Machine learning in predicting respiratory failure in patients with COVID-19 pneumonia-Challenges, strengths, and opportunities in a global health emergency

Feature Identification Using Interpretability Machine Learning Predicting Risk Factors for Disease Severity of In-Patients with COVID-19 in South Florida

Objective: The objective of the study was to establish an AI-driven decision support system by identifying the most important features in the severity of disease for Intensive Care Unit (ICU) with Mechanical Ventilation (MV) requirement, ICU, and InterMediate Care Unit (IMCU) admission for hospitalized patients with COVID-19 in South Florida. The features implicated in the risk factors identified by the model interpretability can be used to forecast treatment plans faster before critical conditions exacerbate. Methods: We analyzed eHR data from 5371 patients diagnosed with COVID-19 from South Florida Memorial Healthcare Systems admitted between March 2020 and January 2021 to predict the need for ICU with MV, ICU, and IMCU admission. A Random Forest classifier was trained on patients' data augmented by SMOTE, collected at hospital admission. We then compared the importance of features utilizing different model interpretability analyses, such as SHAP, MDI, and Permutation Importance. Results: The models for ICU with MV, ICU, and IMCU admission identified the following factors overlapping as the most important predictors among the three outcomes: age, race, sex, BMI, diarrhea, diabetes, hypertension, early stages of kidney disease, and pneumonia. It was observed that individuals over 65 years ('older adults'), males, current smokers, and BMI classified as 'overweight' and 'obese' were at greater risk of severity of illness. The severity was intensified by the co-occurrence of two interacting features (e.g., diarrhea and diabetes). Conclusions: The top features identified by the models' interpretability were from the 'sociodemographic characteristics', 'pre-hospital comorbidities', and 'medications' categories. However, 'pre-hospital comorbidities' played a vital role in different critical conditions. In addition to individual feature importance, the feature interactions also provide crucial information for predicting the most likely outcome of patients' conditions when urgent treatment plans are needed during the surge of patients during the pandemic.

Understanding contextual and practical factors to inform WHO recommendations on using chest imaging to monitor COVID-19 pulmonary sequelae: a qualitative study exploring stakeholders' perspective

BACKGROUND

A recommendation by the World Health Organization (WHO) was issued about the use of chest imaging to monitor pulmonary sequelae following recovery from COVID-19. This qualitative study aimed to explore the perspective of key stakeholders to understand their valuation of the outcome of the proposition, preferences for the modalities of chest imaging, acceptability, feasibility, impact on equity and practical considerations influencing the implementation of using chest imaging.

METHODS

A qualitative descriptive design using in-depth interviews approach. Key stakeholders included adult patients who recovered from the acute illness of COVID-19, and providers caring for those patients. The Evidence to Decision (EtD) conceptual framework was used to guide data collection of contextual and practical factors related to monitoring using imaging. Data analysis was based on the framework thematic analysis approach.

RESULTS

33 respondents, including providers and patients, were recruited from 15 different countries. Participants highly valued the ability to monitor progression and resolution of long-term sequelae but recommended the avoidance of overuse of imaging. Their preferences for the imaging modalities were recorded along with pros and cons. Equity concerns were reported across countries (e.g., access to resources) and within countries (e.g., disadvantaged groups lacked access to insurance). Both providers and patients accepted the use of imaging, some patients were concerned about affordability of the test. Facilitators included post- recovery units and protocols. Barriers to feasibility included low number of specialists in some countries, access to imaging tests among elderly living in nursing homes, experience of poor coordination of care, emotional exhaustion, and transportation challenges driving to a monitoring site.

CONCLUSION

We were able to demonstrate that there is a high value and acceptability using imaging but there were factors influencing feasibility, equity and some practical considerations associated with implementation. We had a few suggestions to be considered by the expert panel in the formulation of the guideline to facilitate its implementation such as using validated risk score predictive tools for lung complications to recommend the appropriate imaging modality and complementary pulmonary function test.

Predictive health monitoring: Leveraging artificial intelligence for early detection of infectious diseases in nursing home residents through discontinuous vital signs analysis

This research addresses the problem of detecting acute respiratory, urinary tract, and other infectious diseases in elderly nursing home residents using machine learning algorithms. The study analyzes data extracted from multiple vital signs and other contextual information for diagnostic purposes. The daily data collection process encounters sampling constraints due to weekends, holidays, shift changes, staff turnover, and equipment breakdowns, resulting in numerous nulls, repeated readings, outliers, and meaningless values. The short time series generated also pose a challenge to analysis, preventing the extraction of seasonal information or consistent trends. Blind data collection results in most of the data coming from periods when residents are healthy, resulting in excessively imbalanced data. This study proposes a data cleaning process and then builds a mechanism that reproduces the basal activity of the residents to improve the classification of the disease. The results show that the proposed basal module-assisted machine learning techniques allow anticipating diagnostics 2, 3 or 4 days before doctors decide to start treatment with antibiotics, achieving a performance measured by the area-under-the-curve metric of 0.857. The contributions of this work are: (1) a new data cleaning process; (2) the analysis of contextual information to improve data quality; (3) the generation of a baseline measure for relative comparison; and (4) the use of either binary (disease/no disease) or multiclass classification, differentiating among types of infections and showing the advantages of multiclass versus binary classification. From a medical point of view, the anticipated detection of infectious diseases in institutionalized individuals is brand new.

Reintubation After Lung Cancer Resection: Development and External Validation of a Predictive Score

BACKGROUND

Reintubation after lung cancer resection is an important quality metric because of increased disability, mortality and cost. However, no validated predictive instrument is in use to reduce reintubation after lung resection. This study aimed to create and validate the PRediction Of REintubation After Lung cancer resection (PROREAL) score.

