A predictive paradigm for COVID-19 prognosis based on the longitudinal measure of biomarkers

Tree-based ensemble machine learning models in the prediction of acute respiratory distress syndrome following cardiac surgery: a multicenter cohort study

BACKGROUND

Acute respiratory distress syndrome (ARDS) after cardiac surgery is a severe respiratory complication with high mortality and morbidity. Traditional clinical approaches may lead to under recognition of this heterogeneous syndrome, potentially resulting in diagnosis delay. This study aims to develop and external validate seven machine learning (ML) models, trained on electronic health records data, for predicting ARDS after cardiac surgery.

METHODS

This multicenter, observational cohort study included patients who underwent cardiac surgery in the training and testing cohorts (data from Nanjing First Hospital), as well as those patients who had cardiac surgery in a validation cohort (data from Shanghai General Hospital). The number of important features was determined using the sliding windows sequential forward feature selection method (SWSFS). We developed a set of tree-based ML models, including Decision Tree, GBDT, AdaBoost, XGBoost, LightGBM, Random Forest, and Deep Forest. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC) and Brier score. The SHapley Additive exPlanation (SHAP) techinque was employed to interpret the ML model. Furthermore, a comparison was made between the ML models and traditional scoring systems. ARDS is defined according to the Berlin definition.

RESULTS

A total of 1996 patients who had cardiac surgery were included in the study. The top five important features identified by the SWSFS were chronic obstructive pulmonary disease, preoperative albumin, central venous pressure_T4, cardiopulmonary bypass time, and left ventricular ejection fraction. Among the seven ML models, Deep Forest demonstrated the best performance, with an AUC of 0.882 and a Brier score of 0.809 in the validation cohort. Notably, the SHAP values effectively illustrated the contribution of the 13 features attributed to the model output and the individual feature's effect on model prediction. In addition, the ensemble ML models demonstrated better performance than the other six traditional scoring systems.

CONCLUSIONS

Our study identified 13 important features and provided multiple ML models to enhance the risk stratification for ARDS after cardiac surgery. Using these predictors and ML models might provide a basis for early diagnostic and preventive strategies in the perioperative management of ARDS patients.

Bioinformatics and molecular biology tools for diagnosis, prevention, treatment and prognosis of COVID-19

Since December 2019, a new form of Severe Acute Respiratory Syndrome (SARS) has emerged worldwide, caused by SARS coronavirus 2 (SARS-CoV-2). This disease was called COVID-19 and was declared a pandemic by the World Health Organization in March 2020. Symptoms can vary from a common cold to severe pneumonia, hypoxemia, respiratory distress, and death. During this period of world stress, the medical and scientific community were able to acquire information and generate scientific data at unprecedented speed, to better understand the disease and facilitate vaccines and therapeutics development. Notably, bioinformatics tools were instrumental in decoding the viral genome and identifying critical targets for COVID-19 diagnosis and therapeutics. Through the integration of omics data, bioinformatics has also improved our understanding of disease pathogenesis and virus-host interactions, facilitating the development of targeted treatments and vaccines. Furthermore, molecular biology techniques have accelerated the design of sensitive diagnostic tests and the characterization of immune responses, paving the way for precision medicine approaches in treating COVID-19. Our analysis highlights the indispensable contributions of bioinformatics and molecular biology to the global effort against COVID-19. In this review, we aim to revise the COVID-19 features, diagnostic, prevention, treatment options, and how molecular biology, modern bioinformatic tools, and collaborations have helped combat this pandemic. An integrative literature review was performed, searching articles on several sites, including PUBMED and Google Scholar indexed in referenced databases, prioritizing articles from the last 3 years. The lessons learned from this COVID-19 pandemic will place the world in a much better position to respond to future pandemics.

On-admission and dynamic trend of laboratory profiles as prognostic biomarkers in COVID-19 inpatients

This large-scale study aimed to investigate the trend of laboratory tests of patients with COVID-19. Hospitalized confirmed and probable COVID-19 patients in three general hospitals were examined from March 20, 2020, to June 18, 2021. The confirmed and probable COVID-19 patients with known outcomes and valid laboratory results were included. The least absolute shrinkage and selection operator (LASSO) and Cox regression were used to select admittance prognostic features. Parallel Pairwise Comparison of mortality versus survival was used to examine the trend of markers. In the final cohort, 11,944 patients were enrolled, with an in-hospital mortality rate of 21.8%, mean age of 59.4 ± 18.0, and a male-to-female ratio of 1.3. Abnormal admittance level of white blood cells, neutrophils, lymphocytes, mean cellular volume, urea, creatinine, bilirubin, creatine kinase-myoglobin binding, lactate dehydrogenase (LDH), Troponin, c-reactive protein (CRP), potassium, and creatinine phosphokinase reduced the survival of COVID-19 inpatients. Moreover, the trend analysis showed lymphocytes, platelet, urea, CRP, alanine transaminase (ALT), and LDH have a dissimilar trend in non-survivors compared to survived patients. This study proposed a novel approach to find serial laboratory markers. Serial examination of platelet count, creatinine, CRP, LDH, and ALT can guide healthcare professionals in finding patients at risk of deterioration.

