Machine Learning and Prediction of All-Cause Mortality in COPD

Predicting COPD exacerbations based on quantitative CT analysis: an external validation study

Purpose

Quantitative computed tomography (CT) analysis is an important method for diagnosis and severity evaluation of lung diseases. However, the association between CT-derived biomarkers and chronic obstructive pulmonary disease (COPD) exacerbations remains unclear. We aimed to investigate its potential in predicting COPD exacerbations.

Methods

Patients with COPD were consecutively enrolled, and their data were analyzed in this retrospective study. Body composition and thoracic abnormalities were analyzed from chest CT scans. Logistic regression analysis was performed to identify independent risk factors of exacerbation. Based on 2-year follow-up data, the deep learning system (DLS) was developed to predict future exacerbations. Receiver operating characteristic (ROC) curve analysis was conducted to assess the diagnostic performance. Finally, the survival analysis was performed to further evaluate the potential of the DLS in risk stratification.

Results

A total of 1,150 eligible patients were included and followed up for 2 years. Multivariate analysis revealed that CT-derived high affected lung volume/total lung capacity (ALV/TLC) ratio, high visceral adipose tissue area (VAT), and low pectoralis muscle cross-sectional area (CSA) were independent risk factors causing COPD exacerbations. The DLS outperformed exacerbation history and the BMI, airflow obstruction, dyspnea, and exercise capacity (BODE) index, with an area under the ROC (AUC) value of 0.88 (95%CI, 0.82-0.92) in the internal cohort and 0.86 (95%CI, 0.81-0.89) in the external cohort. The DeLong test revealed significance between this system and conventional scores in the test cohorts (p < 0.05). In the survival analysis, patients with higher risk were susceptible to exacerbation events.

Conclusion

The DLS could allow accurate prediction of COPD exacerbations. The newly identified CT biomarkers (ALV/TLC ratio, VAT, and pectoralis muscle CSA) could potentially enable investigation into underlying mechanisms responsible for exacerbations.

Development and Validation of an In-Hospital Mortality Prediction Model for Patients with Acute Exacerbation of Chronic Obstructive Pulmonary Disease

Purpose

Patients with chronic obstructive pulmonary disease (COPD) often face unknown risks during acute exacerbation of the disease (AECOPD), which could potentially result in mortality. This study aimed to develop and validate a nomogram model for predicting the risk of in-hospital mortality in AECOPD patients.

Patients and Methods

Clinical data of patients hospitalized at The Second People's Hospital of Wuhu City for AECOPD between January 2013 and December 2022 were retrospectively collected. Variables underwent selection through LASSO regression and multivariable logistic regression to develop a nomogram model. The model's predictive performance was assessed using the concordance index, calibration curve, and decision curve analysis (DCA), with internal validation conducted using the bootstrap method.

Results

A total of 1224 patients were included in this study, with 98 (8%) deaths occurring during hospitalization. LASSO regression identified 11 variables, used to construct model A. Further multivariable logistic regression was conducted to select variables with P < 0.05 to establish model B. model B was selected as the final model based on discrimination, calibration, and clinical utility, encompassing variables including acute respiratory failure, lung cancer, heart rate, hemoglobin, absolute neutrophil count, serum albumin, blood urea nitrogen, and serum chloride. The nomogram model achieved a concordance index of 0.858. Internal validation of the model was conducted using the bootstrap method with 500 repetitions, resulting in a concordance index of 0.851 (95% CI: 0.805, 0.893). The calibration curve demonstrated a good fit, with a Hosmer-Lemeshow goodness-of-fit test P-value of 0.520. Moreover, DCA findings suggested patient benefit within a threshold probability range of 0.02 to 0.73, with a maximum net benefit of 0.07.

Conclusion

The model constructed in this study has good predictive performance, which helps clinical doctors identify patients at high risk of death early.

Specific mortality in patients with diffuse large B-cell lymphoma: a retrospective analysis based on the surveillance, epidemiology, and end results database

BACKGROUND

The full potential of competing risk modeling approaches in the context of diffuse large B-cell lymphoma (DLBCL) patients has yet to be fully harnessed. This study aims to address this gap by developing a sophisticated competing risk model specifically designed to predict specific mortality in DLBCL patients.

METHODS

We extracted DLBCL patients' data from the SEER (Surveillance, Epidemiology, and End Results) database. To identify relevant variables, we conducted a two-step screening process using univariate and multivariate Fine and Gray regression analyses. Subsequently, a nomogram was constructed based on the results. The model's consistency index (C-index) was calculated to assess its performance. Additionally, calibration curves and receiver operator characteristic (ROC) curves were generated to validate the model's effectiveness.

RESULTS

This study enrolled a total of 24,402 patients. The feature selection analysis identified 13 variables that were statistically significant and therefore included in the model. The model validation results demonstrated that the area under the receiver operating characteristic (ROC) curve (AUC) for predicting 6-month, 1-year, and 3-year DLBCL-specific mortality was 0.748, 0.718, and 0.698, respectively, in the training cohort. In the validation cohort, the AUC values were 0.747, 0.721, and 0.697. The calibration curves indicated good consistency between the training and validation cohorts.

CONCLUSION

The most significant predictor of DLBCL-specific mortality is the age of the patient, followed by the Ann Arbor stage and the administration of chemotherapy. This predictive model has the potential to facilitate the identification of high-risk DLBCL patients by clinicians, ultimately leading to improved prognosis.

Development and application of a deep learning-based comprehensive early diagnostic model for chronic obstructive pulmonary disease

BACKGROUND

Chronic obstructive pulmonary disease (COPD) is a frequently diagnosed yet treatable condition, provided it is identified early and managed effectively. This study aims to develop an advanced COPD diagnostic model by integrating deep learning and radiomics features.

METHODS

We utilized a dataset comprising CT images from 2,983 participants, of which 2,317 participants also provided epidemiological data through questionnaires. Deep learning features were extracted using a Variational Autoencoder, and radiomics features were obtained using the PyRadiomics package. Multi-Layer Perceptrons were used to construct models based on deep learning and radiomics features independently, as well as a fusion model integrating both. Subsequently, epidemiological questionnaire data were incorporated to establish a more comprehensive model. The diagnostic performance of standalone models, the fusion model and the comprehensive model was evaluated and compared using metrics including accuracy, precision, recall, F1-score, Brier score, receiver operating characteristic curves, and area under the curve (AUC).

RESULTS

The fusion model exhibited outstanding performance with an AUC of 0.952, surpassing the standalone models based solely on deep learning features (AUC = 0.844) or radiomics features (AUC = 0.944). Notably, the comprehensive model, incorporating deep learning features, radiomics features, and questionnaire variables demonstrated the highest diagnostic performance among all models, yielding an AUC of 0.971.

CONCLUSION

We developed and implemented a data fusion strategy to construct a state-of-the-art COPD diagnostic model integrating deep learning features, radiomics features, and questionnaire variables. Our data fusion strategy proved effective, and the model can be easily deployed in clinical settings.

TRIAL REGISTRATION

Not applicable. This study is NOT a clinical trial, it does not report the results of a health care intervention on human participants.

Association Between Serum Anion Gap and Mortality in Critically Ill Patients with COPD in ICU: Data from the MIMIC IV Database

Background

Serum anion gap (AG) has been proven to be associated with prognosis in critically ill patients. However, few studies have investigated the association between AG and all-cause mortality in critically ill patients with chronic obstructive pulmonary disease (COPD).

Objective

We hypothesized that the initial AG level would predict the mortality risk in critically ill patients with COPD.

Methods

This retrospective cohort study was based on the Medical Information Mart for Intensive Care (MIMIC) IV database. We extracted demographics, vital signs, laboratory tests, comorbidity, and scoring systems from the first 24 hours after patient ICU admission. Multivariable logistic regression analysis models were used to explore the association between serum AG levels and mortality. Interaction and stratified analyses were conducted including age, gender and comorbidity.

Results

A total of 5531 critically ill patients with COPD were enrolled, composed of 53.6% male and 46.4% female with a median age of 73 years. The all-cause mortality of these patients during ICU hospitalization was 13.7%. The risk of all-cause mortality increased as the AG level increased in the univariate logistic regression analysis (OR=1.13, 95% CI: 1.11-1.15, p<0.01). After adjusting for all the covariates in multivariate logistic regression analysis, the odds ratio was 1.06 (95% CI: 1.04-1.09, p<0.01). Compared with the lowest AG group Q1 (≤11mmol/L), the adjusted OR value for AG and mortality in Q2 (12-13mmol/L) was 0.89 (95% CI: 0.63-1.25, p=0.502), Q3 (14-15mmol/L) was 0.95 (95% CI: 0.68-1.34, p=0.788), and Q4 (≥16mmol/L) was 1.49 (95% CI: 1.10-2.02, p=0.009) respectively. In addition, the results of the subgroup and stratified analyses were robust.

Conclusion

AG is positively related to all-cause mortality in critically ill patients with COPD.

The prediction of in-hospital mortality in elderly patients with sepsis-associated acute kidney injury utilizing machine learning models

Background

Sepsis-associated acute kidney injury (SA-AKI) is a severe complication associated with poorer prognosis and increased mortality, particularly in elderly patients. Currently, there is a lack of accurate mortality risk prediction models for these patients in clinic.

Objectives

This study aimed to develop and validate machine learning models for predicting in-hospital mortality risk in elderly patients with SA-AKI.

Methods

Machine learning models were developed and validated using the public, high-quality Medical Information Mart for Intensive Care (MIMIC)-IV critically ill database. The recursive feature elimination (RFE) algorithm was employed for key feature selection. Eleven predictive models were compared, with the best one selected for further validation. Shapley Additive Explanations (SHAP) values were used for visualization and interpretation, making the machine learning models clinically interpretable.