METHODS

The study analyzed lung resection cases from 2 university hospitals. The primary end point was reintubation within 7 days after surgery. Predictors were selected through backward stepwise logistic regression and bootstrap resampling. The investigators used reclassification and receiver-operating characteristic (ROC) curve analyses to assess score performance and compare it with an established score for all surgical patients (Score for Prediction of Postoperative Respiratory Complications [SPORC]).

RESULTS

The study included 2672 patients who underwent resection for lung cancer (1754, development cohort; 918, validation cohort) between 2008 and 2020, of whom 71 (2.7%) were reintubated within 7 days after surgery. Identified score variables were surgical extent and approach, American Society of Anesthesiologists physical status, heart failure, renal disease, and diffusing capacity of the lung for carbon monoxide. The score achieved excellent discrimination in the development cohort (ROC AUC, 0.90; 95% CI, 0.87-0.94) and good discrimination in the validation cohort (ROC AUC, 0.74, 95% CI; 0.66-0.82), thus outperforming the SPORC in both cohorts (P < .001 and P = .018, respectively; validation cohort net reclassification improvement, 0.39; 95% CI, 0.18-0.60; P = .001). The score cutoff of ≥5 yielded a sensitivity of 88% (95% CI, 72-95) and a specificity of 81% (95% CI,79-83) in the development cohort.

CONCLUSIONS

A simple score (PROREAL) specific to lung cancer predicts postoperative reintubation more accurately than the nonspecific SPORC score. Operative candidates at risk may be identified for preventive intervention or alternative oncologic therapy.

Making the Improbable Possible: Generalizing Models Designed for a Syndrome-Based, Heterogeneous Patient Landscape

Syndromic conditions, such as sepsis, are commonly encountered in the intensive care unit. Although these conditions are easy for clinicians to grasp, these conditions may limit the performance of machine-learning algorithms. Individual hospital practice patterns may limit external generalizability. Data missingness is another barrier to optimal algorithm performance and various strategies exist to mitigate this. Recent advances in data science, such as transfer learning, conformal prediction, and continual learning, may improve generalizability of machine-learning algorithms in critically ill patients. Randomized trials with these approaches are indicated to demonstrate improvements in patient-centered outcomes at this point.

A data-driven model for early prediction of need for invasive mechanical ventilation in pediatric intensive care unit patients

RATIONALE

Acute respiratory failure is a life-threatening clinical outcome in critically ill pediatric patients. In severe cases, patients can require mechanical ventilation (MV) for survival. Early recognition of these patients can potentially help clinicians alter the clinical course and lead to improved outcomes.

OBJECTIVES

To build a data-driven model for early prediction of the need for mechanical ventilation in pediatric intensive care unit (PICU) patients.

METHODS

The study consists of a single-center retrospective observational study on a cohort of 13,651 PICU patients admitted between 1/01/2010 and 5/15/2018 with a prevalence of 8.06% for MV due to respiratory failure. XGBoost (extreme gradient boosting) and a convolutional neural network (CNN) using medication history were used to develop a prediction model that could yield a time-varying "risk-score"-a continuous probability of whether a patient will receive MV-and an ideal global threshold was calculated from the receiver operating characteristics (ROC) curve. The early prediction point (EPP) was the first time the risk-score surpassed the optimal threshold, and the interval between the EPP and the start of the MV was the early warning period (EWT). Spectral clustering identified patient groups based on risk-score trajectories after EPP.

RESULTS

A clinical and medication history-based model achieved a 0.89 area under the ROC curve (AUROC), 0.6 sensitivity, 0.95 specificity, 0.55 positive predictive value (PPV), and 0.95 negative predictive value (NPV). Early warning time (EWT) median [inter-quartile range] of this model was 9.9[4.2-69.2] hours. Clustering risk-score trajectories within a six-hour window after the early prediction point (EPP) established three patient groups, with the highest risk group's PPV being 0.92.

CONCLUSIONS

This study uses a unique method to extract and apply medication history information, such as time-varying variables, to identify patients who may need mechanical ventilation for respiratory failure and provide an early warning period to avert it.

Becalm: Intelligent Monitoring of Respiratory Patients

The Becalm project is an open and low-cost solution for the remote monitoring of respiratory support therapies like the ones used in COVID-19 patients. Becalm combines a decision-making system based on Case-Based Reasoning with a low-cost, non-invasive mask that enables the remote monitoring, detection, and explanation of risk situations for respiratory patients. This paper first describes the mask and the sensors that allow remote monitoring. Then, it describes the intelligent decision-making system that detects anomalies and raises early warnings. This detection is based on the comparison of cases that represent patients using a set of static variables plus the dynamic vector of the patient time series from sensors. Finally, personalized visual reports are created to explain the causes of the warning, data patterns, and patient context to the healthcare professional. To evaluate the case-based early-warning system, we use a synthetic data generator that simulates patients' clinical evolution from the physiological features and factors described in healthcare literature. This generation process has been verified with a real dataset and allows the validation of the reasoning system with noisy and incomplete data, threshold values, and life/death situations. The evaluation demonstrates promising results and good accuracy (0.91) for the proposed low-cost solution to monitor respiratory patients.

Using machine learning to identify patient characteristics to predict mortality of in-patients with COVID-19 in south Florida

Introduction

The SARS-CoV-2 (COVID-19) pandemic has created substantial health and economic burdens in the US and worldwide. As new variants continuously emerge, predicting critical clinical events in the context of relevant individual risks is a promising option for reducing the overall burden of COVID-19. This study aims to train an AI-driven decision support system that helps build a model to understand the most important features that predict the "mortality" of patients hospitalized with COVID-19.

Methods

We conducted a retrospective analysis of "5,371" patients hospitalized for COVID-19-related symptoms from the South Florida Memorial Health Care System between March 14th, 2020, and January 16th, 2021. A data set comprising patients' sociodemographic characteristics, pre-existing health information, and medication was analyzed. We trained Random Forest classifier to predict "mortality" for patients hospitalized with COVID-19.

