Prevalence and Predictors of Subclinical Cardiomyopathy in Patients With Type 2 Diabetes in a Health System.

INTRODUCTION: Diabetic cardiomyopathy (DbCM) is characterized by subclinical abnormalities in cardiac structure/function and is associated with a higher risk of overt heart failure (HF). However, there are limited data on optimal strategies to identify individuals with DbCM in contemporary health systems. The aim of this study was to evaluate the prevalence of DbCM in a health system using existing data from the electronic health record (EHR). METHODS: Adult patients with type 2 diabetes mellitus free of cardiovascular disease (CVD) with available data on HF risk in a single-center EHR were included. The presence of DbCM was defined using different definitions: (1) least restrictive: ≥1 echocardiographic abnormality (left atrial enlargement, left ventricle hypertrophy, diastolic dysfunction); (2) intermediate restrictive: ≥2 echocardiographic abnormalities; (3) most restrictive: 3 echocardiographic abnormalities. DbCM prevalence was compared across age, sex, race, and ethnicity-based subgroups, with differences assessed using the chi-squared test. Adjusted logistic regression models were constructed to evaluate significant predictors of DbCM. RESULTS: Among 1921 individuals with type 2 diabetes mellitus, the prevalence of DbCM in the overall cohort was 8.7% and 64.4% in the most and least restrictive definitions, respectively. Across all definitions, older age and Hispanic ethnicity were associated with a higher proportion of DbCM. Females had a higher prevalence than males only in the most restrictive definition. In multivariable-adjusted logistic regression, higher systolic blood pressure, higher creatinine, and longer QRS duration were associated with a higher risk of DbCM across all definitions. CONCLUSIONS: In this single-center, EHR cohort, the prevalence of DbCM varies from 9% to 64%, with a higher prevalence with older age and Hispanic ethnicity.

Air pollution and its effects on emergency room visits in tertiary respiratory care centers in Delhi, India.

Environmental pollution has harmful effects on human health, particularly the respiratory system. We aimed to study the impact of daily ambient air pollution on daily emergency room visits for acute respiratory symptoms. This study was conducted in two tertiary respiratory care centres in Delhi, India. Daily counts of emergency room visits were collected. All patients attending the emergency room were screened for acute onset (less than 2 weeks) of respiratory symptoms and were recruited if they were staying in Delhi continuously for at least 4 weeks and having onset (≤2 weeks) of respiratory symptoms. Daily average air pollution data for the study period was obtained from four continuous ambient air quality monitoring stations. A total of 61,285 patients were screened and 11,424 were enrolled from June 2017 to February 2019. Cough and difficulty in breathing were most common respiratory symptoms. Poor air quality was observed during the months of October to December. Emergency room visits with acute respiratory symptoms significantly increased per standard deviation increase in PM10 from lag days 2-7. Increase in wheezing was primarily seen with increase in NO2. Pollutant levels have effect on acute respiratory symptoms and thus influence emergency room visits. *************************************************************** *Appendix Authors list Kamal Singhal,1 Kana Ram Jat,2 Karan Madan,3 Mohan P. George,4 Kalaivani Mani,5 Randeep Guleria,3 Ravindra Mohan Pandey,5 Rupinder Singh Dhaliwal,6 Rakesh Lodha,2 Varinder Singh1 1Department of Paediatrics, Lady Hardinge Medical College and associated Kalawati Saran Children's Hospital, New Delhi, India 2Department of Paediatrics, All India Institute of Medical Sciences, New Delhi, India 3Department of Pulmonary Medicine, Critical Care and Sleep Disorders, All India Institute of Medical Sciences, New Delhi, India 4Department of Environment, Delhi Pollution Control Committee, Kashmere Gate, New Delhi, India 5Department of Biostatistics, All India Institute of Medical Sciences, New Delhi, India 6Department of Non-communicable Diseases, Indian Council of Medical Research, New Delhi, India.

Early prediction of hypothermia in pediatric intensive care units using machine learning.

