A Multimodality Machine Learning Approach to Differentiate Severe and Nonsevere COVID-19: Model Development and Validation

DeepASD: a deep adversarial-regularized graph learning method for ASD diagnosis with multimodal data

Autism Spectrum Disorder (ASD) is a prevalent neurological condition with multiple co-occurring comorbidities that seriously affect mental health. Precisely diagnosis of ASD is crucial to intervention and rehabilitation. A single modality may not fully reflect the complex mechanisms underlying ASD, and combining multiple modalities enables a more comprehensive understanding. Here, we propose, DeepASD, an end-to-end trainable regularized graph learning method for ASD prediction, which incorporates heterogeneous multimodal data and latent inter-patient relationships to better understand the pathogenesis of ASD. DeepASD first learns cross-modal feature representations through a multimodal adversarial-regularized encoder, and then constructs adaptive patient similarity networks by leveraging the representations of each modality. DeepASD exploits inter-patient relationships to boost the ASD diagnosis that is implemented by a classifier compositing of graph neural networks. We apply DeepASD to the benchmarking Autism Brain Imaging Data Exchange (ABIDE) data with four modalities. Experimental results show that the proposed DeepASD outperforms eight state-of-the-art baselines on the benchmarking ABIDE data, showing an improvement of 13.25% in accuracy, 7.69% in AUC-ROC, and 17.10% in specificity. DeepASD holds promise for a more comprehensive insight of the complex mechanisms of ASD, leading to improved diagnosis performance.

A predictive model to explore risk factors for severe COVID-19

With the rapid spread of the novel coronavirus (COVID-19), a sustained global pandemic has emerged. Globally, the cumulative death toll is in the millions. The rising number of COVID-19 infections and deaths has severely impacted the lives of people worldwide, healthcare systems, and economic development. We conducted a retrospective analysis of the characteristics of COVID-19 patients. This analysis includes clinical features upon initial hospital admission, relevant laboratory test results, and imaging findings. We aimed to identify risk factors for severe illness and to construct a predictive model for assessing the risk of severe COVID-19. We collected and analyzed electronic medical records of confirmed COVID-19 patients admitted to the Affiliated Hospital of Jiangsu University (Zhenjiang, China) between December 18, 2022, and February 28, 2023. According to the WHO diagnostic criteria for the novel coronavirus, we divided the patients into two groups: severe and non-severe, and compared their clinical, laboratory, and imaging data. Logistic regression analysis, the least absolute shrinkage and selection operator (LASSO) regression, and receiver operating characteristic (ROC) curve analysis were used to identify the relevant risk factors for severe COVID-19 patients. Patients were divided into a training cohort and a validation cohort. A nomogram model was constructed using the "rms" package in R software. Among the 346 patients, the severe group exhibited significantly higher respiratory rates, breathlessness, altered consciousness, neutrophil-to-lymphocyte ratio (NLR), and lactate dehydrogenase (LDH) levels compared to the non-severe group. Imaging findings indicated that the severe group had a higher proportion of bilateral pulmonary inflammation and ground-glass opacities compared to the non-severe group. NLR and LDH were identified as independent risk factors for severe patients. The diagnostic performance was maximized when NLR, respiratory rate (RR), and LDH were combined. Based on the statistical analysis results, we developed a COVID-19 severity risk prediction model. The total score is calculated by adding up the scores for each of the twelve independent variables. By mapping the total score to the lowest scale, we can estimate the risk of COVID-19 severity. In addition, the calibration plots and DCA analysis showed that the nomogram had better discrimination power for predicting the severity of COVID-19. Our results showed that the development and validation of the predictive nomogram had good predictive value for severe COVID-19.