Results

Based on the interpretability of the model, age emerged as the primary predictor of "mortality", followed by diarrhea, diabetes, hypertension, BMI, early stages of kidney disease, smoking status, sex, pneumonia, and race in descending order of importance. Notably, individuals aged over 65 years (referred to as "older adults"), males, Whites, Hispanics, and current smokers were identified as being at higher risk of death. Additionally, BMI, specifically in the overweight and obese categories, significantly predicted "mortality". These findings indicated that the model effectively learned from various categories, such as patients' sociodemographic characteristics, pre-hospital comorbidities, and medications, with a predominant focus on characterizing pre-hospital comorbidities. Consequently, the model demonstrated the ability to predict "mortality" with transparency and reliability.

Conclusion

AI can potentially provide healthcare workers with the ability to stratify patients and streamline optimal care solutions when time is of the essence and resources are limited. This work sets the platform for future work that forecasts patient responses to treatments at various levels of disease severity and assesses health disparities and patient conditions that promote improved health care in a broader context. This study contributed to one of the first predictive analyses applying AI/ML techniques to COVID-19 data using a vast sample from South Florida.

Machine learning algorithm for early-stage prediction of severe morbidity in COVID-19 pneumonia patients based on bio-signals

BACKGROUND

Paralysis of medical systems has emerged as a major problem not only in Korea but also globally because of the COVID-19 pandemic. Therefore, early identification and treatment of COVID-19 are crucial. This study aims to develop a machine-learning algorithm based on bio-signals that predicts the infection three days in advance before it progresses from mild to severe, which may necessitate high-flow oxygen therapy or mechanical ventilation.

METHODS

The study included 2758 hospitalized patients with mild severity COVID-19 between July 2020 and October 2021. Bio-signals, clinical information, and laboratory findings were retrospectively collected from the electronic medical records of patients. Machine learning methods included random forest, random forest ranger, gradient boosting machine, and support vector machine (SVM).

RESULTS

SVM showed the best performance in terms of accuracy, kappa, sensitivity, detection rate, balanced accuracy, and run-time; the area under the receiver operating characteristic curve was also quite high at 0.96. Body temperature and SpO2 three and four days before discharge or exacerbation were ranked high among SVM features.

CONCLUSIONS

The proposed algorithm can predict the exacerbation of severity three days in advance in patients with mild COVID-19. This prediction can help effectively manage the reallocation of appropriate medical resources in clinical settings. Therefore, this algorithm can facilitate adequate oxygen therapy and mechanical ventilator preparation, thereby improving patient prognosis, increasing the efficiency of medical systems, and mitigating the damage caused by a global pandemic.

The innovative model based on artificial intelligence algorithms to predict recurrence risk of patients with postoperative breast cancer

Purpose

This study aimed to develop a machine learning model to retrospectively study and predict the recurrence risk of breast cancer patients after surgery by extracting the clinicopathological features of tumors from unstructured clinical electronic health record (EHR) data.

Methods

This retrospective cohort included 1,841 breast cancer patients who underwent surgical treatment. To extract the principal features associated with recurrence risk, the clinical notes and histopathology reports of patients were collected and feature engineering was used. Predictive models were next conducted based on this important information. All algorithms were implemented using Python software. The accuracy of prediction models was further verified in the test cohort. The area under the curve (AUC), precision, recall, and F1 score were adopted to evaluate the performance of each model.

Results

A training cohort with 1,289 patients and a test cohort with 552 patients were recruited. From 2011 to 2019, a total of 1,841 textual reports were included. For the prediction of recurrence risk, both LSTM, XGBoost, and SVM had favorable accuracies of 0.89, 0.86, and 0.78. The AUC values of the micro-average ROC curve corresponding to LSTM, XGBoost, and SVM were 0.98 ± 0.01, 0.97 ± 0.03, and 0.92 ± 0.06. Especially the LSTM model achieved superior execution than other models. The accuracy, F1 score, macro-avg F1 score (0.87), and weighted-avg F1 score (0.89) of the LSTM model produced higher values. All P values were statistically significant. Patients in the high-risk group predicted by our model performed more resistant to DNA damage and microtubule targeting drugs than those in the intermediate-risk group. The predicted low-risk patients were not statistically significant compared with intermediate- or high-risk patients due to the small sample size (188 low-risk patients were predicted via our model, and only two of them were administered chemotherapy alone after surgery). The prognosis of patients predicted by our model was consistent with the actual follow-up records.

Conclusions

The constructed model accurately predicted the recurrence risk of breast cancer patients from EHR data and certainly evaluated the chemoresistance and prognosis of patients. Therefore, our model can help clinicians to formulate the individualized management of breast cancer patients.

Spacer exchange in persistent periprosthetic joint infection: microbiological evaluation and survivorship analysis

PURPOSE

The purposes of this study were to determine demographics and characteristics of patients who underwent spacer exchange for persistent infection in the setting of two-stage arthroplasty for periprosthetic joint infection, to describe the microbiology of pathogens involved, to analyze survivorship free from infection in these patients.

METHODS

The institutional prospectively collected database was reviewed to enroll patients with minimum 2 years follow-up. Patients who underwent two-stage procedure for septic arthritis were excluded, as were patients who had spacer fracture or dislocation.

RESULTS

A total of 34 patients (41 procedures) were included. Mean age was 65.0 ± 12.8 years. Mean follow-up was 53.4 ± 24.8 months. Mean number of previous procedures was 3.6 ± 1.2. A total of 27 (79.4%) patients underwent final reimplantation. The most frequently isolated pathogen in spacer exchange was Staphylococcus epidermidis (10 cases, 28.6%). Polymicrobial cultures were obtained from 9 (25.71%) patients, 10 (28.6%) presented culture-negative infections. A total of 11 (32.4%) resistant pathogens were isolated, and 16 (47.0%) difficult to treat pathogens were detected. Eradication rate was 78.8%. Overall survivorship of implants after final reimplantation was 72.8% at 51.8 months.