Results

There were 16,154 patients with SA-AKI in the MIMIC-IV database, and 8426 SA-AKI patients were included in this study (median age: 77.0 years; female: 45%). 7728 patients excluded based on these criteria. They were randomly divided into a training cohort (5,934, 70%) and a validation cohort (2,492, 30%). Nine key features were selected by the RFE algorithm. The CatBoost model achieved the best performance, with an AUC of 0.844 in the training cohort and 0.804 in the validation cohort. SHAP values revealed that AKI stage, PaO2, and lactate were the top three most important features contributing to the CatBoost model.

Conclusion

We developed a model capable of predicting the risk of in-hospital mortality in elderly patients with SA-AKI.

Development and validation of a nomogram model for mortality prediction in stable chronic obstructive pulmonary disease patients: A prospective observational study in the RealDTC cohort

Background

Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide. There is no nomogram model available for mortality prediction of stable COPD. We intended to develop and validate a nomogram model to predict mortality risk in stable COPD patients for personalised prognostic assessment.

Methods

A prospective observational study was made of COPD outpatients registered in the RealDTC study between December 2016 and December 2019. Patients were randomly assigned to the training cohort and validation cohort in a ratio of 7:3. We used Lasso regression to screen predicted variables. Further, we evaluated the prognostic performance using the area under the time-dependent receiver operating characteristic curve (AUC) and calibration curve. We used the AUC, concordance index, and decision curve analysis to evaluate the net benefits and utility of the nomogram compared with three earlier prediction models.

Results

Of 2499 patients, the median follow-up was 38 months. The characteristics of the patients between the training cohort (n = 1743) and the validation cohort (n = 756) were similar. ABEODS nomogram model, combining age, body mass index, educational level, airflow obstruction, dyspnoea, and severe exacerbation in the first year, was constructed to predict mortality in stable COPD patients. In the integrative analysis of training and validation cohorts of the nomogram model, the three-year mortality prediction achieved AUC = 0.84; 95% confidence interval (CI) = 0.81, 0.88 and AUC = 0.80; 95% CI = 0.74, 0.86, respectively. The ABEODS nomogram model preserved excellent calibration in both the training cohort and validation cohort. The time-dependent AUC, concordance index, and net benefit of the nomogram model were higher than those of BODEx, updated ADO, and DOSE, respectively.

Conclusions

We developed and validated a prognostic nomogram model that accurately predicts mortality across the COPD severity spectrum. The proposed ABEODS nomogram model performed better than earlier models, including BODEx, updated ADO, and DOSE in Chinese patients with COPD.

Registration

ChiCTR-POC-17010431.

Specific patterns and potential risk factors to predict 3-year risk of death among non-cancer patients with advanced chronic kidney disease by machine learning

Chronic kidney disease (CKD) is a major public health concern. But there are limited machine learning studies on non-cancer patients with advanced CKD, and the results of machine learning studies on cancer patients with CKD may not apply directly on non-cancer patients. We aimed to conduct a comprehensive investigation of risk factors for a 3-year risk of death among non-cancer advanced CKD patients with an estimated glomerular filtration rate < 60.0 mL/min/1.73m2 by several machine learning algorithms. In this retrospective cohort study, we collected data from in-hospital and emergency care patients from 2 hospitals in Taiwan from 2009 to 2019, including their international classification of disease at admission and laboratory data from the hospital's electronic medical records (EMRs). Several machine learning algorithms were used to analyze the potential impact and degree of influence of each factor on mortality and survival. Data from 2 hospitals in northern Taiwan were collected with 6565 enrolled patients. After data cleaning, 26 risk factors and approximately 3887 advanced CKD patients from Shuang Ho Hospital were used as the training set. The validation set contained 2299 patients from Taipei Medical University Hospital. Predictive variables, such as albumin, PT-INR, and age, were the top 3 significant risk factors with paramount influence on mortality prediction. In the receiver operating characteristic curve, the random forest had the highest values for accuracy above 0.80. MLP, and Adaboost had better performance on sensitivity and F1-score compared to other methods. Additionally, SVM with linear kernel function had the highest specificity of 0.9983, while its sensitivity and F1-score were poor. Logistic regression had the best performance, with an area under the curve of 0.8527. Evaluating Taiwanese advanced CKD patients' EMRs could provide physicians with a good approximation of the patients' 3-year risk of death by machine learning algorithms.

Machine learning in the prediction of in-hospital mortality in patients with first acute myocardial infarction

BACKGROUND

Persistent efforts are required to further reduce the in-hospital mortality of patients suffering from acute myocardial infarction (AMI), even in the face of a global trend of declining AMI-related fatalities. We investigated deep machine learning models for in-hospital death prediction in patients on their first AMI.

METHOD

In this 2-center retrospective analysis, first AMI patients from Hospital I and Hospital II were included; 4783 patients from Hospital 1 were split in a 7:3 ratio between the training and testing sets. Data from 1053 AMI patients in Hospital II was used for further validation. 70 clinical information and laboratory examination parameters as predictive indicators were included. Logistic Regression Classifier (LR), Random Forests Classifier (RF), eXtreme Gradient Boosting (XGB), Support Vector Machine Classifier (SVM), Multilayer Perceptron (MLP), Gradient Boosting Machine (GBM), Bootstrapped Aggregation (Bagging) models with AMI patients were developed. The importance of selected variables was obtained through the Shapley Additive exPlanations (SHAP) method. The performance of each model was shown using the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (Average Precision; AP).

RESULT

The in-hospital mortality for AMI in the training, testing, and validation sets were 5.7 %, 5.6 %, and 6.0 %, respectively. The top 8 predictors were D-dimer, brain natriuretic peptide, cardiogenic shock, neutrophil, prothrombin time, blood urea nitrogen, cardiac arrest, and phosphorus. In the testing cohort, the models of LR, RF, XGB, SVM, MLP, GBM, and Bagging yielded AUROC values of 0.929, 0.931, 0.907, 0.868, 0.907, 0.923, and 0.932, respectively. Bagging has good predictive value and certain clinical value in external validation with AUROC 0.893.

CONCLUSION

In order to improve the forecasting accuracy of the risk of AMI patients, guide clinical nursing practice, and lower AMI inpatient mortality, this study looked into significant indicators and the optimal models for predicting AMI inpatient mortality.

Identifying metabolic dysfunction-associated steatotic liver disease in patients with hypertension and pre-hypertension: An interpretable machine learning approach

Objective

Metabolic dysfunction-associated steatotic liver disease (MASLD) is one of the most prevalent liver diseases and is associated with pre-hypertension and hypertension. Our research aims to develop interpretable machine learning (ML) models to accurately identify MASLD in hypertensive and pre-hypertensive populations.

Methods

The dataset for 4722 hypertensive and pre-hypertensive patients is from subjects in the NAGALA study. Six ML models, including the decision tree, K-nearest neighbor, gradient boosting, naive Bayes, support vector machine, and random forest (RF) models, were used in this study. The optimal model was constructed according to the performances of models evaluated by K-fold cross-validation (k = 5), the area under the receiver operating characteristic curve (AUC), average precision (AP), accuracy, sensitivity, specificity, and F1. Shapley additive explanation (SHAP) values were employed for both global and local interpretation of the model results.

Results

The prevalence of MASLD in hypertensive and pre-hypertensive patients was 44.3% (362 cases) and 28.3% (1107 cases), respectively. The RF model outperformed the other five models with an AUC of 0.889, AP of 0.800, accuracy of 0.819, sensitivity of 0.816, specificity of 0.821, and F1 of 0.729. According to the SHAP analysis, the top five important features were alanine aminotransferase, body mass index, waist circumference, high-density lipoprotein cholesterol, and total cholesterol. Further analysis of the feature selection in the RF model revealed that incorporating all features leads to optimal model performance.

Conclusions

ML algorithms, especially RF algorithm, improve the accuracy of MASLD identification, and the global and local interpretation of the RF model results enables us to intuitively understand how various features affect the chances of MASLD in patients with hypertension and pre-hypertension.

Chest MRI and CT Predictors of 10-Year All-Cause Mortality in COPD

Pulmonary imaging measurements using magnetic resonance imaging (MRI) and computed tomography (CT) have the potential to deepen our understanding of chronic obstructive pulmonary disease (COPD) by measuring airway and parenchymal pathologic information that cannot be provided by spirometry. Currently, MRI and CT measurements are not included in mortality risk predictions, diagnosis, or COPD staging. We evaluated baseline pulmonary function, MRI and CT measurements alongside imaging texture-features to predict 10-year all-cause mortality in ex-smokers with (n = 93; 31 females; 70 ± 9years) and without (n = 69; 29 females, 69 ± 9years) COPD. CT airway and vessel measurements, helium-3 (3He) MRI ventilation defect percent (VDP) and apparent diffusion coefficients (ADC) were quantified. MRI and CT texture-features were extracted using PyRadiomics (version2.2.0). Associations between 10-year all-cause mortality and all clinical and imaging measurements were evaluated using multivariable regression model odds-ratios. Machine-learning predictive models for 10-year all-cause mortality were evaluated using area-under-receiver-operator-characteristic-curve (AUC), sensitivity and specificity analyses. DLCO (%pred) (HR = 0.955, 95%CI: 0.934-0.976, p < 0.001), MRI ADC (HR = 1.843, 95%CI: 1.260-2.871, p < 0.001), and CT informational-measure-of-correlation (HR = 3.546, 95% CI: 1.660-7.573, p = 0.001) were the strongest predictors of 10-year mortality. A machine-learning model trained on clinical, imaging, and imaging textures was the best predictive model (AUC = 0.82, sensitivity = 83%, specificity = 84%) and outperformed the solely clinical model (AUC = 0.76, sensitivity = 77%, specificity = 79%). In ex-smokers, regardless of COPD status, addition of CT and MR imaging texture measurements to clinical models provided unique prognostic information of mortality risk that can allow for better clinical management.Clinical Trial Registration: www.clinicaltrials.gov NCT02279329.

Machine learning and deep learning predictive models for long-term prognosis in patients with chronic obstructive pulmonary disease: a systematic review and meta-analysis

BACKGROUND

Machine learning and deep learning models have been increasingly used to predict long-term disease progression in patients with chronic obstructive pulmonary disease (COPD). We aimed to summarise the performance of such prognostic models for COPD, compare their relative performances, and identify key research gaps.