Comprehensive analysis of immunogenic cell death-related gene and construction of prediction model based on WGCNA and multiple machine learning in severe COVID-19

The death of coronavirus disease 2019 (COVID-19) is primarily due to from critically ill patients, especially from ARDS complications caused by SARS-CoV-2. Therefore, it is essential to contribute an in-depth understanding of the pathogenesis of the disease and to identify biomarkers for predicting critically ill patients at the molecular level. Immunogenic cell death (ICD), as a specific variant of regulatory cell death driven by stress, can induce adaptive immune responses against cell death antigens in the host. Studies have confirmed that both innate and adaptive immune pathways are involved in the pathogenesis of SARS-CoV-2 infection. However, the role of ICD in the pathogenesis of severe COVID-19 has rarely been explored. In this study, we systematically evaluated the role of ICD-related genes in COVID-19. We conducted consensus clustering, immune infiltration analysis, and functional enrichment analysis based on ICD differentially expressed genes. The results showed that immune infiltration characteristics were altered in severe and non-severe COVID-19. In addition, we used multiple machine learning methods to screen for five risk genes (KLF5, NSUN7, APH1B, GRB10 and CD4), which are used to predict COVID-19 severity. Finally, we constructed a nomogram to predict the risk of severe COVID-19 based on the classification and recognition model, and validated the model with external data sets. This study provides a valuable direction for the exploration of the pathogenesis and progress of COVID-19, and helps in the early identification of severe cases of COVID-19 to reduce mortality.

Machine learning versus multivariate logistic regression for predicting severe COVID-19 in hospitalized children with Omicron variant infection

With the emergence of the Omicron variant, the number of pediatric Coronavirus Disease 2019 (COVID-19) cases requiring hospitalization and developing severe or critical illness has significantly increased. Machine learning and multivariate logistic regression analysis were used to predict risk factors and develop prognostic models for severe COVID-19 in hospitalized children with the Omicron variant in this study. Of the 544 hospitalized children including 243 and 301 in the mild and severe groups, respectively. Fever (92.3%) was the most common symptom, followed by cough (79.4%), convulsions (36.8%), and vomiting (23.2%). The multivariate logistic regression analysis showed that age (1-3 years old, odds ratio (OR): 3.193, 95% confidence interval (CI): 1.778-5.733], comorbidity (OR: 1.993, 95% CI:1.154-3.443), cough (OR: 0.409, 95% CI:0.236-0.709), and baseline neutrophil-to-lymphocyte ratio (OR: 1.108, 95% CI: 1.023-1.200), lactate dehydrogenase (OR: 1.993, 95% CI: 1.154-3.443), blood urea nitrogen (OR: 1.002, 95% CI: 1.000-1.003) and total bilirubin (OR: 1.178, 95% CI: 1.005-3.381) were independent risk factors for severe COVID-19. The area under the curve (AUC) of the prediction models constructed by multivariate logistic regression analysis and machine learning (RandomForest + TomekLinks) were 0.7770 and 0.8590, respectively. The top 10 most important variables of random forest variables were selected to build a prediction model, with an AUC of 0.8210. Compared with multivariate logistic regression, machine learning models could more accurately predict severe COVID-19 in children with Omicron variant infection.

Derivation and external validation of a nomogram predicting the occurrence of severe illness among hospitalized coronavirus disease 2019 patients: a 2020 Chinese multicenter retrospective study

Background

The worldwide pandemic of coronavirus disease 2019 (COVID-19) has still been an overwhelming public health challenge, and it is vital to identify determinants early to forecast the risk of severity using indicators easily available at admission. The current multicenter retrospective study aimed to derive and validate a user-friendly and effective nomogram to address this issue.

Methods

A training cohort consisting of 437 confirmed COVID-19 cases from three hospitals in Hubei province (Tongji Hospital affiliated with Huazhong University of Science and Technology, Wuhan Third Hospital of Wuhan University and Wuhan Jinyintan Hospital in Hubei province) was retrospectively analyzed to construct a predicting model, and another cohort of 161 hospitalized patients from Public Health Clinical Center of Shanghai was selected as an external validation cohort from January 1, 2020 to March 8, 2020. Determinants of developing into severe COVID-19 were probed using univariate regression together with a multivariate stepwise regression model. The risk of progression to severe COVID-19 was forecasted using the derived nomogram. The performances of the nomogram regarding the discrimination and calibration were assessed in the cohort of training as well as the cohort of external validation, respectively.