CONCLUSION

Surgeons should be aware that subjects necessitating spacer exchange often present multiple comorbidities, previous staged revision failures, soft-tissue impairment and difficult to treat infection. In these patients, spacer exchange provides good clinical results and infection eradication, preventing arthrodesis or amputation.

SARS-CoV-2 severity classification from plasma sample using confocal Raman spectroscopy

The world is on the brink of facing coronavirus's (COVID-19) fourth wave as the mutant forms of viruses are escaping neutralizing antibodies in spite of being vaccinated. As we have already witnessed that it has encumbered our health system, with hospitals swamped with infected patients observed during the viral outbreak. Rapid triage of patients infected with SARS-CoV-2 is required during hospitalization to prioritize and provide the best point of care. Traditional diagnostics techniques such as RT-PCR and clinical parameters such as symptoms, comorbidities, sex and age are not enough to identify the severity of patients. Here, we investigated the potential of confocal Raman microspectroscopy as a powerful tool to generate an expeditious blood-based test for the classification of COVID-19 disease severity using 65 patients plasma samples from cohorts infected with SARS-CoV-2. We designed an easy manageable blood test where we used a small volume (8 μl) of inactivated whole plasma samples from infected patients without any extra solvent usage in plasma processing. Raman spectra of plasma samples were acquired and multivariate exploratory analysis PC-LDA (principal component based linear discriminant analysis) was used to build a model, which segregated the severe from the non-severe COVID-19 group with a sensitivity of 83.87%, specificity of 70.60% and classification efficiency of 76.92%. Among the bands expressed in both the cohorts, the study led to the identification of Raman fingerprint regions corresponding to lipids (1661, 1742), proteins amide I and amide III (1555, 1247), proteins (Phe) (1006, 1034), and nucleic acids (760) to be differentially expressed in severe COVID-19 patient's samples. In summary, the current study exhibits the potential of confocal Raman to generate simple, rapid, and less expensive blood tests to triage the severity of patients infected with SARS-CoV-2.

Machine learning predicts the short-term requirement for invasive ventilation among Australian critically ill COVID-19 patients

OBJECTIVE(S)

To use machine learning (ML) to predict short-term requirements for invasive ventilation in patients with COVID-19 admitted to Australian intensive care units (ICUs).

DESIGN

A machine learning study within a national ICU COVID-19 registry in Australia.

PARTICIPANTS

Adult patients who were spontaneously breathing and admitted to participating ICUs with laboratory-confirmed COVID-19 from 20 February 2020 to 7 March 2021. Patients intubated on day one of their ICU admission were excluded.

MAIN OUTCOME MEASURES

Six machine learning models predicted the requirement for invasive ventilation by day three of ICU admission from variables recorded on the first calendar day of ICU admission; (1) random forest classifier (RF), (2) decision tree classifier (DT), (3) logistic regression (LR), (4) K neighbours classifier (KNN), (5) support vector machine (SVM), and (6) gradient boosted machine (GBM). Cross-validation was used to assess the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity of machine learning models.

RESULTS

300 ICU admissions collected from 53 ICUs across Australia were included. The median [IQR] age of patients was 59 [50-69] years, 109 (36%) were female and 60 (20%) required invasive ventilation on day two or three. Random forest and Gradient boosted machine were the best performing algorithms, achieving mean (SD) AUCs of 0.69 (0.06) and 0.68 (0.07), and mean sensitivities of 77 (19%) and 81 (17%), respectively.

CONCLUSION

Machine learning can be used to predict subsequent ventilation in patients with COVID-19 who were spontaneously breathing and admitted to Australian ICUs.

Real-world data mining meets clinical practice: Research challenges and perspective

As Big Data Analysis meets healthcare applications, domain-specific challenges and opportunities materialize in all aspects of data science. Advanced statistical methods and Artificial Intelligence (AI) on Electronic Health Records (EHRs) are used both for knowledge discovery purposes and clinical decision support. Such techniques enable the emerging Predictive, Preventative, Personalized, and Participatory Medicine (P4M) paradigm. Working with the Infectious Disease Clinic of the University Hospital of Modena, Italy, we have developed a range of Data-Driven (DD) approaches to solve critical clinical applications using statistics, Machine Learning (ML) and Big Data Analytics on real-world EHR. Here, we describe our perspective on the challenges we encountered. Some are connected to medical data and their sparse, scarce, and unbalanced nature. Others are bound to the application environment, as medical AI tools can affect people's health and life. For each of these problems, we report some available techniques to tackle them, present examples drawn from our experience, and propose which approaches, in our opinion, could lead to successful real-world, end-to-end implementations.

DESY report number

DESY-22-153.

Predicting the Level of Respiratory Support in COVID-19 Patients Using Machine Learning

In this paper, a machine learning-based system for the prediction of the required level of respiratory support in COVID-19 patients is proposed. The level of respiratory support is divided into three classes: class 0 which refers to minimal support, class 1 which refers to non-invasive support, and class 2 which refers to invasive support. A two-stage classification system is built. First, the classification between class 0 and others is performed. Then, the classification between class 1 and class 2 is performed. The system is built using a dataset collected retrospectively from 3491 patients admitted to tertiary care hospitals at the University of Louisville Medical Center. The use of the feature selection method based on analysis of variance is demonstrated in the paper. Furthermore, a dimensionality reduction method called principal component analysis is used. XGBoost classifier achieves the best classification accuracy (84%) in the first stage. It also achieved optimal performance in the second stage, with a classification accuracy of 83%.