METHODS

We conducted a systematic review and meta-analysis to compare the performance of machine learning and deep learning prognostic models and identify pathways for future research. We searched PubMed, Embase, the Cochrane Library, ProQuest, Scopus, and Web of Science from database inception to April 6, 2023, for studies in English using machine learning or deep learning to predict patient outcomes at least 6 months after initial clinical presentation in those with COPD. We included studies comprising human adults aged 18-90 years and allowed for any input modalities. We reported area under the receiver operator characteristic curve (AUC) with 95% CI for predictions of mortality, exacerbation, and decline in forced expiratory volume in 1 s (FEV1). We reported the degree of interstudy heterogeneity using Cochran's Q test (significant heterogeneity was defined as p≤0·10 or I2>50%). Reporting quality was assessed using the TRIPOD checklist and a risk-of-bias assessment was done using the PROBAST checklist. This study was registered with PROSPERO (CRD42022323052).

FINDINGS

We identified 3620 studies in the initial search. 18 studies were eligible, and, of these, 12 used conventional machine learning and six used deep learning models. Seven models analysed exacerbation risk, with only six reporting AUC and 95% CI on internal validation datasets (pooled AUC 0·77 [95% CI 0·69-0·85]) and there was significant heterogeneity (I2 97%, p<0·0001). 11 models analysed mortality risk, with only six reporting AUC and 95% CI on internal validation datasets (pooled AUC 0·77 [95% CI 0·74-0·80]) with significant degrees of heterogeneity (I2 60%, p=0·027). Two studies assessed decline in lung function and were unable to be pooled. Machine learning and deep learning models did not show significant improvement over pre-existing disease severity scores in predicting exacerbations (p=0·24). Three studies directly compared machine learning models against pre-existing severity scores for predicting mortality and pooled performance did not differ (p=0·57). Of the five studies that performed external validation, performance was worse than or equal to regression models. Incorrect handling of missing data, not reporting model uncertainty, and use of datasets that were too small relative to the number of predictive features included provided the largest risks of bias.

INTERPRETATION

There is limited evidence that conventional machine learning and deep learning prognostic models demonstrate superior performance to pre-existing disease severity scores. More rigorous adherence to reporting guidelines would reduce the risk of bias in future studies and aid study reproducibility.

FUNDING

None.

Demystification of artificial intelligence for respiratory clinicians managing patients with obstructive lung diseases

INTRODUCTION

Asthma and chronic obstructive pulmonary disease (COPD) are leading causes of morbidity and mortality worldwide. Despite all available diagnostics and treatments, these conditions pose a significant individual, economic and social burden. Artificial intelligence (AI) promises to support clinical decision-making processes by optimizing diagnosis and treatment strategies of these heterogeneous and complex chronic respiratory diseases. Its capabilities extend to predicting exacerbation risk, disease progression and mortality, providing healthcare professionals with valuable insights for more effective care. Nevertheless, the knowledge gap between respiratory clinicians and data scientists remains a major constraint for wide application of AI and may hinder future progress. This narrative review aims to bridge this gap and encourage AI deployment by explaining its methodology and added value in asthma and COPD diagnosis and treatment.

AREAS COVERED

This review offers an overview of the fundamental concepts of AI and machine learning, outlines the key steps in building a model, provides examples of their applicability in asthma and COPD care, and discusses barriers to their implementation.

EXPERT OPINION

Machine learning can advance our understanding of asthma and COPD, enabling personalized therapy and better outcomes. Further research and validation are needed to ensure the development of clinically meaningful and generalizable models.

Hybridizing machine learning in survival analysis of cardiac PET/CT imaging

BACKGROUND

Machine Learning (ML) allows integration of the numerous variables delivered by cardiac PET/CT, while traditional survival analysis can provide explainable prognostic estimates from a restricted number of input variables. We implemented a hybrid ML-and-survival analysis of multimodal PET/CT data to identify patients who developed myocardial infarction (MI) or death in long-term follow up.

METHODS

Data from 739 intermediate risk patients who underwent coronary CT and selectively stress 15O-water-PET perfusion were analyzed for the occurrence of MI and all-cause mortality. Images were evaluated segmentally for atherosclerosis and absolute myocardial perfusion through 75 variables that were integrated through ML into an ML-CCTA and an ML-PET score. These scores were then modeled along with clinical variables through Cox regression. This hybridized model was compared against an expert interpretation-based and a calcium score-based model.

RESULTS

Compared with expert- and calcium score-based models, the hybridized ML-survival model showed the highest performance (CI .81 vs .71 and .64). The strongest predictor for outcomes was the ML-CCTA score.

CONCLUSION

Prognostic modeling of PET/CT data for the long-term occurrence of adverse events may be improved through ML imaging score integration and subsequent traditional survival analysis with clinical variables. This hybridization of methods offers an alternative to traditional survival modeling of conventional expert image scoring and interpretation.

A Macrophage-Related Gene Signature for Identifying COPD Based on Bioinformatics and ex vivo Experiments

Background

This study aims to investigate the association between immune cells and the development of COPD, while providing a new method for the diagnosis of COPD according to the changes in immune microenvironment.

Methods

In this study, the "CIBERSORT" algorithm was used to estimate the tissue infiltration of 22 types of immune cells in GSE20257 and GSE10006. The "limma" package was used for differentially expressed analysis. The key modules associated with vital immune cells were identified using WGCNA. GO and KEGG enrichment analysis revealed the biological functions of the candidate genes. Ultimately, a novel diagnostic prediction model was constructed via machine learning methods and multivariate logistic regression analysis based on GSE20257. Furthermore, we examined the stability of the model on one internal test set (GSE10006), three external test sets (GSE8545, GSE57148 and GSE76925), one single-cell transcriptome dataset (GSE167295), macrophages (THP-M cells) and lung tissue from COPD patients.

Results

M0 macrophages (AUC > 0.7 in GSE20257 and GSE10006) were considered as the most important immune cells through exploring the immune microenvironment landscapes in COPD patients and healthy controls. The differentially expressed genes from GSE20257 and GSE10006 were divided into six and five modules via WGCNA, respectively. The green module in GSE20257 (cor = 0.41, P < 0.001) and the brown module in GSE10006 (cor = 0.67, P < 0.001) were highly correlated with M0 macrophages and were selected as key modules. Forty-one intersected genes obtained from two modules were primarily involved in regulation of cytokine production, regulation of innate immune response, specific granule, phagosome, lysosome, ferroptosis, and other biological processes. On the basis of the candidate genetic markers further characterized via the "Boruta" and "LASSO" algorithm for COPD, a diagnostic model comprising CLEC5A, FTL and SLC2A3 was constructed, which could accurately distinguish COPD patients from healthy controls in multiple datasets. GSE20257 as the training set has an AUC of 0.916. The AUCs of the internal test set and three external test sets were 0.873, 0.932, 0.675 and 0.688, respectively. Single-cell sequencing analysis suggested that CLEC5A, FTL and SLC2A3 were expressed in macrophages from COPD patients. The expressions of CLEC5A, FTL and SLC2A3 were up-regulated in THP-M cells and lung tissue from COPD patients.

Conclusion

According to the variations of immune microenvironment in COPD patients, we constructed and validated a novel macrophage M0-associated diagnostic model with satisfactory predictive value. CLEC5A, FTL and SLC2A3 are expected to be promising targets of immunotherapy in COPD.

The utility of quantitative computed tomography in cohort studies of chronic obstructive pulmonary disease: a narrative review

Background and Objective

Chronic obstructive pulmonary disease (COPD) is a significant contributor to global morbidity and mortality. Quantitative computed tomography (QCT), a non-invasive imaging modality, offers the potential to assess lung structure and function in COPD patients. Amidst the coronavirus disease 2019 (COVID-19) pandemic, chest computed tomography (CT) scans have emerged as a viable alternative for assessing pulmonary function (e.g., spirometry), minimizing the risk of aerosolized virus transmission. However, the clinical application of QCT measurements is not yet widespread enough, necessitating broader validation to determine its usefulness in COPD management.

Methods

We conducted a search in the PubMed database in English from January 1, 2013 to April 20, 2023, using keywords and controlled vocabulary related to QCT, COPD, and cohort studies.

Key Content and Findings

Existing studies have demonstrated the potential of QCT in providing valuable information on lung volume, airway geometry, airway wall thickness, emphysema, and lung tissue density in COPD patients. Moreover, QCT values have shown robust correlations with pulmonary function tests, and can predict exacerbation risk and mortality in patients with COPD. QCT can even discern COPD subtypes based on phenotypic characteristics such as emphysema predominance, supporting targeted management and interventions.

Conclusions

QCT has shown promise in cohort studies related to COPD, since it can provide critical insights into the pathogenesis and progression of the disease. Further research is necessary to determine the clinical significance of QCT measurements for COPD management.

Development of a Machine Learning Model for Predicting Weaning Outcomes Based Solely on Continuous Ventilator Parameters during Spontaneous Breathing Trials

Discontinuing mechanical ventilation remains challenging. We developed a machine learning model to predict weaning outcomes using only continuous monitoring parameters obtained from ventilators during spontaneous breathing trials (SBTs). Patients who received mechanical ventilation in the medical intensive care unit at a tertiary university hospital from 2019-2021 were included in this study. During the SBTs, three waveforms and 25 numerical data were collected as input variables. The proposed convolutional neural network (CNN)-based weaning prediction model extracts features from input data with diverse lengths. Among 138 enrolled patients, 35 (25.4%) experienced weaning failure. The dataset was randomly divided into training and test sets (8:2 ratio). The area under the receiver operating characteristic curve for weaning success by the prediction model was 0.912 (95% confidence interval [CI], 0.795-1.000), with an area under the precision-recall curve of 0.767 (95% CI, 0.434-0.983). Furthermore, we used gradient-weighted class activation mapping technology to provide visual explanations of the model's prediction, highlighting influential features. This tool can assist medical staff by providing intuitive information regarding readiness for extubation without requiring any additional data collection other than SBT data. The proposed predictive model can assist clinicians in making ventilator weaning decisions in real time, thereby improving patient outcomes.