Prediction Model of hospitalization time of COVID-19 patients based on Gradient Boosted Regression Trees

When an outbreak of COVID-19 occurs, it will cause a shortage of medical resources and the surge of demand for hospital beds. Predicting the length of stay (LOS) of COVID-19 patients is helpful to the overall coordination of hospital management and improves the utilization rate of medical resources. The purpose of this paper is to predict LOS for patients with COVID-19, so as to provide hospital management with auxiliary decision-making of medical resource scheduling. We collected the data of 166 COVID-19 patients in a hospital in Xinjiang from July 19, 2020, to August 26, 2020, and carried out a retrospective study. The results showed that the median LOS was 17.0 days, and the average of LOS was 18.06 days. Demographic data and clinical indicators were included as predictive variables to construct a model for predicting the LOS using gradient boosted regression trees (GBRT). The MSE, MAE and MAPE of the model are 23.84, 4.12 and 0.76 respectively. The importance of all the variables involved in the prediction of the model was analyzed, and the clinical indexes creatine kinase-MB (CK-MB), C-reactive protein (CRP), creatine kinase (CK), white blood cell count (WBC) and the age of patients had a higher contribution to the LOS. We found our GBRT model can accurately predict the LOS of COVID-19 patients, which will provide good assistant decision-making for medical management.

Interpretable artificial intelligence model for accurate identification of medical conditions using immune repertoire

Underlying medical conditions, such as cancer, kidney disease and heart failure, are associated with a higher risk for severe COVID-19. Accurate classification of COVID-19 patients with underlying medical conditions is critical for personalized treatment decision and prognosis estimation. In this study, we propose an interpretable artificial intelligence model termed VDJMiner to mine the underlying medical conditions and predict the prognosis of COVID-19 patients according to their immune repertoires. In a cohort of more than 1400 COVID-19 patients, VDJMiner accurately identifies multiple underlying medical conditions, including cancers, chronic kidney disease, autoimmune disease, diabetes, congestive heart failure, coronary artery disease, asthma and chronic obstructive pulmonary disease, with an average area under the receiver operating characteristic curve (AUC) of 0.961. Meanwhile, in this same cohort, VDJMiner achieves an AUC of 0.922 in predicting severe COVID-19. Moreover, VDJMiner achieves an accuracy of 0.857 in predicting the response of COVID-19 patients to tocilizumab treatment on the leave-one-out test. Additionally, VDJMiner interpretively mines and scores V(D)J gene segments of the T-cell receptors that are associated with the disease. The identified associations between single-cell V(D)J gene segments and COVID-19 are highly consistent with previous studies. The source code of VDJMiner is publicly accessible at https://github.com/TencentAILabHealthcare/VDJMiner. The web server of VDJMiner is available at https://gene.ai.tencent.com/VDJMiner/.

Factors associated to the duration of COVID-19 lockdowns in Chile

During the first year of the COVID-19 pandemic, several countries have implemented non-pharmacologic measures, mainly lockdowns and social distancing, to reduce the spread of the SARS-CoV-2 virus. These strategies varied widely across nations, and their efficacy is currently being studied. This study explores demographic, socioeconomic, and epidemiological factors associated with the duration of lockdowns applied in Chile between March 25th and December 25th, 2020. Joint models for longitudinal and time-to-event data were used. In this case, the number of days under lockdown for each Chilean commune and longitudinal information were modeled jointly. Our results indicate that overcrowding, number of active cases, and positivity index are significantly associated with the duration of lockdowns, being identified as risk factors for longer lockdown duration. In short, joint models for longitudinal and time-to-event data permit the identification of factors associated with the duration of lockdowns in Chile. Indeed, our findings suggest that demographic, socioeconomic, and epidemiological factors should be used to define both entering and exiting lockdown.

Learning and Predicting from Dynamic Models for COVID-19 Patient Monitoring

COVID-19 has challenged health systems to learn how to learn. This paper describes the context, methods and challenges for learning to improve COVID-19 care at one academic health center. Challenges to learning include: (1) choosing a right clinical target; (2) designing methods for accurate predictions by borrowing strength from prior patients' experiences; (3) communicating the methodology to clinicians so they understand and trust it; (4) communicating the predictions to the patient at the moment of clinical decision; and (5) continuously evaluating and revising the methods so they adapt to changing patients and clinical demands. To illustrate these challenges, this paper contrasts two statistical modeling approaches - prospective longitudinal models in common use and retrospective analogues complementary in the COVID-19 context - for predicting future biomarker trajectories and major clinical events. The methods are applied to and validated on a cohort of 1,678 patients who were hospitalized with COVID-19 during the early months of the pandemic. We emphasize graphical tools to promote physician learning and inform clinical decision making.

Rethinking Autism Intervention Science: A Dynamic Perspective

Recent advances in longitudinal methodologies for observational studies have contributed to a better understanding of Autism as a neurodevelopmental condition characterized by within-person and between-person variability over time across behavioral domains. However, this finer-grained approach to the study of developmental variability has yet to be applied to Autism intervention science. The widely adopted experimental designs in the field-randomized control trials and quasi-experimental designs-hold value for inferring treatment effects; at the same time, they are limited in elucidating what works for whom, why, and when, given the idiosyncrasies of neurodevelopmental disorders where predictors and outcomes are often dynamic in nature. This perspective paper aims to serve as a primer for Autism intervention scientists to rethink the way we approach predictors of treatment response and treatment-related change using a dynamic lens. We discuss several empirical gaps, and potential methodological challenges and opportunities pertaining to: (1) capturing finer-grained treatment effects in specific behavioral domains as indexed by micro-level within-person changes during and beyond intervention; and (2) examining and modeling dynamic prediction of treatment response. Addressing these issues can contribute to enhanced study designs and methodologies that generate evidence to inform the development of more personalized interventions and stepped care approaches for individuals on the heterogeneous spectrum of Autism with changing needs across development.