A novel machine learning model to predict respiratory failure and invasive mechanical ventilation in critically ill patients suffering from COVID-19

In hypoxemic patients at risk for developing respiratory failure, the decision to initiate invasive mechanical ventilation (IMV) may be extremely difficult, even more so among patients suffering from COVID-19. Delayed recognition of respiratory failure may translate into poor outcomes, emphasizing the need for stronger predictive models for IMV necessity. We developed a two-step model; the first step was to train a machine learning predictive model on a large dataset of non-COVID-19 critically ill hypoxemic patients from the United States (MIMIC-III). The second step was to apply transfer learning and adapt the model to a smaller COVID-19 cohort. An XGBoost algorithm was trained on data from the MIMIC-III database to predict if a patient would require IMV within the next 6, 12, 18 or 24 h. Patients’ datasets were used to construct the model as time series of dynamic measurements and laboratory results obtained during the previous 6 h with additional static variables, applying a sliding time-window once every hour. We validated the adaptation algorithm on a cohort of 1061 COVID-19 patients from a single center in Israel, of whom 160 later deteriorated and required IMV. The new XGBoost model for the prediction of the IMV onset was trained and tested on MIMIC-III data and proved to be predictive, with an AUC of 0.83 on a shortened set of features, excluding the clinician’s settings, and an AUC of 0.91 when the clinician settings were included. Applying these models “as is” (no adaptation applied) on the dataset of COVID-19 patients degraded the prediction results to AUCs of 0.78 and 0.80, without and with the clinician’s settings, respectively. Applying the adaptation on the COVID-19 dataset increased the prediction power to an AUC of 0.94 and 0.97, respectively. Good AUC results get worse with low overall precision. We show that precision of the prediction increased as prediction probability was higher. Our model was successfully trained on a specific dataset, and after adaptation it showed promise in predicting outcome on a completely different dataset. This two-step model successfully predicted the need for invasive mechanical ventilation 6, 12, 18 or 24 h in advance in both general ICU population and COVID-19 patients. Using the prediction probability as an indicator of the precision carries the potential to aid the decision-making process in patients with hypoxemic respiratory failure despite the low overall precision.

An Unsupervised Machine Learning Clustering and Prediction of Differential Clinical Phenotypes of COVID-19 Patients Based on Blood Tests—A Hong Kong Population Study

Background

To better understand the different clinical phenotypes across the disease spectrum in patients with COVID-19 using an unsupervised machine learning clustering approach.

Materials and Methods

A population-based retrospective study was conducted utilizing demographics, clinical characteristics, comorbidities, and clinical outcomes of 7,606 COVID-19–positive patients on admission to public hospitals in Hong Kong in the year 2020. An unsupervised machine learning clustering was used to explore this large cohort.

Results

Four clusters of differing clinical phenotypes based on data at initial admission was derived in which 86.6% of the deceased cases were aggregated in one of the clusters without prior knowledge of their clinical outcomes. Other distinctive clinical characteristics of this cluster were old age and high concurrent comorbidities as well as laboratory characteristics of lower hemoglobin/hematocrit levels, higher neutrophil, C-reactive protein, lactate dehydrogenase, and creatinine levels. The clinical patterns captured by the cluster analysis was validated on other temporally distinct cohorts in 2021. The phenotypes aligned with existing literature.

Conclusion

The study demonstrated the usefulness of unsupervised machine learning techniques with the potential to uncover latent clinical phenotypes. It could serve as a more robust classification for patient triaging and patient-tailored treatment strategies.

Classification of COVID-19 individuals using adaptive neuro-fuzzy inference system

Coronavirus is a fatal disease that affects mammals and birds. Usually, this virus spreads in humans through aerial precipitation of any fluid secreted from the infected entity's body part. This type of virus is fatal than other unpremeditated viruses. Meanwhile, another class of coronavirus was developed in December 2019, named Novel Coronavirus (2019-nCoV), first seen in Wuhan, China. From January 23, 2020, the number of affected individuals from this virus rapidly increased in Wuhan and other countries. This research proposes a system for classifying and analyzing the predictions obtained from symptoms of this virus. The proposed system aims to determine those attributes that help in the early detection of Coronavirus Disease (COVID-19) using the Adaptive Neuro-Fuzzy Inference System (ANFIS). This work computes the accuracy of different machine learning classifiers and selects the best classifier for COVID-19 detection based on comparative analysis. ANFIS is used to model and control ill-defined and uncertain systems to predict this globally spread disease's risk factor. COVID-19 dataset is classified using Support Vector Machine (SVM) because it achieved the highest accuracy of 100% among all classifiers. Furthermore, the ANFIS model is implemented on this classified dataset, which results in an 80% risk prediction for COVID-19.

Coronavirus Disease 2019 as Cause of Viral Sepsis: A Systematic Review and Meta-Analysis

OBJECTIVE

Coronavirus disease 2019 is a heterogeneous disease most frequently causing respiratory tract infection, which can induce respiratory failure and multiple organ dysfunction syndrome in its severe forms. The prevalence of coronavirus disease 2019-related sepsis is still unclear; we aimed to describe this in a systematic review.

DATA SOURCES

MEDLINE (PubMed), Cochrane, and Google Scholar databases were searched based on a prespecified protocol (International Prospective Register for Systematic Reviews: CRD42020202018).

STUDY SELECTION

Studies reporting on patients with confirmed coronavirus disease 2019 diagnosed with sepsis according to sepsis-3 or according to the presence of infection-related organ dysfunctions necessitating organ support/replacement were included in the analysis. The primary end point was prevalence of coronavirus disease 2019-related sepsis among adults hospitalized in the ICU and the general ward. Among secondary end points were the need for ICU admission among patients initially hospitalized in the general ward and the prevalence of new onset of organ dysfunction in the ICU. Outcomes were expressed as proportions with respective 95% CI.

DATA EXTRACTION

Two reviewers independently screened and reviewed existing literature and assessed study quality with the Newcastle-Ottawa Scale and the Methodological index for nonrandomized studies.