Hypothermia is a life-threatening condition where the temperature of the body drops below 35°C and is a key source of concern in Intensive Care Units (ICUs). Early identification can help to nudge clinical management to initiate early interventions. Despite its importance, very few studies have focused on the early prediction of hypothermia. In this study, we aim to monitor and predict Hypothermia (30 min-4 h) ahead of its onset using machine learning (ML) models developed on physiological vitals and to prospectively validate the best performing model in the pediatric ICU. We developed and evaluated ML algorithms for the early prediction of hypothermia in a pediatric ICU. Sepsis advanced forecasting engine ICU Database (SafeICU) data resource is an in-house ICU source of data built in the Pediatric ICU at the All-India Institute of Medical Science (AIIMS), New Delhi. Each time-stamp at 1-min resolution was labeled for the presence of hypothermia to construct a retrospective cohort of pediatric patients in the SafeICU data resource. The training set consisted of windows of the length of 4.2 h with a lead time of 30 min-4 h from the onset of hypothermia. A set of 3,835 hand-engineered time-series features were calculated to capture physiological features from the time series. Features selection using the Boruta algorithm was performed to select the most important predictors of hypothermia. A battery of models such as gradient boosting machine, random forest, AdaBoost, and support vector machine (SVM) was evaluated utilizing five-fold test sets. The best-performing model was prospectively validated. A total of 148 patients with 193 ICU stays were eligible for the model development cohort. Of 3,939 features, 726 were statistically significant in the Boruta analysis for the prediction of Hypothermia. The gradient boosting model performed best with an Area Under the Receiver Operating Characteristic curve (AUROC) of 85% (SD = 1.6) and a precision of 59.2% (SD = 8.8) for a 30-min lead time before the onset of Hypothermia onset. As expected, the model showed a decline in model performance at higher lead times, such as AUROC of 77.2% (SD = 2.3) and precision of 41.34% (SD = 4.8) for 4 h ahead of Hypothermia onset. Our GBM(gradient boosting machine) model produced equal and superior results for the prospective validation, where an AUROC of 79.8% and a precision of 53% for a 30-min lead time before the onset of Hypothermia whereas an AUROC of 69.6% and a precision of 38.52% for a (30 min-4 h) lead time prospective validation of Hypothermia. Therefore, this work establishes a pipeline termed ThermoGnose for predicting hypothermia, a major complication in pediatric ICUs.

Predicting Emerging Themes in Rapidly Expanding COVID-19 Literature With Unsupervised Word Embeddings and Machine Learning: Evidence-Based Study.

BACKGROUND: Evidence from peer-reviewed literature is the cornerstone for designing responses to global threats such as COVID-19. In massive and rapidly growing corpuses, such as COVID-19 publications, assimilating and synthesizing information is challenging. Leveraging a robust computational pipeline that evaluates multiple aspects, such as network topological features, communities, and their temporal trends, can make this process more efficient. OBJECTIVE: We aimed to show that new knowledge can be captured and tracked using the temporal change in the underlying unsupervised word embeddings of the literature. Further imminent themes can be predicted using machine learning on the evolving associations between words. METHODS: Frequently occurring medical entities were extracted from the abstracts of more than 150,000 COVID-19 articles published on the World Health Organization database, collected on a monthly interval starting from February 2020. Word embeddings trained on each month's literature were used to construct networks of entities with cosine similarities as edge weights. Topological features of the subsequent month's network were forecasted based on prior patterns, and new links were predicted using supervised machine learning. Community detection and alluvial diagrams were used to track biomedical themes that evolved over the months. RESULTS: We found that thromboembolic complications were detected as an emerging theme as early as August 2020. A shift toward the symptoms of long COVID complications was observed during March 2021, and neurological complications gained significance in June 2021. A prospective validation of the link prediction models achieved an area under the receiver operating characteristic curve of 0.87. Predictive modeling revealed predisposing conditions, symptoms, cross-infection, and neurological complications as dominant research themes in COVID-19 publications based on the patterns observed in previous months. CONCLUSIONS: Machine learning-based prediction of emerging links can contribute toward steering research by capturing themes represented by groups of medical entities, based on patterns of semantic relationships over time.

Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos.