Results

A total of 144 (32.95%) and 54 (33.54%) patients, respectively, in cohorts of training and validation progressed to severe COVID-19 during hospitalization. Multivariable analyses showed determinants of severity consisted of hypertension, shortness of breath, platelet count, alanine aminotransferase (ALT), potassium, cardiac troponin I (cTnI), myohemoglobin, procalcitonin (PCT) and intervals from onset to diagnosis. The nomogram had good discrimination with concordance indices being 0.887 (95% CI: 0.854-0.919) and 0.850 (95% CI: 0.815-0.885) in internal and external validation, respectively. Calibration curves exhibited excellent concordance between the predictions by nomogram and actual observations in two cohorts.

Conclusions

We have established and validated an early predicting nomogram model, which can contribute to determine COVID-19 cases at risk of progression to severe illness.

Using artificial intelligence algorithms to predict the overall survival of hemodialysis patients during the COVID-19 pandemic: A prospective cohort study

BACKGROUND

Hemodialysis (HD) patients are a vulnerable population at high risk for severe complications from COVID-19. The impact of partial COVID-19 vaccination on the survival of HD patients remains uncertain. This prospective cohort study was designed to use artificial intelligence algorithms to predict the survival impact of partial COVID-19 vaccination in HD patients.

METHODS

A cohort of 433 HD patients was used to develop machine-learning models based on a subset of clinical features assessed between July 1, 2021, and April 29, 2022. The patient cohort was randomly split into training (80%) and testing (20%) sets for model development and evaluation. Machine-learning models, including categorical boosting (CatBoost), light gradient boosting machines (LightGBM), RandomForest, and extreme gradient boosting models (XGBoost), were applied to evaluate their discriminative performance using the patient cohorts.

RESULTS

Among these models, LightGBM achieved the highest F1 score of 0.95, followed by CatBoost, RandomForest, and XGBoost, with area under the receiver operating characteristic curve values of 0.94 on the testing dataset. The SHapley Additive explanation summary plot derived from the XGBoost model indicated that key features such as age, albumin, and vaccination details had a significant impact on survival. Furthermore, the fully vaccinated group exhibited higher levels of anti-spike (S) receptor-binding domain antibodies.

CONCLUSION

This prospective cohort study involved using artificial intelligence algorithms to predict overall survival in HD patients during the COVID-19 pandemic. These predictive models assisted in identifying high-risk individuals and guiding vaccination strategies for HD patients, ultimately improving overall prognosis. Further research is warranted to validate and refine these predictive models in larger and more diverse populations of HD patients.

Discovering associations between radiological features and COVID-19 patients' deterioration

Background and Aims

Data mining methods are effective and well-known tools for developing predictive models and extracting useful information from various data of patients. The present study aimed to predict the severity of patients with COVID-19 by applying the rule mining method using characteristics of medical images.

Methods

This retrospective study has analyzed the radiological data from 104 COVID-19 hospitalized patients diagnosed with COVID-19 in a hospital in Iran. A data set containing 75 binary features was generated. Apriori method is utilized for association rule mining on this data set. Only rules with confidence equal to one were generated. The performance of rules is calculated by support, coverage, and lift indexes.

Results

Ten rules were extracted with only X-ray-related features on cases referred to ICU. The Support and Coverage index of all of these rules was 0.087, and the Lift index of them was 1.58. Thirteen rules were extracted from only CT scan-related features on cases referred to ICU. The CXR_Pleural effusion feature has appeared in all the rules. The CXR_Left upper zone feature appears in 9 rules out of 10. The Support and Coverage index of all rules was 0.15, and the Lift index of all rules was 1.63. the CT_Adjacent pleura thickening feature has appeared in all rules, and the CT_Right middle lobe appeared in 9 rules out of 13.

Conclusion

This study could reveal the application and efficacy of CXR and CT scan imaging modalities in predicting ICU admission to a major COVID-19 infection via data mining methods. The findings of this study could help data scientists, radiologists, and clinicians in the future development and implementation of these methods in similar conditions and timely and appropriately save patients from adverse disease outcomes.