DATA SYNTHESIS

Of 3,825 articles, 151 were analyzed, only five of which directly reported sepsis prevalence. Noting the high heterogeneity observed, coronavirus disease 2019-related sepsis prevalence was 77.9% (95% CI, 75.9-79.8; I2 = 91%; 57 studies) in the ICU, and 33.3% (95% CI, 30.3-36.4; I2 = 99%; 86 studies) in the general ward. ICU admission was required for 17.7% (95% CI, 12.9-23.6; I2 = 100%) of ward patients. Acute respiratory distress syndrome was the most common organ dysfunction in the ICU (87.5%; 95% CI, 83.3-90.7; I2 = 98%).

CONCLUSIONS

The majority of coronavirus disease 2019 patients hospitalized in the ICU meet Sepsis-3 criteria and present infection-associated organ dysfunction. The medical and scientific community should be aware and systematically report viral sepsis for prognostic and treatment implications.

Novel Application of Machine Learning Algorithms and Model-Agnostic Methods to Identify Factors Influencing Childhood Blood Lead Levels

Blood lead (Pb) poisoning remains a global concern, particularly for children in their early developmental years. Broken Hill is Australia's oldest operating silver-zinc-lead mine. In this study, we utilized recent advances in machine learning to assess multiple algorithms and identify the most optimal model for predicting childhood blood Pb levels (BLL) using Broken Hill children's (<5 years of age) data (n = 23,749) from 1991 to 2015, combined with demographic, socio-economic, and environmental influencing factors. We applied model-agnostic methods to interpret the most optimal model, investigating different environmental and human factors influencing childhood BLL. Algorithm assessment showed that stacked ensemble, a method for automatically and optimally combining multiple prediction algorithms, enhanced predictive performance by 1.1% with respect to mean absolute error (p < 0.01) and 2.6% for root-mean-squared error (p < 0.01) compared to the best performing constituent algorithm (random forest). By interpreting the model, the following information was acquired: children had higher BLL if they resided within 1.0 km to the central mine area or 1.37 km to the railroad; year of testing had the greatest interactive strength with all other factors; BLL increased faster in Aboriginal than in non-Aboriginal children at 9-10 and 12-18 months of age. This "stacked ensemble + model-agnostic interpretation" framework achieved both prediction accuracy and model interpretability, identifying previously unconnected variables associated with elevated childhood BLL, offering a marked advantage over previous works. Thus, this approach has a clear value and potential for application to other environmental health issues.

Rapid Classification of COVID-19 Severity by ATR-FTIR Spectroscopy of Plasma Samples

The coronavirus disease 2019 (COVID-19) pandemic continues to ravage the world, with many hospitals overwhelmed by the large number of patients presenting during major outbreaks. A rapid triage for COVID-19 patient requiring hospitalization and intensive care is urgently needed. Age and comorbidities have been associated with a higher risk of severe COVID-19 but are not sufficient to triage patients. Here, we investigated the potential of attenuated total reflectance Fourier-transform infrared (ATR-FTIR) spectroscopy as a rapid blood test for classification of COVID-19 disease severity using a cohort of 160 COVID-19 patients. A simple plasma processing and ATR-FTIR data acquisition procedure was established using 75% ethanol for viral inactivation. Next, partial least-squares-discriminant analysis (PLS-DA) models were developed and tested using data from 130 and 30 patients, respectively. Addition of the ATR-FTIR spectra to the clinical parameters (age, sex, diabetes mellitus, and hypertension) increased the area under the ROC curve (C-statistics) for both the training and test data sets, from 69.3% (95% CI 59.8-78.9%) to 85.7% (78.6-92.8%) and 77.8% (61.3-94.4%) to 85.1% (71.3-98.8%), respectively. The independent test set achieved 69.2% specificity (42.4-87.3%) and 94.1% sensitivity (73.0-99.0%). Diabetes mellitus was the strongest predictor in the model, followed by FTIR regions 1020-1090 and 1588-1592 cm-1. In summary, this study demonstrates the potential of ATR-FTIR spectroscopy as a rapid, low-cost COVID-19 severity triage tool to facilitate COVID-19 patient management during an outbreak.

Predicting critical state after COVID-19 diagnosis: model development using a large US electronic health record dataset

As the COVID-19 pandemic is challenging healthcare systems worldwide, early identification of patients with a high risk of complication is crucial. We present a prognostic model predicting critical state within 28 days following COVID-19 diagnosis trained on data from US electronic health records (IBM Explorys), including demographics, comorbidities, symptoms, and hospitalization. Out of 15753 COVID-19 patients, 2050 went into critical state or deceased. Non-random train-test splits by time were repeated 100 times and led to a ROC AUC of 0.861 [0.838, 0.883] and a precision-recall AUC of 0.434 [0.414, 0.485] (median and interquartile range). The interpretability analysis confirmed evidence on major risk factors (e.g., older age, higher BMI, male gender, diabetes, and cardiovascular disease) in an efficient way compared to clinical studies, demonstrating the model validity. Such personalized predictions could enable fine-graded risk stratification for optimized care management.

Explaining multivariate molecular diagnostic tests via Shapley values

Background

Machine learning (ML) can be an effective tool to extract information from attribute-rich molecular datasets for the generation of molecular diagnostic tests. However, the way in which the resulting scores or classifications are produced from the input data may not be transparent. Algorithmic explainability or interpretability has become a focus of ML research. Shapley values, first introduced in game theory, can provide explanations of the result generated from a specific set of input data by a complex ML algorithm.

Methods

For a multivariate molecular diagnostic test in clinical use (the VeriStrat® test), we calculate and discuss the interpretation of exact Shapley values. We also employ some standard approximation techniques for Shapley value computation (local interpretable model-agnostic explanation (LIME) and Shapley Additive Explanations (SHAP) based methods) and compare the results with exact Shapley values.