Shock is one of the major killers in intensive care units, and early interventions can potentially reverse it. In this study, we advance a noncontact thermal imaging modality for continuous monitoring of hemodynamic shock working on 1,03,936 frames from 406 videos recorded longitudinally upon 22 pediatric patients. Deep learning was used to preprocess and extract the Center-to-Peripheral Difference (CPD) in temperature values from the videos. This time-series data along with the heart rate was finally analyzed using Long-Short Term Memory models to predict the shock status up to the next 6 h. Our models achieved the best area under the receiver operating characteristic curve of 0.81 ± 0.06 and area under the precision-recall curve of 0.78 ± 0.05 at 5 h, providing sufficient time to stabilize the patient. Our approach, thus, provides a reliable shock prediction using an automated decision pipeline that can provide better care and save lives.

VacSIM: Learning effective strategies for COVID-19 vaccine distribution using reinforcement learning.

A COVID-19 vaccine is our best bet for mitigating the ongoing onslaught of the pandemic. However, vaccine is also expected to be a limited resource. An optimal allocation strategy, especially in countries with access inequities and temporal separation of hot-spots, might be an effective way of halting the disease spread. We approach this problem by proposing a novel pipeline VacSIM that dovetails Deep Reinforcement Learning models into a Contextual Bandits approach for optimizing the distribution of COVID-19 vaccine. Whereas the Reinforcement Learning models suggest better actions and rewards, Contextual Bandits allow online modifications that may need to be implemented on a day-to-day basis in the real world scenario. We evaluate this framework against a naive allocation approach of distributing vaccine proportional to the incidence of COVID-19 cases in five different States across India (Assam, Delhi, Jharkhand, Maharashtra and Nagaland) and demonstrate up to 9039 potential infections prevented and a significant increase in the efficacy of limiting the spread over a period of 45 days through the VacSIM approach. Our models and the platform are extensible to all states of India and potentially across the globe. We also propose novel evaluation strategies including standard compartmental model-based projections and a causality-preserving evaluation of our model. Since all models carry assumptions that may need to be tested in various contexts, we open source our model VacSIM and contribute a new reinforcement learning environment compatible with OpenAI gym to make it extensible for real-world applications across the globe.

Estimating the impact of health systems factors on antimicrobial resistance in priority pathogens.

OBJECTIVES: Antimicrobial resistance (AMR) is the next big pandemic that threatens humanity. The One Health approach to AMR requires quantification of interactions between health, demographic, socioeconomic, environmental, and geopolitical factors to design interventions. This study is focused on learning health system factors on global AMR. METHODS: This study analysed longitudinal data (2004-2017) of AMR having 6 33 820 isolates from 70 middle and high-income countries. We integrated AMR data with the Global Burden of Disease (GBD), Governance (WGI), and Finance data sets to find AMR's unbiased and actionable determinants. We chose a Bayesian decision network (BDN) approach within the causal modelling framework to quantify determinants of AMR. Further, we integrated Bayesian networks' global knowledge discovery approach with discriminative machine learning to predict individual-level antibiotic susceptibility in patients. RESULTS: From MAR (multiple antibiotic resistance) scores, we found a non-uniform spread pattern of AMR. Components-level analysis revealed that governance, finance, and disease burden variables strongly correlate with AMR. From the Bayesian network analysis, we found that access to immunization, obstetric care, and government effectiveness are strong, actionable factors in reducing AMR, confirmed by what-if analysis. Finally, our discriminative machine learning models achieved an individual-level AUROC (Area under receiver operating characteristic curve) of 0.94 (SE = 0.01) and 0.89 (SE = 0.002) to predict Staphylococcus aureus resistance to ceftaroline and oxacillin, respectively. CONCLUSION: Causal machine learning revealed that immunisation strategies and quality of governance are vital, actionable interventions to reduce AMR.

Genomic Surveillance of COVID-19 Variants With Language Models and Machine Learning.