Effectiveness and safety of opioids on breathlessness and exercise endurance in patients with chronic obstructive pulmonary disease: A systematic review and meta-analysis of randomised controlled trials

BACKGROUND

Opioids are recommended to treat advanced refractory dyspnoea despite optimal therapy by the American Thoracic Society clinical practice guidelines, while newly published randomised controlled trials of opioids in chronic obstructive pulmonary disease yield conflicting results.

AIM

This study aimed to evaluate the effectiveness and safety of opioids for patients with chronic obstructive pulmonary disease.

DESIGN

Systematic review and meta-analysis (PROSPERO CRD42021272556).

DATA SOURCES

Databases of PubMed, EMBASE and CENTRAL were searched from inception to 2022 for eligible randomised controlled trials.

RESULTS

Twenty-four studies including 975 patients, were included. In cross-over studies, opioids improved breathlessness (standardised mean difference, -0.43; 95% CI, -0.55 to -0.30; I2 = 18%) and exercise endurance (standardised mean difference, 0.22; 95% CI, 0.02-0.41; I2 = 70%). However, opioids failed to improve dyspnoea (standardised mean difference, -0.02; 95% CI, -0.22 to 0.19; I2 = 39%) and exercise endurance (standardised mean difference, 0.00; 95% CI, -0.27 to 0.27; I2 = 0%) in parallel control studies that administered sustained-release opioids for more than 1 week. The opioids used in most crossover studies were short-acting and rarely associated with serious adverse effects. Only minor side effects such as dizziness, nausea, constipation and vomiting were identified for short-acting opioids.

CONCLUSIONS

Sustained-release opioids did not improve dyspnoea and exercise endurance. Short-acting opioids appeared to be safe, have potential to lessen dyspnoea and improve exercise endurance, supporting benefit in managing episodes of breathlessness and providing prophylactic treatment for exertional dyspnoea.

Prognostic models for upper urinary tract urothelial carcinoma patients after radical nephroureterectomy based on a novel systemic immune-inflammation score with machine learning

PURPOSE

This study aimed to evaluate the clinical significance of a novel systemic immune-inflammation score (SIIS) to predict oncological outcomes in upper urinary tract urothelial carcinoma(UTUC) after radical nephroureterectomy(RNU).

METHOD

The clinical data of 483 patients with nonmetastatic UTUC underwent surgery in our center were analyzed. Five inflammation-related biomarkers were screened in the Lasso-Cox model and then aggregated to generate the SIIS based on the regression coefficients. Overall survival (OS) was assessed using Kaplan-Meier analyses. The Cox proportional hazards regression and random survival forest model were adopted to build the prognostic model. Then we established an effective nomogram for UTUC after RNU based on SIIS. The discrimination and calibration of the nomogram were evaluated using the concordance index (C-index), area under the time-dependent receiver operating characteristic curve (time-dependent AUC), and calibration curves. Decision curve analysis (DCA) was used to assess the net benefits of the nomogram at different threshold probabilities.

RESULT

According to the median value SIIS computed by the lasso Cox model, the high-risk group had worse OS (p<0.0001) than low risk-group. Variables with a minimum depth greater than the depth threshold or negative variable importance were excluded, and the remaining six variables were included in the model. The area under the ROC curve (AUROC) of the Cox and random survival forest models were 0.801 and 0.872 for OS at five years, respectively. Multivariate Cox analysis showed that elevated SIIS was significantly associated with poorer OS (p<0.001). In terms of predicting overall survival, a nomogram that considered the SIIS and clinical prognostic factors performed better than the AJCC staging.

CONCLUSION

The pretreatment levels of SIIS were an independent predictor of prognosis in upper urinary tract urothelial carcinoma after RNU. Therefore, incorporating SIIS into currently available clinical parameters helps predict the long-term survival of UTUC.

Digital health in chronic obstructive pulmonary disease

Chronic obstructive pulmonary disease (COPD) can be prevented and treated through effective care, reducing exacerbations and hospitalizations. Early identification of individuals at high risk of COPD exacerbation is an opportunity for preventive measures. However, many patients struggle to follow their treatment plans because of a lack of knowledge about the disease, limited access to resources, and insufficient clinical support. The growth of digital health-which encompasses advancements in health information technology, artificial intelligence, telehealth, the Internet of Things, mobile health, wearable technology, and digital therapeutics-offers opportunities for improving the early diagnosis and management of COPD. This study reviewed the field of digital health in terms of COPD. The findings showed that despite significant advances in digital health, there are still obstacles impeding its effectiveness. Finally, we highlighted some of the major challenges and possibilities for developing and integrating digital health in COPD management.

The minimal important difference of the constant work rate cycle test in severe COPD

BACKGROUND

The Constant Work Rate Cycle Test (CWRT) is a commonly used and sensitive test to detect treatment success in patients with Chronic Obstructive Pulmonary Disease (COPD). Earlier, the Minimal Important Difference (MID) of the CWRT was estimated at 101 s (or 34%) change from baseline based on one well executed study. However, this study was performed in a population of patients with mild-to-moderate COPD, and we have learned that MIDs might be quite different in patients with severe COPD. Therefore, we aimed to establish the MID of the CWRT in patients with severe COPD.

METHODS

We included 141 patients with severe COPD, who underwent either pulmonary rehabilitation, bronchoscopic lung volume reduction with endobronchial valves, or a sham bronchoscopy as a control group. CWRT workload was set at 75% of the peak work capacity, as determined by an incremental cycle test. We used the change in 6-min walking test (6-MWT), forced expiratory volume in 1s (FEV₁), residual volume (RV), and St. George's Respiratory Questionnaire (SGRQ) total score as anchors to calculate the MID.

RESULTS

All anchors had an association of ≥0.41 with change in CWRT. The MID estimates for the different anchors were: 6-MWT 278 s (95%), FEV₁ 273 s (90%), RV 240 s (84%), and SGRQ 208 s (71%). The average of these four MID estimates resulted in an MID of 250 s (or 85%).

CONCLUSION

We established the MID for CWRT at 250 s (or 85%) change from baseline in patients with severe COPD.

Impact of the Covid-19 pandemic on the performance of machine learning algorithms for predicting perioperative mortality

BACKGROUND

Machine-learning models are susceptible to external influences which can result in performance deterioration. The aim of our study was to elucidate the impact of a sudden shift in covariates, like the one caused by the Covid-19 pandemic, on model performance.

METHODS

After ethical approval and registration in Clinical Trials (NCT04092933, initial release 17/09/2019), we developed different models for the prediction of perioperative mortality based on preoperative data: one for the pre-pandemic data period until March 2020, one including data before the pandemic and from the first wave until May 2020, and one that covers the complete period before and during the pandemic until October 2021. We applied XGBoost as well as a Deep Learning neural network (DL). Performance metrics of each model during the different pandemic phases were determined, and XGBoost models were analysed for changes in feature importance.

RESULTS

XGBoost and DL provided similar performance on the pre-pandemic data with respect to area under receiver operating characteristic (AUROC, 0.951 vs. 0.942) and area under precision-recall curve (AUPR, 0.144 vs. 0.187). Validation in patient cohorts of the different pandemic waves showed high fluctuations in performance from both AUROC and AUPR for DL, whereas the XGBoost models seemed more stable. Change in variable frequencies with onset of the pandemic were visible in age, ASA score, and the higher proportion of emergency operations, among others. Age consistently showed the highest information gain. Models based on pre-pandemic data performed worse during the first pandemic wave (AUROC 0.914 for XGBoost and DL) whereas models augmented with data from the first wave lacked performance after the first wave (AUROC 0.907 for XGBoost and 0.747 for DL). The deterioration was also visible in AUPR, which worsened by over 50% in both XGBoost and DL in the first phase after re-training.

CONCLUSIONS

A sudden shift in data impacts model performance. Re-training the model with updated data may cause degradation in predictive accuracy if the changes are only transient. Too early re-training should therefore be avoided, and close model surveillance is necessary.

Medical record data-enabled machine learning can enhance prediction of left atrial appendage thrombosis in nonvalvular atrial fibrillation

BACKGROUND

As a major complication of non-valvular atrial fibrillation (NVAF), left atrial appendage (LAA) thrombosis is associated with cerebral ischemic strokes, as well as high morbidity. Due to insufficient incorporation of risk factors, most current scoring methods are limited to the analysis of relationships between clinical characteristics and LAA thrombosis rather than detecting potential risk. Therefore, this study proposes a clinical data-driven machine learning method to predict LAA thrombosis of NVAF.

METHODS

Patients with NVAF from January 2014 to June 2022 were enrolled from Southwest Hospital. We selected 40 variables for analysis, including demographic data, medical history records, laboratory results, and the structure of LAA. Three machine learning algorithms were adopted to construct classifiers for the prediction of LAA thrombosis risk. The most important variables related to LAA thrombosis and their influences were recognized by SHapley Addictive exPlanations method. In addition, we compared our model with CHADS2 and CHADS2-VASc scoring methods.

RESULTS

A total of 713 participants were recruited, including 127 patients with LAA thrombosis and 586 patients with no obvious thrombosis. The consensus models based on Random Forest and eXtreme Gradient Boosting LAA thrombosis prediction (RXTP) achieved the best accuracy of 0.865, significantly outperforming CHADS2 score and CHA2DS2-VASc score (0.757 and 0.754, respectively). The SHAP results showed that B-type natriuretic peptide, left atrial appendage width, C-reactive protein, Fibrinogen and estimated glomerular filtration rate are closely related to the risk of LAA thrombosis in nonvalvular atrial fibrillation.

CONCLUSIONS

The RXTP-NVAF model is the most effective model with the greatest ROC value and recall rate. The summarized risk factors obtained from SHAP enable the optimization of the treatment strategy, thereby preventing thromboembolism events and the occurrence of cardiogenic ischemic stroke.