Results

Exact Shapley values calculated for data collected from a cohort of 256 patients showed that the relative importance of attributes for test classification varied by sample. While all eight features used in the VeriStrat® test contributed equally to classification for some samples, other samples showed more complex patterns of attribute importance for classification generation. Exact Shapley values and Shapley-based interaction metrics were able to provide interpretable classification explanations at the sample or patient level, while patient subgroups could be defined by comparing Shapley value profiles between patients. LIME and SHAP approximation approaches, even those seeking to include correlations between attributes, produced results that were quantitatively and, in some cases qualitatively, different from the exact Shapley values.

Conclusions

Shapley values can be used to determine the relative importance of input attributes to the result generated by a multivariate molecular diagnostic test for an individual sample or patient. Patient subgroups defined by Shapley value profiles may motivate translational research. However, correlations inherent in molecular data and the typically small ML training sets available for molecular diagnostic test development may cause some approximation methods to produce approximate Shapley values that differ both qualitatively and quantitatively from exact Shapley values. Hence, caution is advised when using approximate methods to evaluate Shapley explanations of the results of molecular diagnostic tests.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-021-01569-9.

A Machine Learning Approach for Mortality Prediction in COVID-19 Pneumonia: Development and Evaluation of the Piacenza Score

BACKGROUND

Several models have been developed to predict mortality in patients with COVID-19 pneumonia, but only a few have demonstrated enough discriminatory capacity. Machine learning algorithms represent a novel approach for the data-driven prediction of clinical outcomes with advantages over statistical modeling.

OBJECTIVE

We aimed to develop a machine learning-based score-the Piacenza score-for 30-day mortality prediction in patients with COVID-19 pneumonia.

METHODS

The study comprised 852 patients with COVID-19 pneumonia, admitted to the Guglielmo da Saliceto Hospital in Italy from February to November 2020. Patients' medical history, demographics, and clinical data were collected using an electronic health record. The overall patient data set was randomly split into derivation and test cohorts. The score was obtained through the naïve Bayes classifier and externally validated on 86 patients admitted to Centro Cardiologico Monzino (Italy) in February 2020. Using a forward-search algorithm, 6 features were identified: age, mean corpuscular hemoglobin concentration, PaO2/FiO2 ratio, temperature, previous stroke, and gender. The Brier index was used to evaluate the ability of the machine learning model to stratify and predict the observed outcomes. A user-friendly website was designed and developed to enable fast and easy use of the tool by physicians. Regarding the customization properties of the Piacenza score, we added a tailored version of the algorithm to the website, which enables an optimized computation of the mortality risk score for a patient when some of the variables used by the Piacenza score are not available. In this case, the naïve Bayes classifier is retrained over the same derivation cohort but using a different set of patient characteristics. We also compared the Piacenza score with the 4C score and with a naïve Bayes algorithm with 14 features chosen a priori.

RESULTS

The Piacenza score exhibited an area under the receiver operating characteristic curve (AUC) of 0.78 (95% CI 0.74-0.84, Brier score=0.19) in the internal validation cohort and 0.79 (95% CI 0.68-0.89, Brier score=0.16) in the external validation cohort, showing a comparable accuracy with respect to the 4C score and to the naïve Bayes model with a priori chosen features; this achieved an AUC of 0.78 (95% CI 0.73-0.83, Brier score=0.26) and 0.80 (95% CI 0.75-0.86, Brier score=0.17), respectively.

CONCLUSIONS

Our findings demonstrated that a customizable machine learning-based score with a purely data-driven selection of features is feasible and effective for the prediction of mortality among patients with COVID-19 pneumonia.

Development and validation of a prediction model for tocilizumab failure in hospitalized patients with SARS-CoV-2 infection

BACKGROUND

The aim of this secondary analysis of the TESEO cohort is to identify, early in the course of treatment with tocilizumab, factors associated with the risk of progressing to mechanical ventilation and death and develop a risk score to estimate the risk of this outcome according to patients' profile.

METHODS

Patients with COVID-19 severe pneumonia receiving standard of care + tocilizumab who were alive and free from mechanical ventilation at day 6 after treatment initiation were included in this retrospective, multicenter cohort study. Multivariable logistic regression models were built to identify predictors of mechanical ventilation or death by day-28 from treatment initiation and β-coefficients were used to develop a risk score. Secondary outcome was mortality. Patients with the same inclusion criteria as the derivation cohort from 3 independent hospitals were used as validation cohort.

RESULTS

266 patients treated with tocilizumab were included. By day 28 of hospital follow-up post treatment initiation, 40 (15%) underwent mechanical ventilation or died [26 (10%)]. At multivariable analysis, sex, day-4 PaO2/FiO2 ratio, platelets and CRP were independently associated with the risk of developing the study outcomes and were used to generate the proposed risk score. The accuracy of the score in AUC was 0.80 and 0.70 in internal validation and test for the composite endpoint and 0.92 and 0.69 for death, respectively.

CONCLUSIONS

Our score could assist clinicians in identifying, early after tocilizumab administration, patients who are likely to progress to mechanical ventilation or death, so that they could be selected for eventual rescue therapies.

A Machine Learning Prediction Model of Respiratory Failure Within 48 Hours of Patient Admission for COVID-19: Model Development and Validation

BACKGROUND

Predicting early respiratory failure due to COVID-19 can help triage patients to higher levels of care, allocate scarce resources, and reduce morbidity and mortality by appropriately monitoring and treating the patients at greatest risk for deterioration. Given the complexity of COVID-19, machine learning approaches may support clinical decision making for patients with this disease.

OBJECTIVE

Our objective is to derive a machine learning model that predicts respiratory failure within 48 hours of admission based on data from the emergency department.

