altering the design of the wearable sensor to increase comfort. We built the user interface of the assessment tool based on the core functionalities selected by workshop attendees. The interface was subsequently tested for usability in an iterative manner by the clinical staff members. Lessons learnt: The development and implementation of digital health technologies need an interoperable and appropriate plan for data management including collection, sharing and integration. Engagements and implementation studies should be conceptualized and conducted alongside the digital health technology development. The priorities of end-users, and understanding context and regulatory landscape are crucial for success.

Monkeypox Virus Infection in 2 Female Travelers Returning to Vietnam from Dubai, United Arab Emirates, 2022.

Mpox was diagnosed in 2 women returning to Vietnam from the United Arab Emirates. The monkeypox viruses belonged to an emerging sublineage, A.2.1, distinct from B.1, which is responsible for the ongoing multicountry outbreak. Women could contribute to mpox transmission, and enhanced genomic surveillance is needed to clarify pathogen evolution.

Improving Classification of Tetanus Severity for Patients in Low-Middle Income Countries Wearing ECG Sensors by Using a CNN-Transformer Network.

Tetanus is a life-threatening infectious disease, which is still common in low- and middle-income countries, including in Vietnam. This disease is characterized by muscle spasm and in severe cases is complicated by autonomic dysfunction. Ideally continuous vital sign monitoring using bedside monitors allows the prompt detection of the onset of autonomic nervous system dysfunction or avoiding rapid deterioration. Detection can be improved using heart rate variability analysis from ECG signals. Recently, characteristic ECG and heart rate variability features have been shown to be of value in classifying tetanus severity. However, conventional manual analysis of ECG is time-consuming. The traditional convolutional neural network (CNN) has limitations in extracting the global context information, due to its fixed-sized kernel filters. In this work, we propose a novel hybrid CNN-Transformer model to automatically classify tetanus severity using tetanus monitoring from low-cost wearable sensors. This model can capture the local features from the CNN and the global features from the Transformer. The time series imaging - spectrogram - is transformed from one-dimensional ECG signal and input to the proposed model. The CNN-Transformer model outperforms state-of-the-art methods in tetanus classification, achieves results with a F1 score of 0.82±0.03, precision of 0.94±0.03, recall of 0.73±0.07, specificity of 0.97±0.02, accuracy of 0.88±0.01 and AUC of 0.85±0.03. In addition, we found that Random Forest with enough manually selected features can be comparable with the proposed CNN-Transformer model.

vital_sqi: A Python package for physiological signal quality control.

Electrocardiogram (ECG) and photoplethysmogram (PPG) are commonly used to determine the vital signs of heart rate, respiratory rate, and oxygen saturation in patient monitoring. In addition to simple observation of those summarized indexes, waveform signals can be analyzed to provide deeper insights into disease pathophysiology and support clinical decisions. Such data, generated from continuous patient monitoring from both conventional bedside and low-cost wearable monitors, are increasingly accessible. However, the recorded waveforms suffer from considerable noise and artifacts and, hence, are not necessarily used prior to certain quality control (QC) measures, especially by those with limited programming experience. Various signal quality indices (SQIs) have been proposed to indicate signal quality. To facilitate and harmonize a wider usage of SQIs in practice, we present a Python package, named vital_sqi, which provides a unified interface to the state-of-the-art SQIs for ECG and PPG signals. The vital_sqi package provides with seven different peak detectors and access to more than 70 SQIs by using different settings. The vital_sqi package is designed with pipelines and graphical user interfaces to enable users of various programming fluency to use the package. Multiple SQI extraction pipelines can take the PPG and ECG waveforms and generate a bespoke SQI table. As these SQI scores represent the signal features, they can be input in any quality classifier. The package provides functions to build simple rule-based decision systems for signal segment quality classification using user-defined SQI thresholds. An experiment with a carefully annotated PPG dataset suggests thresholds for relevant PPG SQIs.

Classification of Tetanus Severity in Intensive-Care Settings for Low-Income Countries Using Wearable Sensing.