Machine learning-based prediction of COVID-19 mortality using immunological and metabolic biomarkers

COVID-19 mortality prediction Background COVID-19 has become a major global public health problem, despite prevention and efforts. The daily number of COVID-19 cases rapidly increases, and the time and financial costs associated with testing procedure are burdensome. Method To overcome this, we aim to identify immunological and metabolic biomarkers to predict COVID-19 mortality using a machine learning model. We included inpatients from Hong Kong's public hospitals between January 1, and September 30, 2020, who were diagnosed with COVID-19 using RT-PCR. We developed three machine learning models to predict the mortality of COVID-19 patients based on data in their electronic medical records. We performed statistical analysis to compare the trained machine learning models which are Deep Neural Networks (DNN), Random Forest Classifier (RF) and Support Vector Machine (SVM) using data from a cohort of 5,059 patients (median age = 46 years; 49.3% male) who had tested positive for COVID-19 based on electronic health records and data from 532,427 patients as controls. Result We identified top 20 immunological and metabolic biomarkers that can accurately predict the risk of mortality from COVID-19 with ROC-AUC of 0.98 (95% CI 0.96-0.98). Of the three models used, our result demonstrate that the random forest (RF) model achieved the most accurate prediction of mortality among COVID-19 patients with age, glomerular filtration, albumin, urea, procalcitonin, c-reactive protein, oxygen, bicarbonate, carbon dioxide, ferritin, glucose, erythrocytes, creatinine, lymphocytes, PH of blood and leukocytes among the most important biomarkers identified. A cohort from Kwong Wah Hospital (131 patients) was used for model validation with ROC-AUC of 0.90 (95% CI 0.84-0.92). Conclusion We recommend physicians closely monitor hematological, coagulation, cardiac, hepatic, renal and inflammatory factors for potential progression to severe conditions among COVID-19 patients. To the best of our knowledge, no previous research has identified important immunological and metabolic biomarkers to the extent demonstrated in our study.

Supplementary Information

The online version contains supplementary material available at 10.1186/s44247-022-00001-0.

Identification of Molecular Mechanisms Involved in Viral Infection Progression Based on Text Mining: Case Study for HIV Infection

Viruses cause various infections that may affect human lifestyle for durations ranging from several days to for many years. Although preventative and therapeutic remedies are available for many viruses, they may still have a profound impact on human life. The human immunodeficiency virus type 1 is the most common cause of HIV infection, which represents one of the most dangerous and complex diseases since it affects the immune system and causes its disruption, leading to secondary complications and negatively influencing health-related quality of life. While highly active antiretroviral therapy may decrease the viral load and the velocity of HIV infection progression, some individual peculiarities may affect viral load control or the progression of T-cell malfunction induced by HIV. Our study is aimed at the text-based identification of molecular mechanisms that may be involved in viral infection progression, using HIV as a case study. Specifically, we identified human proteins and genes which commonly occurred, overexpressed or underexpressed, in the collections of publications relevant to (i) HIV infection progression and (ii) acute and chronic stages of HIV infection. Then, we considered biological processes that are controlled by the identified protein and genes. We verified the impact of the identified molecules in the associated clinical study.

Interpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, Iran

BACKGROUND

The high number of COVID-19 deaths is a serious threat to the world. Demographic and clinical biomarkers are significantly associated with the mortality risk of this disease. This study aimed to implement Generalized Neural Additive Model (GNAM) as an interpretable machine learning method to predict the COVID-19 mortality of patients.

METHODS

This cohort study included 2181 COVID-19 patients admitted from February 2020 to July 2021 in Sina and Besat hospitals in Hamadan, west of Iran. A total of 22 baseline features including patients' demographic information and clinical biomarkers were collected. Four strategies including removing missing values, mean, K-Nearest Neighbor (KNN), and Multivariate Imputation by Chained Equations (MICE) imputation methods were used to deal with missing data. Firstly, the important features for predicting binary outcome (1: death, 0: recovery) were selected using the Random Forest (RF) method. Also, synthetic minority over-sampling technique (SMOTE) method was used for handling imbalanced data. Next, considering the selected features, the predictive performance of GNAM for predicting mortality outcome was compared with logistic regression, RF, generalized additive model (GAMs), gradient boosting decision tree (GBDT), and deep neural networks (DNNs) classification models. Each model trained on fifty different subsets of a train-test dataset to ensure a model performance. The average accuracy, F1-score and area under the curve (AUC) evaluation indices were used for comparison of the predictive performance of the models.