METHODS

Data were collected from patients with COVID-19 who were admitted to Northwell Health acute care hospitals and were discharged, died, or spent a minimum of 48 hours in the hospital between March 1 and May 11, 2020. Of 11,525 patients, 933 (8.1%) were placed on invasive mechanical ventilation within 48 hours of admission. Variables used by the models included clinical and laboratory data commonly collected in the emergency department. We trained and validated three predictive models (two based on XGBoost and one that used logistic regression) using cross-hospital validation. We compared model performance among all three models as well as an established early warning score (Modified Early Warning Score) using receiver operating characteristic curves, precision-recall curves, and other metrics.

RESULTS

The XGBoost model had the highest mean accuracy (0.919; area under the curve=0.77), outperforming the other two models as well as the Modified Early Warning Score. Important predictor variables included the type of oxygen delivery used in the emergency department, patient age, Emergency Severity Index level, respiratory rate, serum lactate, and demographic characteristics.

CONCLUSIONS

The XGBoost model had high predictive accuracy, outperforming other early warning scores. The clinical plausibility and predictive ability of XGBoost suggest that the model could be used to predict 48-hour respiratory failure in admitted patients with COVID-19.

Applications of artificial intelligence in battling against covid-19: A literature review

Colloquially known as coronavirus, the Severe Acute Respiratory Syndrome CoronaVirus 2 (SARS-CoV-2), that causes CoronaVirus Disease 2019 (COVID-19), has become a matter of grave concern for every country around the world. The rapid growth of the pandemic has wreaked havoc and prompted the need for immediate reactions to curb the effects. To manage the problems, many research in a variety of area of science have started studying the issue. Artificial Intelligence is among the area of science that has found great applications in tackling the problem in many aspects. Here, we perform an overview on the applications of AI in a variety of fields including diagnosis of the disease via different types of tests and symptoms, monitoring patients, identifying severity of a patient, processing covid-19 related imaging tests, epidemiology, pharmaceutical studies, etc. The aim of this paper is to perform a comprehensive survey on the applications of AI in battling against the difficulties the outbreak has caused. Thus we cover every way that AI approaches have been employed and to cover all the research until the writing of this paper. We try organize the works in a way that overall picture is comprehensible. Such a picture, although full of details, is very helpful in understand where AI sits in current pandemonium. We also tried to conclude the paper with ideas on how the problems can be tackled in a better way and provide some suggestions for future works.

Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal

OBJECTIVE

To review and appraise the validity and usefulness of published and preprint reports of prediction models for diagnosing coronavirus disease 2019 (covid-19) in patients with suspected infection, for prognosis of patients with covid-19, and for detecting people in the general population at increased risk of covid-19 infection or being admitted to hospital with the disease.

DESIGN

Living systematic review and critical appraisal by the COVID-PRECISE (Precise Risk Estimation to optimise covid-19 Care for Infected or Suspected patients in diverse sEttings) group.

DATA SOURCES

PubMed and Embase through Ovid, up to 1 July 2020, supplemented with arXiv, medRxiv, and bioRxiv up to 5 May 2020.

STUDY SELECTION

Studies that developed or validated a multivariable covid-19 related prediction model.

DATA EXTRACTION

At least two authors independently extracted data using the CHARMS (critical appraisal and data extraction for systematic reviews of prediction modelling studies) checklist; risk of bias was assessed using PROBAST (prediction model risk of bias assessment tool).

RESULTS

37 421 titles were screened, and 169 studies describing 232 prediction models were included. The review identified seven models for identifying people at risk in the general population; 118 diagnostic models for detecting covid-19 (75 were based on medical imaging, 10 to diagnose disease severity); and 107 prognostic models for predicting mortality risk, progression to severe disease, intensive care unit admission, ventilation, intubation, or length of hospital stay. The most frequent types of predictors included in the covid-19 prediction models are vital signs, age, comorbidities, and image features. Flu-like symptoms are frequently predictive in diagnostic models, while sex, C reactive protein, and lymphocyte counts are frequent prognostic factors. Reported C index estimates from the strongest form of validation available per model ranged from 0.71 to 0.99 in prediction models for the general population, from 0.65 to more than 0.99 in diagnostic models, and from 0.54 to 0.99 in prognostic models. All models were rated at high or unclear risk of bias, mostly because of non-representative selection of control patients, exclusion of patients who had not experienced the event of interest by the end of the study, high risk of model overfitting, and unclear reporting. Many models did not include a description of the target population (n=27, 12%) or care setting (n=75, 32%), and only 11 (5%) were externally validated by a calibration plot. The Jehi diagnostic model and the 4C mortality score were identified as promising models.

CONCLUSION

Prediction models for covid-19 are quickly entering the academic literature to support medical decision making at a time when they are urgently needed. This review indicates that almost all pubished prediction models are poorly reported, and at high risk of bias such that their reported predictive performance is probably optimistic. However, we have identified two (one diagnostic and one prognostic) promising models that should soon be validated in multiple cohorts, preferably through collaborative efforts and data sharing to also allow an investigation of the stability and heterogeneity in their performance across populations and settings. Details on all reviewed models are publicly available at https://www.covprecise.org/. Methodological guidance as provided in this paper should be followed because unreliable predictions could cause more harm than benefit in guiding clinical decisions. Finally, prediction model authors should adhere to the TRIPOD (transparent reporting of a multivariable prediction model for individual prognosis or diagnosis) reporting guideline.

SYSTEMATIC REVIEW REGISTRATION

Protocol https://osf.io/ehc47/, registration https://osf.io/wy245.

READERS' NOTE

This article is a living systematic review that will be updated to reflect emerging evidence. Updates may occur for up to two years from the date of original publication. This version is update 3 of the original article published on 7 April 2020 (BMJ 2020;369:m1328). Previous updates can be found as data supplements (https://www.bmj.com/content/369/bmj.m1328/related#datasupp). When citing this paper please consider adding the update number and date of access for clarity.