The global efforts to control COVID-19 are threatened by the rapid emergence of novel SARS-CoV-2 variants that may display undesirable characteristics such as immune escape, increased transmissibility or pathogenicity. Early prediction for emergence of new strains with these features is critical for pandemic preparedness. We present Strainflow, a supervised and causally predictive model using unsupervised latent space features of SARS-CoV-2 genome sequences. Strainflow was trained and validated on 0.9 million sequences for the period December, 2019 to June, 2021 and the frozen model was prospectively validated from July, 2021 to December, 2021. Strainflow captured the rise in cases 2 months ahead of the Delta and Omicron surges in most countries including the prediction of a surge in India as early as beginning of November, 2021. Entropy analysis of Strainflow unsupervised embeddings clearly reveals the explore-exploit cycles in genomic feature-space, thus adding interpretability to the deep learning based model. We also conducted codon-level analysis of our model for interpretability and biological validity of our unsupervised features. Strainflow application is openly available as an interactive web-application for prospective genomic surveillance of COVID-19 across the globe.

Geographic information system-based mapping of air pollution & emergency room visits of patients for acute respiratory symptoms in Delhi, India (March 2018-February 2019).

Background & objectives: Studies assessing the spatial and temporal association of ambient air pollution with emergency room visits of patients having acute respiratory symptoms in Delhi are lacking. Therefore, the present study explored the relationship between spatio-temporal variation of particulate matter (PM)2.5 concentrations and air quality index (AQI) with emergency room (ER) visits of patients having acute respiratory symptoms in Delhi using the geographic information system (GIS) approach. Methods: The daily number of ER visits of patients having acute respiratory symptoms (less than or equal to two weeks) was recorded from the ER of four hospitals of Delhi from March 2018 to February 2019. Daily outdoor PM2.5 concentrations and air quality index (AQI) were obtained from the Delhi Pollution Control Committee. Spatial distribution of patients with acute respiratory symptoms visiting ER, PM2.5 concentrations and AQI were mapped for three seasons of Delhi using ArcGIS software. Results: Of the 70,594 patients screened from ER, 18,063 eligible patients were enrolled in the study. Winter days had poor AQI compared to moderate and satisfactory AQI during summer and monsoon days, respectively. None of the days reported good AQI (<50). During winters, an increase in acute respiratory ER visits of patients was associated with higher PM2.5 concentrations in the highly polluted northwest region of Delhi. In contrast, a lower number of acute respiratory ER visits of patients were seen from the 'moderately polluted' south-west region of Delhi with relatively lower PM2.5 concentrations. Interpretation & conclusions: Acute respiratory ER visits of patients were related to regional PM2.5 concentrations and AQI that differed during the three seasons of Delhi. The present study provides support for identifying the hotspots and implementation of focused, intensive decentralized strategies to control ambient air pollution in worst-affected areas, in addition to the general city-wise strategies.

Physical Growth and Its Determinants in Indian Children with Down Syndrome, from 3 Months to 5 Years of Age.

OBJECTIVE: To compare the growth of Indian children with Down syndrome (DS), with typically developing Indian children. The effect of comorbidities on their physical growth was also evaluated, so that factors affecting growth can be identified early and timely interventions be planned. METHODS: A cross-sectional study was conducted at the All India Institute of Medical Sciences, New Delhi from June 2015 to June 2017. Children with karyotype-proven DS within age group of 3 mo to 5 y were enrolled as study subjects. Children with DS having mosaic karyotype were excluded. Anthropometry and associated comorbidities were assessed. RESULTS: Hundred and eight children with DS were enrolled, mean WHO z scores were-WAZ: -2.31 (SD 1.44), HAZ: -2.51 (SD 1.47), BAZ: -1.07 (SD 1.8), and HCZ: -2.79 (SD 1.21). Congenital heart disease (in 44.5% children), hypothyroidism (in 27.7%), and anemia (in 50%) were the common comorbidities. Growth parameters of children with and without any comorbidity were significantly different, mean WHO z scores were WAZ -2.61 vs. -1.09 (p = 0.005), HAZ -2.43 vs. -2.41 (p = 0.3), BAZ -1.49 vs. -0.38 (p = 0.001), and HCZ -3.13 vs. -2.33 (p = 0.001). CONCLUSION: Growth of Indian children with DS is significantly less compared to normally growing children. Weight was affected maximum during infancy, length was more affected as the age progressed, head circumference was affected similarly in all age groups, whereas BMI showed almost progressive increase with age. Children with severe heart disease had significantly lower BMI whereas children with treated hypothyroidism had better growth. There is a need for a large longitudinal study on Indian children with DS to construct Indian DS-specific growth charts.