RESULTS

Out of the 2181 COVID-19 patients, 624 died during hospitalization and 1557 recovered. The missing rate was 3 percent for each patient. The mean age of dead patients (71.17 ± 14.44 years) was statistically significant higher than recovered patients (58.25 ± 16.52 years). Based on RF, 10 features with the highest relative importance were selected as the best influential features; including blood urea nitrogen (BUN), lymphocytes (Lym), age, blood sugar (BS), serum glutamic-oxaloacetic transaminase (SGOT), monocytes (Mono), blood creatinine (CR), neutrophils (NUT), alkaline phosphatase (ALP) and hematocrit (HCT). The results of predictive performance comparisons showed GNAM with the mean accuracy, F1-score, and mean AUC in the test dataset of 0.847, 0.691, and 0.774, respectively, had the best performance. The smooth function graphs learned from the GNAM were descending for the Lym and ascending for the other important features.

CONCLUSIONS

Interpretable GNAM can perform well in predicting the mortality of COVID-19 patients. Therefore, the use of such a reliable model can help physicians to prioritize some important demographic and clinical biomarkers by identifying the effective features and the type of predictive trend in disease progression.

Influence of Co-morbidities During SARS-CoV-2 Infection in an Indian Population

Background

Since the outbreak of COVID-19 pandemic the interindividual variability in the course of the disease has been reported, indicating a wide range of factors influencing it. Factors which were the most often associated with increased COVID-19 severity include higher age, obesity and diabetes. The influence of cytokine storm is complex, reflecting the complexity of the immunological processes triggered by SARS-CoV-2 infection. A modern challenge such as a worldwide pandemic requires modern solutions, which in this case is harnessing the machine learning for the purpose of analysing the differences in the clinical properties of the populations affected by the disease, followed by grading its significance, consequently leading to creation of tool applicable for assessing the individual risk of SARS-CoV-2 infection.

Methods

Biochemical and morphological parameters values of 5,000 patients (Curisin Healthcare (India) were gathered and used for calculation of eGFR, SII index and N/L ratio. Spearman's rank correlation coefficient formula was used for assessment of correlations between each of the features in the population and the presence of the SARS-CoV-2 infection. Feature importance was evaluated by fitting a Random Forest machine learning model to the data and examining their predictive value. Its accuracy was measured as the F1 Score.

Results

The parameters which showed the highest correlation coefficient were age, random serum glucose, serum urea, gender and serum cholesterol, whereas the highest inverse correlation coefficient was assessed for alanine transaminase, red blood cells count and serum creatinine. The accuracy of created model for differentiating positive from negative SARS-CoV-2 cases was 97%. Features of highest importance were age, alanine transaminase, random serum glucose and red blood cells count.

Conclusion

The current analysis indicates a number of parameters available for a routine screening in clinical setting. It also presents a tool created on the basis of these parameters, useful for assessing the individual risk of developing COVID-19 in patients. The limitation of the study is the demographic specificity of the studied population, which might restrict its general applicability.

Prediction of clinical outcome in CLL based on recurrent gene mutations, CLL-IPI variables, and (para)clinical data

A highly variable clinical course, immune dysfunction, and a complex genetic blueprint pose challenges for treatment decisions and the management of risk of infection in patients with chronic lymphocytic leukemia (CLL). In recent years, the use of machine learning (ML) technologies has made it possible to attempt to untangle such heterogeneous disease entities. In this study, using 3 classes of variables (international prognostic index for CLL [CLL-IPI] variables, baseline [para]clinical data, and data on recurrent gene mutations), we built ML predictive models to identify the individual risk of 4 clinical outcomes: death, treatment, infection, and the combined outcome of treatment or infection. Using the predictive models, we assessed to what extent the different classes of variables are predictive of the 4 different outcomes, within both a short-term 2-year outlook and a long-term 5-year outlook after CLL diagnosis. By adding the baseline (para)clinical data to CLL-IPI variables, predictive performance was improved, whereas no further improvement was observed when including the data on recurrent genetic mutations. We discovered 2 main clusters of variables predictive of treatment and infection. Further emphasizing the high mortality resulting from infection in CLL, we found a close similarity between variables predictive of infection in the short-term outlook and those predictive of death in the long-term outlook. We conclude that at the time of CLL diagnosis, routine (para)clinical data are more predictive of patient outcome than recurrent mutations. Future studies on modeling genetics and clinical outcome should always consider the inclusion of several (para)clinical data to improve performance.