Diffusing Capacity and Mortality in Chronic Obstructive Pulmonary Disease

Rationale: Chronic obstructive pulmonary disease (COPD) mortality risk is often estimated using the BODE (body mass index, obstruction, dyspnea, exercise capacity) index, including body mass index, forced expiratory volume in 1 second, dyspnea score, and 6-minute walk distance. Diffusing capacity of the lung for carbon monoxide (DlCO) is a potential predictor of mortality that reflects physiology distinct from that in the BODE index. Objectives: This study evaluated DlCO as a predictor of mortality using participants from the COPDGene study. Methods: We performed time-to-event analyses of individuals with COPD (former or current smokers with forced expiratory volume in 1 second/forced vital capacity < 0.7) and DlCO measurements from the COPDGene phase 2 visit. Cox proportional hazard methods were used to model survival, adjusting for age, sex, pack-years, smoking status, BODE index, computed tomography (CT) percent emphysema (low attenuation areas below -950 Hounsfield units), CT airway wall thickness, and history of cardiovascular or kidney diseases. C statistics for models with DlCO and BODE scores were used to compare discriminative accuracy. Results: Of 2,329 participants, 393 (16.8%) died during the follow-up period (median = 4.9 yr). In adjusted analyses, for every 10% decrease in DlCO percent predicted, mortality increased by 28% (hazard ratio = 1.28; 95% confidence interval, 1.17-1.41, P < 0.001). When compared with other clinical predictors, DlCO percent predicted performed similarly to BODE (C statistic DlCO = 0.68; BODE = 0.70), and the addition of DlCO to BODE improved its discriminative accuracy (C statistic = 0.71). Conclusions: Diffusing capacity, a measure of gas transfer, strongly predicted all-cause mortality in individuals with COPD, independent of BODE index and CT evidence of emphysema and airway wall thickness. These findings support inclusion of DlCO in prognostic models for COPD.

Relationship between Antithrombin III Activity and Mortality in Patients with Acute Exacerbation of Chronic Obstructive Pulmonary Disease

We aimed to explore the role of antithrombin III (AT-III) activity in diagnosing patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD) and chronic bronchitis, and its relationship with all-cause mortality of AECOPD patients. We performed univariate and multivariate Cox regression analyses of the factors determining all-cause mortality. We recruited 279 patients with AECOPD and 91 with chronic bronchitis. On admission, patients with AECOPD had lower AT-III activity (80.7 vs. 86.35%, p = 0.002) and higher neutrophil percentages (70.12 vs. 66.40%, p = 0.02) than those with chronic bronchitis. The patients who died were older (78 vs. 73 years, p < 0.001); had higher CRP (39.05 vs. 5.65 mg/L, p < 0.001), D-dimer (1.72 vs. 0.46 mg/L, p < 0.001), FIB (3.56 vs. 3.05 g/L, p = 0.01) levels; and exhibited lower AT-III activity (71.29 vs. 82.94%, p < 0.001) than the survivors. The AT-III area under the receiver operating characteristic curve for predicting COPD all-cause mortality was 0.75 (p < 0.001), optimal cutoff point 79.75%, sensitivity 86.8%, and specificity 57.1%. Multivariate Cox regression analyses showed that increased levels of CRP (HR = 1.005, p = 0.02), D-dimer (HR = 1.17, p = 0.01), WBC count (HR = 1.11, p = 0.002), and reduced AT-III activity (HR = 0.97, p = 0.02) were independent prognostic factors for all-cause mortality. Patients with AT-III ≤ 79.75% were 4.52 times (p = 0.001) more likely to die than those with AT-III > 79.75%. AT-III activity was lower in patients with AECOPD than in those with chronic bronchitis and is potentially useful as an independent predictor of all-cause mortality in patients with AECOPD: reduced AT-III activity and increased CRP and D-dimer levels indicate a higher risk of all-cause mortality.

Blood urea nitrogen to serum albumin ratio: a good predictor of in-hospital and 90-day all-cause mortality in patients with acute exacerbations of chronic obstructive pulmonary disease

BACKGROUND

Previous studies on acute exacerbation of chronic obstructive pulmonary disease (AECOPD) have found that those who died in hospital had higher blood urea nitrogen levels and a worse nutritional status compared to survivors. However, the association between the blood urea nitrogen to serum albumin ratio (BUN/ALB ratio) and in-hospital and short-term prognosis in patients with AECOPD remains unclear. The aim of this study was to explore the usefulness of BUN/ALB ratio in AECOPD as an objective predictor for in-hospital and 90-day all-cause mortality.

METHODS

We recorded the laboratory and clinical data in patients with AECOPD on admission. By drawing the ROC curve for the patients, we obtained the cut-off point for the BUN/ALB ratio for in-hospital death. Multivariate logistic regression was used for analyses of the factors of in-hospital mortality and multivariate Cox regression was used to analyze the factors of 90-day all-cause mortality.

RESULTS

A total of 362 patients were recruited and 319 patients were finally analyzed. Twenty-three patients died during hospitalization and the fatality rate was 7.2%. Furthermore, 14 patients died by the 90-day follow-up. Compared with in-hospital survivors, patients who died in hospital were older (80.78 ± 6.58 vs. 75.09 ± 9.73 years old, P = 0.001), had a higher prevalence of congestive heart failure(69.6% vs. 27.4%, P < 0.001), had a higher BUN/ALB ratio [0.329 (0.250-0.399) vs. 0.145 (0.111-0.210), P < 0.001], had higher neutrophil counts [10.27 (7.21-14.04) vs. 6.58 (4.58-9.04), P < 0.001], higher blood urea nitrogen levels [10.86 (7.10-12.25) vs. 5.35 (4.14-7.40), P < 0.001], a lower albumin level (32.58 ± 3.72 vs. 36.26 ± 4.53, P < 0.001) and a lower lymphocyte count [0.85 (0.58-1.21) vs. 1.22 (0.86-1.72), P = 0.001]. The ROC curve showed that the area under the curve (AUC) of BUN/ALB ratio for in-hospital death was 0.87, (95%CI 0.81-0.93, P < 0.001), the best cut-off point value to discriminate survivors from non-survivors in hospital was 0.249, the sensitivity was 78.3%, the specificity was 86.5%, and Youden's index was 0.648. Having a BUN/ALB ratio ≥ 0.249 was an independent risk factor for both in-hospital and 90-day all-cause mortality after adjustment for relative risk (RR; RR = 15.08, 95% CI 3.80-59.78, P < 0.001 for a multivariate logistic regression analysis) and hazard ratio (HR; HR = 5.34, 95% CI 1.62-17.57, P = 0.006 for a multivariate Cox regression analysis).

CONCLUSION

An elevated BUN/ALB ratio was a strong and independent predictor of in-hospital and 90-day all-cause mortality in patients with AECOPD.

Machine Learning Model Development and Validation for Predicting Outcome in Stage 4 Solid Cancer Patients with Septic Shock Visiting the Emergency Department: A Multi-Center, Prospective Cohort Study

A reliable prognostic score for minimizing futile treatments in advanced cancer patients with septic shock is rare. A machine learning (ML) model to classify the risk of advanced cancer patients with septic shock is proposed and compared with the existing scoring systems. A multi-center, retrospective, observational study of the septic shock registry in patients with stage 4 cancer was divided into a training set and a test set in a 7:3 ratio. The primary outcome was 28-day mortality. The best ML model was determined using a stratified 10-fold cross-validation in the training set. A total of 897 patients were included, and the 28-day mortality was 26.4%. The best ML model in the training set was balanced random forest (BRF), with an area under the curve (AUC) of 0.821 to predict 28-day mortality. The AUC of the BRF to predict the 28-day mortality in the test set was 0.859. The AUC of the BRF was significantly higher than those of the Sequential Organ Failure Assessment score and the Acute Physiology and Chronic Health Evaluation II score (both p < 0.001). The ML model outperformed the existing scores for predicting 28-day mortality in stage 4 cancer patients with septic shock. However, further studies are needed to improve the prediction algorithm and to validate it in various countries. This model might support clinicians in real-time to adopt appropriate levels of care.

Unsupervised Learning Identifies Computed Tomographic Measurements as Primary Drivers of Progression, Exacerbation, and Mortality in Chronic Obstructive Pulmonary Disease

Rationale: Chronic obstructive pulmonary disease (COPD) is a heterogeneous syndrome with phenotypic manifestations that tend to be distributed along a continuum. Unsupervised machine learning based on broad selection of imaging and clinical phenotypes may be used to identify primary variables that define disease axes and stratify patients with COPD. Objectives: To identify primary variables driving COPD heterogeneity using principal component analysis and to define disease axes and assess the prognostic value of these axes across three outcomes: progression, exacerbation, and mortality. Methods: We included 7,331 patients between 39 and 85 years old, of whom 40.3% were Black and 45.8% were female smokers with a mean of 44.6 pack-years, from the COPDGene (Genetic Epidemiology of COPD) phase I cohort (2008-2011) in our analysis. Out of a total of 916 phenotypes, 147 continuous clinical, spirometric, and computed tomography (CT) features were selected. For each principal component (PC), we computed a PC score based on feature weights. We used PC score distributions to define disease axes along which we divided the patients into quartiles. To assess the prognostic value of these axes, we applied logistic regression analyses to estimate 5-year (n = 4,159) and 10-year (n = 1,487) odds of progression. Cox regression and Kaplan-Meier analyses were performed to estimate 5-year and 10-year risk of exacerbation (n = 6,532) and all-cause mortality (n = 7,331). Results: The first PC, accounting for 43.7% of variance, was defined by CT measures of air trapping and emphysema. The second PC, accounting for 13.7% of variance, was defined by spirometric and CT measures of vital capacity and lung volume. The third PC, accounting for 7.9% of the variance, was defined by CT measures of lung mass, airway thickening, and body habitus. Stratification of patients across each disease axis revealed up to 3.2-fold (95% confidence interval [CI] 2.4, 4.3) greater odds of 5-year progression, 5.4-fold (95% CI 4.6, 6.3) greater risk of 5-year exacerbation, and 5.0-fold (95% CI 4.2, 6.0) greater risk of 10-year mortality between the highest and lowest quartiles. Conclusions: Unsupervised learning analysis of the COPDGene cohort reveals that CT measurements may bolster patient stratification along the continuum of COPD phenotypes. Each of the disease axes also individually demonstrate prognostic potential, predictive of future forced expiratory volume in 1 second decline, exacerbation, and mortality.