Infectious diseases remain a common problem in low- and middle-income countries, including in Vietnam. Tetanus is a severe infectious disease characterized by muscle spasms and complicated by autonomic nervous system dysfunction in severe cases. Patients require careful monitoring using electrocardiograms (ECGs) to detect deterioration and the onset of autonomic nervous system dysfunction as early as possible. Machine learning analysis of ECG has been shown of extra value in predicting tetanus severity, however any additional ECG signal analysis places a high demand on time-limited hospital staff and requires specialist equipment. Therefore, we present a novel approach to tetanus monitoring from low-cost wearable sensors combined with a deep-learning-based automatic severity detection. This approach can automatically triage tetanus patients and reduce the burden on hospital staff. In this study, we propose a two-dimensional (2D) convolutional neural network with a channel-wise attention mechanism for the binary classification of ECG signals. According to the Ablett classification of tetanus severity, we define grades 1 and 2 as mild tetanus and grades 3 and 4 as severe tetanus. The one-dimensional ECG time series signals are transformed into 2D spectrograms. The 2D attention-based network is designed to extract the features from the input spectrograms. Experiments demonstrate a promising performance for the proposed method in tetanus classification with an F1 score of 0.79 ± 0.03, precision of 0.78 ± 0.08, recall of 0.82 ± 0.05, specificity of 0.85 ± 0.08, accuracy of 0.84 ± 0.04 and AUC of 0.84 ± 0.03.

Sepsis Mortality Prediction Using Wearable Monitoring in Low-Middle Income Countries.

Sepsis is associated with high mortality-particularly in low-middle income countries (LMICs). Critical care management of sepsis is challenging in LMICs due to the lack of care providers and the high cost of bedside monitors. Recent advances in wearable sensor technology and machine learning (ML) models in healthcare promise to deliver new ways of digital monitoring integrated with automated decision systems to reduce the mortality risk in sepsis. In this study, firstly, we aim to assess the feasibility of using wearable sensors instead of traditional bedside monitors in the sepsis care management of hospital admitted patients, and secondly, to introduce automated prediction models for the mortality prediction of sepsis patients. To this end, we continuously monitored 50 sepsis patients for nearly 24 h after their admission to the Hospital for Tropical Diseases in Vietnam. We then compared the performance and interpretability of state-of-the-art ML models for the task of mortality prediction of sepsis using the heart rate variability (HRV) signal from wearable sensors and vital signs from bedside monitors. Our results show that all ML models trained on wearable data outperformed ML models trained on data gathered from the bedside monitors for the task of mortality prediction with the highest performance (area under the precision recall curve = 0.83) achieved using time-varying features of HRV and recurrent neural networks. Our results demonstrate that the integration of automated ML prediction models with wearable technology is well suited for helping clinicians who manage sepsis patients in LMICs to reduce the mortality risk of sepsis.

Media portrayal of vaccine: a content analysis of Vietnam online news about a pentavalent vaccine in the Expanded Program of Immunization.

Background: Vaccine hesitancy has become a prominent public health concern, particularly within the coronavirus disease 2019 (COVID-19) pandemic context. Worries about vaccine side effects are often cited as a reason for hesitancy, while media reporting about this topic plays an important role in influencing the public's perspectives about vaccines and vaccination. In Vietnam, during 2012-2013, there were several adverse events following immunization (AEFIs) of Quinvaxem- a pentavalent vaccine in the Expanded Immunization Program, which made big headlines in the media. Such incidences have contributed to changes in vaccination policies and influenced parents' concerns to date. This study explores the portrayal of Quinvaxem in Vietnam digital news during four periods marked by important events. Methods: We performed quantitative and qualitative analysis with a coding framework to identify main content focus, sentiments towards Quinvaxem, and emotional tones in these articles. Results: In total, we included 360 articles into analysis. The amount of news coverage about Quinvaxem increased after AEFIs happened, from 7 articles before AEFIs to 98 and 159 articles in the following periods when AEFIs and investigation into vaccine safety occurred. Most articles are neutral in titles (n=255/360) and content (n=215/360) towards Quinvaxem and do not convey emotional expressions (n=271/360). However, articles focusing on side effects contain negative sentiments and emotional expressions more frequently than articles of other contents while AEFIs details were conflicting across articles. Vaccine sentiments are provoked in the information about vaccine quality and safety, health authority, local delivery, and quoted vaccine opinions. Emotion-conveying elements in 89/360 articles included emotional wording and imagery and expressive punctuation. Conclusions: The heterogeneity of information in online news may reinforce uncertainty about vaccine safety and decrease vaccine intention. Our results have important implications for vaccine communication, given the current plan of the Vietnamese government to roll out COVID-19 vaccination to younger children.