Classifying the Mortality of People with Underlying Health Conditions Affected by COVID-19 Using Machine Learning Techniques

The COVID-19 pandemic has greatly affected populations worldwide and has posed a significant challenge to medical systems. With the constant increase in the number of severe COVID-19 infections, an essential area of research has been directed towards predicting the mortality rate of these patients, in order to make informed medical decisions about the necessary healthcare priorities. Although a large amount of research has attempted to predict the mortality rate of COVID-19 patients, the association between the mortality rate of COVID-19 patients and their underlying health conditions has been given significantly less attention. Meanwhile, patients with underlying conditions often face a worse COVID-19 prognosis. Therefore, the goal of this study was to classify the mortality rate of patients diagnosed with COVID-19, who also suffer from underlying health conditions or comorbidities. To achieve our goal, we applied machine learning (ML) models on a new publicly available dataset, not investigated by any existing literature. The dataset provides detailed information on 582 COVID-19 patients and facilitates a robust forecasting model of the mortality rate. The dataset was analysed using seven ML classifiers, namely, Bagging, J48, logistic regression (LR), random forest (RF), support vector machine (SVM), naïve Bayes (NB), and threshold selector. A comparative analysis was performed across the seven ML techniques, and their performance was assessed based on evaluation parameters including classification accuracy, true-positive rate, and false-positive rate. The best performance was demonstrated by the Bagging algorithm with an accuracy of 83.55% when using all the dataset features. The findings are intended to assist researchers and physicians in the early identification of at-risk COVID-19 patients and to make the appropriate intensive care decisions.

COVID-19 Diagnosis from Chest X-ray Images Using a Robust Multi-Resolution Analysis Siamese Neural Network with Super-Resolution Convolutional Neural Network

Chest X-ray (CXR) is becoming a useful method in the evaluation of coronavirus disease 19 (COVID-19). Despite the global spread of COVID-19, utilizing a computer-aided diagnosis approach for COVID-19 classification based on CXR images could significantly reduce the clinician burden. There is no doubt that low resolution, noise and irrelevant annotations in chest X-ray images are a major constraint to the performance of AI-based COVID-19 diagnosis. While a few studies have made huge progress, they underestimate these bottlenecks. In this study, we propose a super-resolution-based Siamese wavelet multi-resolution convolutional neural network called COVID-SRWCNN for COVID-19 classification using chest X-ray images. Concretely, we first reconstruct high-resolution (HR) counterparts from low-resolution (LR) CXR images in order to enhance the quality of the dataset for improved performance of our model by proposing a novel enhanced fast super-resolution convolutional neural network (EFSRCNN) to capture texture details in each given chest X-ray image. Exploiting a mutual learning approach, the HR images are passed to the proposed Siamese wavelet multi-resolution convolutional neural network to learn the high-level features for COVID-19 classification. We validate the proposed COVID-SRWCNN model on public-source datasets, achieving accuracy of 98.98%. Our screening technique achieves 98.96% AUC, 99.78% sensitivity, 98.53% precision, and 98.86% specificity. Owing to the fact that COVID-19 chest X-ray datasets are low in quality, experimental results show that our proposed algorithm obtains up-to-date performance that is useful for COVID-19 screening.