Predicting acute kidney injury risk in acute myocardial infarction patients: An artificial intelligence model using medical information mart for intensive care databases

Background

Predictive models based on machine learning have been widely used in clinical practice. Patients with acute myocardial infarction (AMI) are prone to the risk of acute kidney injury (AKI), which results in a poor prognosis for the patient. The aim of this study was to develop a machine learning predictive model for the identification of AKI in AMI patients.

Methods

Patients with AMI who had been registered in the Medical Information Mart for Intensive Care (MIMIC) III and IV database were enrolled. The primary outcome was the occurrence of AKI during hospitalization. We developed Random Forests (RF) model, Naive Bayes (NB) model, Support Vector Machine (SVM) model, eXtreme Gradient Boosting (xGBoost) model, Decision Trees (DT) model, and Logistic Regression (LR) models with AMI patients in MIMIC-IV database. The importance ranking of all variables was obtained by the SHapley Additive exPlanations (SHAP) method. AMI patients in MIMIC-III databases were used for model evaluation. The area under the receiver operating characteristic curve (AUC) was used to compare the performance of each model.

Results

A total of 3,882 subjects with AMI were enrolled through screening of the MIMIC database, of which 1,098 patients (28.2%) developed AKI. We randomly assigned 70% of the patients in the MIMIC-IV data to the training cohort, which is used to develop models in the training cohort. The remaining 30% is allocated to the testing cohort. Meanwhile, MIMIC-III patient data performs the external validation function of the model. 3,882 patients and 37 predictors were included in the analysis for model construction. The top 5 predictors were serum creatinine, activated partial prothrombin time, blood glucose concentration, platelets, and atrial fibrillation, (SHAP values are 0.670, 0.444, 0.398, 0.389, and 0.381, respectively). In the testing cohort, using top 20 important features, the models of RF, NB, SVM, xGBoost, DT model, and LR obtained AUC of 0.733, 0.739, 0.687, 0.689, 0.663, and 0.677, respectively. Placing RF models of number of different variables on the external validation cohort yielded their AUC of 0.711, 0.754, 0.778, 0.781, and 0.777, respectively.

Conclusion

Machine learning algorithms, particularly the random forest algorithm, have improved the accuracy of risk stratification for AKI in AMI patients and are applied to accurately identify the risk of AKI in AMI patients.

Clustering-based COPD subtypes have distinct longitudinal outcomes and multi-omics biomarkers

Introduction

Chronic obstructive pulmonary disease (COPD) can progress across several domains, complicating the identification of the determinants of disease progression. In our previous work, we applied k-means clustering to spirometric and chest radiological measures to identify four COPD-related subtypes: ‘relatively resistant smokers (RRS)’, ‘mild upper lobe-predominant emphysema (ULE)’, ‘airway-predominant disease (AD)’ and ‘severe emphysema (SE)’. In the current study, we examined the associations of these subtypes to longitudinal COPD-related health measures as well as blood transcriptomic and plasma proteomic biomarkers.

Methods

We included 8266 non-Hispanic white and African-American smokers from the COPDGene study. We used linear regression to investigate cluster associations to 5-year prospective changes in spirometric and radiological measures and to gene expression and protein levels. We used Cox-proportional hazard test to test for cluster associations to prospective exacerbations, comorbidities and mortality.

Results

The RRS, ULE, AD and SE clusters represented 39%, 15%, 26% and 20% of the studied cohort at baseline, respectively. The SE cluster had the greatest 5-year FEV₁ (forced expiratory volume in 1 s) and emphysema progression, and the highest risks of exacerbations, cardiovascular disease and mortality. The AD cluster had the highest diabetes risk. After adjustments, only the SE cluster had an elevated respiratory mortality risk, while the ULE, AD and SE clusters had elevated all-cause mortality risks. These clusters also demonstrated differential protein and gene expression biomarker associations, mostly related to inflammatory and immune processes.

Conclusion

COPD k-means subtypes demonstrate varying rates of disease progression, prospective comorbidities, mortality and associations to transcriptomic and proteomic biomarkers. These findings emphasise the clinical and biological relevance of these subtypes, which call for more study for translation into clinical practice.

Trail registration number

NCT00608764.

Machine Learning Methods for Hypercholesterolemia Long-Term Risk Prediction

Cholesterol is a waxy substance found in blood lipids. Its role in the human body is helpful in the process of producing new cells as long as it is at a healthy level. When cholesterol exceeds the permissible limits, it works the opposite, causing serious heart health problems. When a person has high cholesterol (hypercholesterolemia), the blood vessels are blocked by fats, and thus, circulation through the arteries becomes difficult. The heart does not receive the oxygen it needs, and the risk of heart attack increases. Nowadays, machine learning (ML) has gained special interest from physicians, medical centers and healthcare providers due to its key capabilities in health-related issues, such as risk prediction, prognosis, treatment and management of various conditions. In this article, a supervised ML methodology is outlined whose main objective is to create risk prediction tools with high efficiency for hypercholesterolemia occurrence. Specifically, a data understanding analysis is conducted to explore the features association and importance to hypercholesterolemia. These factors are utilized to train and test several ML models to find the most efficient for our purpose. For the evaluation of the ML models, precision, recall, accuracy, F-measure, and AUC metrics have been taken into consideration. The derived results highlighted Soft Voting with Rotation and Random Forest trees as base models, which achieved better performance in comparison to the other models with an AUC of 94.5%, precision of 92%, recall of 91.8%, F-measure of 91.7% and an accuracy equal to 91.75%.

Data-Driven Machine-Learning Methods for Diabetes Risk Prediction

Diabetes mellitus is a chronic condition characterized by a disturbance in the metabolism of carbohydrates, fats and proteins. The most characteristic disorder in all forms of diabetes is hyperglycemia, i.e., elevated blood sugar levels. The modern way of life has significantly increased the incidence of diabetes. Therefore, early diagnosis of the disease is a necessity. Machine Learning (ML) has gained great popularity among healthcare providers and physicians due to its high potential in developing efficient tools for risk prediction, prognosis, treatment and the management of various conditions. In this study, a supervised learning methodology is described that aims to create risk prediction tools with high efficiency for type 2 diabetes occurrence. A features analysis is conducted to evaluate their importance and explore their association with diabetes. These features are the most common symptoms that often develop slowly with diabetes, and they are utilized to train and test several ML models. Various ML models are evaluated in terms of the Precision, Recall, F-Measure, Accuracy and AUC metrics and compared under 10-fold cross-validation and data splitting. Both validation methods highlighted Random Forest and K-NN as the best performing models in comparison to the other models.

Long-term outcome of chronic obstructive pulmonary disease: A review

Chronic obstructive pulmonary disease (COPD) is a chronic airway inflammation characterized by fixed airflow limitation and chronic respiratory symptoms, such as cough, sputum, and dyspnea. COPD is a progressive disease characterized by a decline in lung function. During the natural course of the disease, acute deterioration of symptoms leading to hospital visits can occur and influence further disease progression and subsequent exacerbation. Moreover, COPD is not only restricted to pulmonary manifestations but can present with other systemic diseases as comorbidities or systemic manifestations, including lung cancer, cardiovascular disease, pulmonary hypertension, sarcopenia, and metabolic abnormalities. These pulmonary and extrapulmonary conditions lead to the aggravation of dyspnea, physical inactivity, decreased exercise capacity, functional decline, reduced quality of life, and increased mortality. In addition, pneumonia, which is attributed to both COPD itself and an adverse effect of treatment (especially the use of inhaled and/or systemic steroids), can occur and lead to further deterioration in the prognosis of COPD. This review summarizes the long-term outcomes of patients with COPD. In addition, recent studies on the prediction of adverse outcomes are summarized in the last part of the review.

Development and validation of a deep learning model to predict the survival of patients in ICU

BACKGROUND

Patients in the intensive care unit (ICU) are often in critical condition and have a high mortality rate. Accurately predicting the survival probability of ICU patients is beneficial to timely care and prioritizing medical resources to improve the overall patient population survival. Models developed by deep learning (DL) algorithms show good performance on many models. However, few DL algorithms have been validated in the dimension of survival time or compared with traditional algorithms.

METHODS

Variables from the Early Warning Score, Sequential Organ Failure Assessment Score, Simplified Acute Physiology Score II, Acute Physiology and Chronic Health Evaluation (APACHE) II, and APACHE IV models were selected for model development. The Cox regression, random survival forest (RSF), and DL methods were used to develop prediction models for the survival probability of ICU patients. The prediction performance was independently evaluated in the MIMIC-III Clinical Database (MIMIC-III), the eICU Collaborative Research Database (eICU), and Shanghai Pulmonary Hospital Database (SPH).

RESULTS

Forty variables were collected in total for model development. 83 943 participants from 3 databases were included in the study. The New-DL model accurately stratified patients into different survival probability groups with a C-index of >0.7 in the MIMIC-III, eICU, and SPH, performing better than the other models. The calibration curves of the models at 3 and 10 days indicated that the prediction performance was good. A user-friendly interface was developed to enable the model's convenience.

CONCLUSIONS

Compared with traditional algorithms, DL algorithms are more accurate in predicting the survival probability during ICU hospitalization. This novel model can provide reliable, individualized survival probability prediction.