Effects of ambient air pollution on emergency room visits of children for acute respiratory symptoms in Delhi, India.

The present study explored the association between daily ambient air pollution and daily emergency room (ER) visits due to acute respiratory symptoms in children of Delhi. The daily counts of ER visits (ERV) of children (≤15 years) having acute respiratory symptoms were obtained from two hospitals of Delhi for 21 months. Simultaneously, data on daily concentrations of particulate matter (PM10 and PM2.5), nitrogen dioxide (NO2), sulfur dioxide (SO2), carbon monoxide (CO), and ozone (O3) and weather variables were provided by the Delhi Pollution Control Committee. K-means clustering with time-series approach and multi-pollutant generalized additive models with Poisson link function was used to estimate the 0-6-day lagged change in daily ER visits with the change in multiple pollutants levels. Out of 1,32,029 children screened, 19,120 eligible children having acute respiratory symptoms for ≤2 weeks and residing in Delhi for the past 4 weeks were enrolled. There was a 29% and 21% increase in ERVs among children on high and moderate level pollution cluster days, respectively, compared to low pollution cluster days on the same day and previous 1-6 days of exposure to air pollutants. There was percentage increase (95% CI) 1.50% (0.76, 2.25) in ERVs for acute respiratory symptoms for 10 µg/m3 increase of NO2 on previous day 1, 46.78% (21.01, 78.05) for 10 µg/m3 of CO on previous day 3, and 13.15% (9.95, 16.45) for 10 µg/m3 of SO2 on same day of exposure. An increase in the daily ER visits of children for acute respiratory symptoms was observed after increase in daily ambient air pollution levels in Delhi.

Genetic variations in olfactory receptor gene OR2AG2 in a large multigenerational family with asthma.

It is estimated from twin studies that heritable factors account for at-least half of asthma-risk, of which genetic variants identified through population studies explain only a small fraction. Multi-generation large families with high asthma prevalence can serve as a model to identify highly penetrant genetic variants in closely related individuals that are missed by population studies. To achieve this, a four-generation Indian family with asthma was identified and recruited for examination and genetic testing. Twenty subjects representing all generations were selected for whole genome genotyping, of which eight were subjected to exome sequencing. Non-synonymous and deleterious variants, segregating with the affected individuals, were identified by exome sequencing. A prioritized deleterious missense common variant in the olfactory receptor gene OR2AG2 that segregated with a risk haplotype in asthma, was validated in an asthma cohort of different ethnicity. Phenotypic tests were conducted to verify expected deficits in terms of reduced ability to sense odors. Pathway-level relevance to asthma biology was tested in model systems and unrelated human lung samples. Our study suggests that OR2AG2 and other olfactory receptors may contribute to asthma pathophysiology. Genetic studies on large families of interest can lead to efficient discovery.

Predicting Hemodynamic Shock from Thermal Images using Machine Learning.

Proactive detection of hemodynamic shock can prevent organ failure and save lives. Thermal imaging is a non-invasive, non-contact modality to capture body surface temperature with the potential to reveal underlying perfusion disturbance in shock. In this study, we automate early detection and prediction of shock using machine learning upon thermal images obtained in a pediatric intensive care unit of a tertiary care hospital. 539 images were recorded out of which 253 had concomitant measurement of continuous intra-arterial blood pressure, the gold standard for shock monitoring. Histogram of oriented gradient features were used for machine learning based region-of-interest segmentation that achieved 96% agreement with a human expert. The segmented center-to-periphery difference along with pulse rate was used in longitudinal prediction of shock at 0, 3, 6 and 12 hours using a generalized linear mixed-effects model. The model achieved a mean area under the receiver operating characteristic curve of 75% at 0 hours (classification), 77% at 3 hours (prediction) and 69% at 12 hours (prediction) respectively. Since hemodynamic shock associated with critical illness and infectious epidemics such as Dengue is often fatal, our model demonstrates an affordable, non-invasive, non-contact and tele-diagnostic decision support system for its reliable detection and prediction.