Deep Learning and Medical Image Analysis for COVID-19 Diagnosis and Prediction

The coronavirus disease 2019 (COVID-19) pandemic has imposed dramatic challenges to health-care organizations worldwide. To combat the global crisis, the use of thoracic imaging has played a major role in diagnosis, prediction, and management for COVID-19 patients with moderate to severe symptoms or with evidence of worsening respiratory status. In response, the medical image analysis community acted quickly to develop and disseminate deep learning models and tools to meet the urgent need of managing and interpreting large amounts of COVID-19 imaging data. This review aims to not only summarize existing deep learning and medical image analysis methods but also offer in-depth discussions and recommendations for future investigations. We believe that the wide availability of high-quality, curated, and benchmarked COVID-19 imaging data sets offers the great promise of a transformative test bed to develop, validate, and disseminate novel deep learning methods in the frontiers of data science and artificial intelligence. Expected final online publication date for the Annual Review of Biomedical Engineering, Volume 24 is June 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

A nomogram predicting the severity of COVID-19 based on initial clinical and radiologic characteristics

Aim: This study aimed to build an easy-to-use nomogram to predict the severity of COVID-19. Patients & methods: From December 2019 to January 2020, patients confirmed with COVID-19 in our hospital were enrolled. The initial clinical and radiological characteristics were extracted. Univariate and multivariate logistic regression were used to identify variables for the nomogram. Results: In total, 104 patients were included. Based on statistical analysis, age, levels of neutrophil count, creatinine, procalcitonin and numbers of involved lung segments were identified for nomogram. The area under the curve was 0.939 (95% CI: 0.893–0.984). The calibration curve showed good agreement between prediction of nomogram and observation in the primary cohort. Conclusion: An easy-to-use nomogram with great discrimination was built to predict the severity of COVID-19.

Comparing different machine learning techniques for predicting COVID-19 severity

Background

Coronavirus disease 2019 (COVID-19) is still ongoing spreading globally, machine learning techniques were used in disease diagnosis and to predict treatment outcomes, which showed favorable performance. The present study aims to predict COVID-19 severity at admission by different machine learning techniques including random forest (RF), support vector machine (SVM), and logistic regression (LR). Feature importance to COVID-19 severity were further identified.

Methods

A retrospective design was adopted in the JinYinTan Hospital from January 26 to March 28, 2020, eighty-six demographic, clinical, and laboratory features were selected with LassoCV method, Spearman’s rank correlation, experts’ opinions, and literature evaluation. RF, SVM, and LR were performed to predict severe COVID-19, the performance of the models was compared by the area under curve (AUC). Additionally, feature importance to COVID-19 severity were analyzed by the best performance model.

Results

A total of 287 patients were enrolled with 36.6% severe cases and 63.4% non-severe cases. The median age was 60.0 years (interquartile range: 49.0–68.0 years). Three models were established using 23 features including 1 clinical, 1 chest computed tomography (CT) and 21 laboratory features. Among three models, RF yielded better overall performance with the highest AUC of 0.970 than SVM of 0.948 and LR of 0.928, RF also achieved a favorable sensitivity of 96.7%, specificity of 69.5%, and accuracy of 84.5%. SVM had sensitivity of 93.9%, specificity of 79.0%, and accuracy of 88.5%. LR also achieved a favorable sensitivity of 92.3%, specificity of 72.3%, and accuracy of 85.2%. Additionally, chest-CT had highest importance to illness severity, and the following features were neutrophil to lymphocyte ratio, lactate dehydrogenase, and D-dimer, respectively.

Conclusions

Our results indicated that RF could be a useful predictive tool to identify patients with severe COVID-19, which may facilitate effective care and further optimize resources.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s40249-022-00946-4.

ICU admission and mortality classifiers for COVID-19 patients based on subgroups of dynamically associated profiles across multiple timepoints

The coronavirus disease 2019 (COVID-19) which is caused by severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) is consistently causing profound wounds in the global healthcare system due to its increased transmissibility. Currently, there is an urgent unmet need to identify the underlying dynamic associations among COVID-19 patients and distinguish patient subgroups with common clinical profiles towards the development of robust classifiers for ICU admission and mortality. To address this need, we propose a four step pipeline which: (i) enhances the quality of multiple timeseries clinical data through an automated data curation workflow, (ii) deploys Dynamic Bayesian Networks (DBNs) for the detection of features with increased connectivity based on dynamic association analysis across multiple points, (iii) utilizes Self Organizing Maps (SOMs) and trajectory analysis for the early identification of COVID-19 patients with common clinical profiles, and (iv) trains robust multiple additive regression trees (MART) for ICU admission and mortality classification based on the extracted homogeneous clusters, to identify risk factors and biomarkers for disease progression. The contribution of the extracted clusters and the dynamically associated clinical data improved the classification performance for ICU admission to sensitivity 0.83 and specificity 0.83, and for mortality to sensitivity 0.74 and specificity 0.76. Additional information was included to enhance the performance of the classifiers yielding an increase by 4% in sensitivity and specificity for mortality. According to the risk factor analysis, the number of lymphocytes, SatO2, PO2/FiO2, and O2 supply type were highlighted as risk factors for ICU admission and the percentage of neutrophils and lymphocytes, PO2/FiO2, LDH, and ALP for mortality, among others. To our knowledge, this is the first study that combines dynamic modeling with clustering analysis to identify homogeneous groups of COVID-19 patients towards the development of robust classifiers for ICU admission and mortality.