Stroke Risk Prediction with Machine Learning Techniques

A stroke is caused when blood flow to a part of the brain is stopped abruptly. Without the blood supply, the brain cells gradually die, and disability occurs depending on the area of the brain affected. Early recognition of symptoms can significantly carry valuable information for the prediction of stroke and promoting a healthy life. In this research work, with the aid of machine learning (ML), several models are developed and evaluated to design a robust framework for the long-term risk prediction of stroke occurrence. The main contribution of this study is a stacking method that achieves a high performance that is validated by various metrics, such as AUC, precision, recall, F-measure and accuracy. The experiment results showed that the stacking classification outperforms the other methods, with an AUC of 98.9%, F-measure, precision and recall of 97.4% and an accuracy of 98%.

Exploring the most important factors related to self-perceived health among older men in Sweden: a cross-sectional study using machine learning

OBJECTIVE

To evaluate which factors are the most strongly related to self-perceived health among older men and describe the shape of the association between the related factors and self-perceived health using machine learning.

DESIGN AND SETTING

This is a cross-sectional study within the population-based VAScular and Chronic Obstructive Lung disease study (VASCOL) conducted in southern Sweden in 2019.

PARTICIPANTS

A total of 475 older men aged 73 years from the VASCOL dataset.

MEASURES

Self-perceived health was measured using the first item of the Short Form 12. An extreme gradient-boosting model was trained to classify self-perceived health as better (rated: excellent or very good) or worse (rated: fair or poor) using self-reported data on 19 prevalent physician-diagnosed health conditions, intensity of 9 symptoms and 9 demographic and lifestyle factors. Importance of factors was measured in SHapley Additive exPlanations absolute mean and higher scores correspond to greater importance.

RESULTS

The most important factors for classifying self-perceived health were: pain (0.629), sleep quality (0.595), breathlessness (0.549), fatigue (0.542) and depression (0.526). Health conditions ranked well below symptoms and lifestyle variables. Low levels of symptoms, good sleep quality, regular exercise, alcohol consumption and a body mass index between 22 and 28 were associated with better self-perceived health.

CONCLUSIONS

Symptoms are more strongly related to self-perceived health than health conditions, which suggests that the impacts of health conditions are mediated through symptoms, which could be important targets to improve self-perceived health. Machine learning offers a new way to assess composite constructs such as well-being or quality of life.

Machine learning-based modeling of acute respiratory failure following emergency general surgery operations

Background

Emergency general surgery (EGS) operations are associated with substantial risk of morbidity including postoperative respiratory failure (PRF). While existing risk models are not widely utilized and rely on traditional statistical methods, application of machine learning (ML) in prediction of PRF following EGS remains unexplored.

Objective

The present study aimed to develop ML-based prediction models for respiratory failure following EGS and compare their performance to traditional regression models using a nationally-representative cohort.

Methods

Non-elective hospitalizations for EGS (appendectomy, cholecystectomy, repair of perforated ulcer, large or small bowel resection, lysis of adhesions) were identified in the 2016–18 Nationwide Readmissions Database. Factors associated with PRF were identified using ML techniques and logistic regression. The performance of XGBoost and logistic regression was evaluated using the receiver operating characteristic curve and coefficient of determination (R²). The impact of PRF on mortality, length of stay (LOS) and hospitalization costs was secondarily assessed using generalized linear models.

Results

Of 1,003,703 hospitalizations, 8.8% developed PRF. The XGBoost model exhibited slightly superior discrimination compared to logistic regression (0.900, 95% CI 0.899–0.901 vs 0.894, 95% CI 0.862–0.896). Compared to logistic regression, XGBoost demonstrated excellent calibration across all risk levels (R²: 0.998 vs 0.962). Congestive heart failure, neurologic disorders, and coagulopathy were significantly associated with increased risk of PRF. After risk-adjustment, PRF was associated with 10-fold greater odds (95% confidence interval (CI) 9.8–11.1) of mortality and incremental increases in LOS by 3.1 days (95% CI 3.0–3.2) and $11,900 (95% CI 11,600–12,300) in costs.

Conclusions

Logistic regression and XGBoost perform similarly in overall classification of PRF risk. However, due to superior calibration at extremes of risk, ML-based models may prove more useful in the clinical setting, where probabilities rather than classifications are desired.

Utility of Noncancerous Chest CT Features for Predicting Overall Survival and Noncancer Death in Patients With Stage I Lung Cancer Treated With Stereotactic Body Radiotherapy

Background: Noncancerous imaging markers can be readily derived from pretreatment diagnostic and radiotherapy planning chest CT examinations. Objective: To explore the ability of noncancerous features on chest CT to predict overall survival (OS) and noncancer-related death in patients with stage I lung cancer treated with stereotactic body radiation therapy (SBRT). Methods: This retrospective study included 282 patients (168 female, 114 male; median age, 75 years) with stage I lung cancer treated with SBRT between January 2009 and June 2017. Pretreatment chest CT was used to quantify coronary artery calcium (CAC) score, pulmonary artery (PA)-to-aorta ratio, emphysema, and body composition in terms of the cross-sectional area and attenuation of skeletal muscle and subcutaneous adipose tissue at the T5, T8, and T10 vertebral levels. Associations of clinical and imaging features with OS were quantified using a multivariable Cox proportional hazards (PH) model. Penalized multivariable Cox PH models to predict OS were constructed using clinical features only and using both clinical and imaging features. Models' discriminatory ability was assessed by constructing time-varying ROC curves and computing AUC at prespecified times. Results: After a median OS of 60.8 months (95% CI 55.8-68.9), 148 (52.5%) patients died, including 83 (56.1%) with noncancer deaths. Higher CAC score (11-399: hazard ratio [HR] 1.83 [95% CI 1.15-2.91], P=.01; ≥400: HR 1.63 [95% CI 1.01-2.63], P=.04), higher PA-to-aorta ratio (HR 1.33 [95% CI 1.16-1.52], P<.001, per 0.1-unit increase), and lower thoracic skeletal muscle index (HR 0.88 [95% CI 0.79-0.98], P=.02, per 10 cm²/m² increase) were independently associated with shorter OS. Discriminatory ability for 5-year OS was greater for the model including clinical and imaging features than for the model including clinical features only (AUC, 0.75 [95% CI 0.68-0.83] versus 0.61 [95% CI 0.53-0.70], p < .01). The model's most important clinical or imaging feature based on mean standardized regression coefficients was the PA-to-aorta ratio. Conclusions: In patients undergoing SBRT for stage I lung cancer, higher CAC score, higher PA-to-aorta ratio, and lower thoracic skeletal muscle index independently predicted worse OS. Clinical Impact: Noncancerous imaging features on chest CT performed before SBRT improve survival prediction compared with clinical features alone.

A Promising Preoperative Prediction Model for Microvascular Invasion in Hepatocellular Carcinoma Based on an Extreme Gradient Boosting Algorithm

Background

The non-invasive preoperative diagnosis of microvascular invasion (MVI) in hepatocellular carcinoma (HCC) is vital for precise surgical decision-making and patient prognosis. Herein, we aimed to develop an MVI prediction model with valid performance and clinical interpretability.

Methods

A total of 2160 patients with HCC without macroscopic invasion who underwent hepatectomy for the first time in West China Hospital from January 2015 to June 2019 were retrospectively included, and randomly divided into training and a validation cohort at a ratio of 8:2. Preoperative demographic features, imaging characteristics, and laboratory indexes of the patients were collected. Five machine learning algorithms were used: logistic regression, random forest, support vector machine, extreme gradient boosting (XGBoost), and multilayer perception. Performance was evaluated using the area under the receiver operating characteristic curve (AUC). We also determined the Shapley Additive exPlanation value to explain the influence of each feature on the MVI prediction model.

Results

The top six important preoperative factors associated with MVI were the maximum image diameter, protein induced by vitamin K absence or antagonist-II, α-fetoprotein level, satellite nodules, alanine aminotransferase (AST)/aspartate aminotransferase (ALT) ratio, and AST level, according to the XGBoost model. The XGBoost model for preoperative prediction of MVI exhibited a better AUC (0.8, 95% confidence interval: 0.74–0.83) than the other prediction models. Furthermore, to facilitate use of the model in clinical settings, we developed a user-friendly online calculator for MVI risk prediction based on the XGBoost model.

Conclusions

The XGBoost model achieved outstanding performance for non-invasive preoperative prediction of MVI based on big data. Moreover, the MVI risk calculator would assist clinicians in conveniently determining the optimal therapeutic remedy and ameliorating the prognosis of patients with HCC.

Chronic obstructive pulmonary disease risk assessment tools: is one better than the others?

PURPOSE OF REVIEW

Risk assessment tools are essential in COPD care to help clinicians identify patients at higher risk of accelerated lung function decline, respiratory exacerbations, hospitalizations, and death.

RECENT FINDINGS

Conventional methods of assessing risk have focused on spirometry, patient-reported symptoms, functional status, and a combination of these tools in composite indices. More recently, qualitatively and quantitatively assessed chest imaging findings, such as emphysema, large and small airways disease, and pulmonary vascular abnormalities have been associated with poor long-term outcomes in COPD patients. Although several blood and sputum biomarkers have been investigated for risk assessment in COPD, most still warrant further validation. Finally, novel remote digital monitoring technologies may be valuable to predict exacerbations but their large-scale performance, ease of implementation, and cost effectiveness remain to be determined.

SUMMARY

Given the complex heterogeneity of COPD, any single metric is unlikely to fully capture the risk of poor long-term outcomes. Therefore, clinicians should review all available clinical data, including spirometry, symptom severity, functional status, chest imaging, and bloodwork, to guide personalized preventive care of COPD patients. The potential of machine learning tools and remote monitoring technologies to refine COPD risk assessment is promising but remains largely untapped pending further investigation.