A Multimodal Approach for the Risk Prediction of Intensive Care and Mortality in Patients with COVID-19

BACKGROUND

Although several studies have been launched towards the prediction of risk factors for mortality and admission in the intensive care unit (ICU) in COVID-19, none of them focuses on the development of explainable AI models to define an ICU scoring index using dynamically associated biological markers.

METHODS

We propose a multimodal approach which combines explainable AI models with dynamic modeling methods to shed light into the clinical features of COVID-19. Dynamic Bayesian networks were used to seek associations among cytokines across four time intervals after hospitalization. Explainable gradient boosting trees were trained to predict the risk for ICU admission and mortality towards the development of an ICU scoring index.

RESULTS

Our results highlight LDH, IL-6, IL-8, Cr, number of monocytes, lymphocyte count, TNF as risk predictors for ICU admission and survival along with LDH, age, CRP, Cr, WBC, lymphocyte count for mortality in the ICU, with prediction accuracy 0.79 and 0.81, respectively. These risk factors were combined with dynamically associated biological markers to develop an ICU scoring index with accuracy 0.9.

CONCLUSIONS

to our knowledge, this is the first multimodal and explainable AI model which quantifies the risk of intensive care with accuracy up to 0.9 across multiple timepoints.

Machine-learning-based COVID-19 mortality prediction model and identification of patients at low and high risk of dying

Background

The coronavirus disease 2019 (COVID-19) pandemic caused by the SARS-Cov2 virus has become the greatest health and controversial issue for worldwide nations. It is associated with different clinical manifestations and a high mortality rate. Predicting mortality and identifying outcome predictors are crucial for COVID patients who are critically ill. Multivariate and machine learning methods may be used for developing prediction models and reduce the complexity of clinical phenotypes.

Methods

Multivariate predictive analysis was applied to 108 out of 250 clinical features, comorbidities, and blood markers captured at the admission time from a hospitalized cohort of patients (N = 250) with COVID-19. Inspired modification of partial least square (SIMPLS)-based model was developed to predict hospital mortality. Prediction accuracy was randomly assigned to training and validation sets. Predictive partition analysis was performed to obtain cutting value for either continuous or categorical variables. Latent class analysis (LCA) was carried to cluster the patients with COVID-19 to identify low- and high-risk patients. Principal component analysis and LCA were used to find a subgroup of survivors that tends to die.

Results

SIMPLS-based model was able to predict hospital mortality in patients with COVID-19 with moderate predictive power (Q² = 0.24) and high accuracy (AUC > 0.85) through separating non-survivors from survivors developed using training and validation sets. This model was obtained by the 18 clinical and comorbidities predictors and 3 blood biochemical markers. Coronary artery disease, diabetes, Altered Mental Status, age > 65, and dementia were the topmost differentiating mortality predictors. CRP, prothrombin, and lactate were the most differentiating biochemical markers in the mortality prediction model. Clustering analysis identified high- and low-risk patients among COVID-19 survivors.

Conclusions

An accurate COVID-19 mortality prediction model among hospitalized patients based on the clinical features and comorbidities may play a beneficial role in the clinical setting to better management of patients with COVID-19. The current study revealed the application of machine-learning-based approaches to predict hospital mortality in patients with COVID-19 and identification of most important predictors from clinical, comorbidities and blood biochemical variables as well as recognizing high- and low-risk COVID-19 survivors.