Applications of artificial intelligence in common pulmonary diseases

Artificial intelligence (AI) is a branch of computer science where machines are trained to imitate human-level intelligence and perform well-defined tasks. AI can provide accurate results as well as analyze vast amounts of data that cannot be analyzed via conventional statistical methods. AI has been utilized in pulmonary medicine for almost two decades and its utilization continues to expand. AI can help in making diagnoses and predicting outcomes in pulmonary diseases based on clinical data, chest imaging, lung pathology, and pulmonary function testing. AI-based applications enable physicians to use enormous amounts of data and improve their precision in the treatment of pulmonary diseases. Given the growing role of AI in pulmonary medicine, it is important for practitioners caring for patients with pulmonary diseases to understand how AI can work in order to implement it into clinical practices and improve patient care. The goal of this mini-review is to discuss the use of AI in pulmonary medicine and imaging in cases of obstructive lung disease, interstitial lung disease, infections, nodules, and lung cancer.

Rationale and design of the Early Chronic Obstructive Pulmonary Disease (ECOPD) study in Guangdong, China: a prospective observational cohort study

Background

Chronic obstructive pulmonary disease (COPD) is a heterogeneous disease and its clinically relevant subtypes are not well understood. Which clinical characteristics are more likely to be present among individuals who develop COPD remains to be studied in depth. Therefore, we designed a prospective observational cohort study, entitled the Early Chronic Obstructive Pulmonary Disease (ECOPD) study, to fill this evidence gap. The ECOPD study has four specific aims: (I) identification of characteristics, parameters, and biomarkers that may predict the development of airflow obstruction and annual decline in lung function with normal spirometry; (II) identification of clinically relevant early COPD subtypes; (III) identification of characteristics, parameters, and biomarkers that may predict disease progression in these early COPD subtypes; (IV) development and validation of machine learning models to predict development of airflow obstruction and disease progression.

Methods

We will recruit approximately 2,000 participants aged 40–80 years, including approximately 1,000 with COPD [post-bronchodilator forced expiratory volume in 1 second (FEV1)/forced vital capacity (FVC) <0.7] and approximately 1,000 without COPD, using a population-based survey for COPD. We will assess all participants using standard respiratory epidemiological questionnaires, pulmonary function tests [pre-bronchodilator and post-bronchodilator spirometry, and impulse oscillometry (IOS)], health outcomes [modified British Medical Research Council (mMRC) dyspnea scale, COPD assessment test (CAT), COPD clinical questionnaire (CCQ)], inspiratory and expiratory chest computed tomography (CT), and biomarker measurements (blood and urine), as well as satellite remote sensing pollutant exposure measures. Subgroup will additionally complete induced sputum, exercise capacity tests [6-minute walk test (6MWT) and cardiopulmonary exercise testing (CPET)] and home monitoring/personal sampling as pollutant exposure measures. Study procedures will be performed at baseline and every 1 year thereafter.

Discussion

The ECOPD study will provide insight into many aspects of early COPD and improve our understanding of COPD development, which may facilitate therapeutic interventions with the potential to modify the course of disease.

Trial Registration

Chinese Clinical Trial Registry, ChiCTR1900024643. Registered on 19 July, 2019.

Development of a Blood-based Transcriptional Risk Score for Chronic Obstructive Pulmonary Disease

Rationale: The ability of peripheral blood biomarkers to assess chronic obstructive pulmonary disease (COPD) risk and progression is unknown. Genetics and gene expression may capture important aspects of COPD-related biology that predict disease activity. Objectives: Develop a transcriptional risk score (TRS) for COPD and assess the contribution of the TRS and a polygenic risk score (PRS) for disease susceptibility and progression. Methods: We randomly split 2,569 COPDGene (Genetic Epidemiology of COPD) participants with whole-blood RNA sequencing into training (n = 1,945) and testing (n = 624) samples and used 468 ECLIPSE (Evaluation of COPD Longitudinally to Identify Predictive Surrogate End-points) COPD cases with microarray data for replication. We developed a TRS using penalized regression (least absolute shrinkage and selection operator) to model FEV1/FVC and studied the predictive value of TRS for COPD (Global Initiative for Chronic Obstructive Lung Disease 2-4), prospective FEV1 change (ml/yr), and additional COPD-related traits. We adjusted for potential confounders, including age and smoking. We evaluated the predictive performance of the TRS in the context of a previously derived PRS and clinical factors. Measurements and Main Results: The TRS included 147 transcripts and was associated with COPD (odds ratio, 3.3; 95% confidence interval [CI], 2.4-4.5; P < 0.001), FEV1 change (β, -17 ml/yr; 95% CI, -28 to -6.6; P = 0.002), and other COPD-related traits. In ECLIPSE cases, we replicated the association with FEV1 change (β, -8.2; 95% CI, -15 to -1; P = 0.025) and the majority of other COPD-related traits. Models including PRS, TRS, and clinical factors were more predictive of COPD (area under the receiver operator characteristic curve, 0.84) and annualized FEV1 change compared with models with one risk score or clinical factors alone. Conclusions: Blood transcriptomics can improve prediction of COPD and lung function decline when added to a PRS and clinical risk factors.

Comparison of Machine Learning Techniques for Mortality Prediction in a Prospective Cohort of Older Adults

As global demographics change, ageing is a global phenomenon which is increasingly of interest in our modern and rapidly changing society. Thus, the application of proper prognostic indices in clinical decisions regarding mortality prediction has assumed a significant importance for personalized risk management (i.e., identifying patients who are at high or low risk of death) and to help ensure effective healthcare services to patients. Consequently, prognostic modelling expressed as all-cause mortality prediction is an important step for effective patient management. Machine learning has the potential to transform prognostic modelling. In this paper, results on the development of machine learning models for all-cause mortality prediction in a cohort of healthy older adults are reported. The models are based on features covering anthropometric variables, physical and lab examinations, questionnaires, and lifestyles, as well as wearable data collected in free-living settings, obtained for the "Healthy Ageing Initiative" study conducted on 2291 recruited participants. Several machine learning techniques including feature engineering, feature selection, data augmentation and resampling were investigated for this purpose. A detailed empirical comparison of the impact of the different techniques is presented and discussed. The achieved performances were also compared with a standard epidemiological model. This investigation showed that, for the dataset under consideration, the best results were achieved with Random UnderSampling in conjunction with Random Forest (either with or without probability calibration). However, while including probability calibration slightly reduced the average performance, it increased the model robustness, as indicated by the lower 95% confidence intervals. The analysis showed that machine learning models could provide comparable results to standard epidemiological models while being completely data-driven and disease-agnostic, thus demonstrating the opportunity for building machine learning models on health records data for research and clinical practice. However, further testing is required to significantly improve the model performance and its robustness.

Analyzing the use of artificial intelligence for the management of chronic obstructive pulmonary disease (COPD)

OBJECTIVE

Chronic obstructive pulmonary disease (COPD) is a disease that causes airflow limitation to the lungs and has a high morbidity around the world. The objective of this study was to evaluate how artificial intelligence (AI) is being applied for the management of the disease, analyzing the objectives that are raised, the algorithms that are used and what results they offer.

METHODS

We conducted a scoping review following the Arksey and O'Malley (2005) and Levac et al. (2010) guidelines. Two reviewers independently searched, analyzed and extracted data from papers of five databases: Web of Science, PubMed, Scopus, Cinahl and Cochrane. To be included, the studies had to apply some AI techniques for the management of at least one stage of the COPD clinical process. In the event of any discrepancy between both reviewers, the criterion of a third reviewer prevailed.

RESULTS

380 papers were identified through database searches. After applying the exclusion criteria, 67 papers were included in the study. The studies were of a different nature and pursued a wide range of objectives, highlighting mainly those focused on the identification, classification and prevention of the disease. Neural nets, support vector machines and decision trees were the AI algorithms most commonly used. The mean and median values of all the performance metrics evaluated were between 80% and 90%.

CONCLUSIONS

The results obtained show a growing interest in the development of medical applications that manage the different phases of the COPD clinical process, especially predictive models. According to the performance shown, these models could be a useful complementary tool in the decision-making by health specialists, although more high-quality ML studies are needed to endorse the findings of this study.

Deep radiomics-based survival prediction in patients with chronic obstructive pulmonary disease

Heterogeneous clinical manifestations and progression of chronic obstructive pulmonary disease (COPD) affect patient health risk assessment, stratification, and management. Pulmonary function tests are used to diagnose and classify the severity of COPD, but they cannot fully represent the type or range of pathophysiologic abnormalities of the disease. To evaluate whether deep radiomics from chest computed tomography (CT) images can predict mortality in patients with COPD, we designed a convolutional neural network (CNN) model for extracting representative features from CT images and then performed random survival forest to predict survival in COPD patients. We trained CNN-based binary classifier based on six-minute walk distance results (> 440 m or not) and extracted high-throughput image features (i.e., deep radiomics) directly from the last fully connected layer of it. The various sizes of fully connected layers and combinations of deep features were experimented using a discovery cohort with 344 patients from the Korean Obstructive Lung Disease cohort and an external validation cohort with 102 patients from Penang General Hospital in Malaysia. In the integrative analysis of discovery and external validation cohorts, with combining 256 deep features from the coronal slice of the vertebral body and two sagittal slices of the left/right lung, deep radiomics for survival prediction achieved concordance indices of 0.8008 (95% CI, 0.7642–0.8373) and 0.7156 (95% CI, 0.7024–0.7288), respectively. Deep radiomics from CT images could be used to predict mortality in COPD patients.

Artificial Intelligence and Machine Learning in Chronic Airway Diseases: Focus on Asthma and Chronic Obstructive Pulmonary Disease

Chronic airway diseases are characterized by airway inflammation, obstruction, and remodeling and show high prevalence, especially in developing countries. Among them, asthma and chronic obstructive pulmonary disease (COPD) show the highest morbidity and socioeconomic burden worldwide. Although there are extensive guidelines for the prevention, early diagnosis, and rational treatment of these lifelong diseases, their value in precision medicine is very limited. Artificial intelligence (AI) and machine learning (ML) techniques have emerged as effective methods for mining and integrating large-scale, heterogeneous medical data for clinical practice, and several AI and ML methods have recently been applied to asthma and COPD. However, very few methods have significantly contributed to clinical practice. Here, we review four aspects of AI and ML implementation in asthma and COPD to summarize existing knowledge and indicate future steps required for the safe and effective application of AI and ML tools by clinicians.