Machine Learning to Predict In-Hospital Mortality in COVID-19 Patients Using Computed Tomography-Derived Pulmonary and Vascular Features.
Ontology highlight
ABSTRACT:
SUBMITTER: Schiaffino S
PROVIDER: S-EPMC8230339 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundComputed tomography pulmonary angiography (CTPA) is a first-line noninvasive method to diagnose acute pulmonary thromboembolism (APE); however, whether chest noncontrast CT (NC-CT) could aid in the diagnosis of APE remains unknown. The aim of this study was to build and evaluate a holistic lung graph-based machine learning (HLG-ML) using NC-CT for the diagnosis of APE and to compare its performance with that of radiologists and the YEARS algorithm.MethodsThis study enrolled 178 cases (77 males; age 63.9±16.7 years) who underwent NC-CT and CTPA in the same day from January 2019 to December 2020. Of these patients, 133 (75% of cases; 58 males; age 65.4±15.6 years) were placed into a training group and 45 (25% of cases; 19 males; age 59.6±19.2 years) into a testing group. The other 43 cases (18 males; age 62.8±20.0 years) were used to externally validate the model between January 2021 and March 2022. A HLG was developed with a pulmonary radiomics descriptor derived from NC-CT images. The approach extracted local radiomics features and encoded these local features into a radiomics descriptor as a characterization of global radiomics feature distribution. Subsequently, 8 ML models were trained and compared based on the radiomics descriptor. In the validation group, area under the curves (AUCs) of the HLG-ML model in the diagnosis of APE were compared with those of the 3 radiologists and the YEARS algorithm.ResultsAmong the 8 ML models, gradient boosting decision tree demonstrated the best classification performance (AUC =0.772) on the training set. In the testing set, the AUC of gradient boosting decision trees was 0.857 [95% confidence intervals (CIs): 0.699-0.951]. In the validation set, the performance of gradient boosting decision tree (AUC =0.810; 95% CI: 0.669-0.952; Youden index =0.621) outperformed 3 radiologists (AUC =0.508, 95% CI: 0.335-0.681, Youden index =0.016; AUC =0.504, 95% CI: 0.354-0.654, Youden index =0.008; AUC =0.527, 95% CI: 0.363-0.691, Youden index =0.050) and the YEARS algorithm (AUC =0.618; 95% CI: 0.469-0.767; Youden index =0.237).ConclusionsCompared to all 3 radiologists and the YEARS algorithm, the proposed HLG-based gradient boosting decision tree model achieved a superior performance in the diagnosis of APE on the NC-CT and may thus serve as a valuable tool for physicians in the diagnosis of APE.
| S-EPMC10585544 | biostudies-literature
Project description:ObjectivesNon-contrast computed tomography (NCCT) markers are robust predictors of parenchymal hematoma expansion in intracerebral hemorrhage (ICH). We investigated whether NCCT features can also identify ICH patients at risk of intraventricular hemorrhage (IVH) growth.MethodsPatients with acute spontaneous ICH admitted at four tertiary centers in Germany and Italy were retrospectively included from January 2017 to June 2020. NCCT markers were rated by two investigators for heterogeneous density, hypodensity, black hole sign, swirl sign, blend sign, fluid level, island sign, satellite sign, and irregular shape. ICH and IVH volumes were semi-manually segmented. IVH growth was defined as IVH expansion > 1 mL (eIVH) or any delayed IVH (dIVH) on follow-up imaging. Predictors of eIVH and dIVH were explored with multivariable logistic regression. Hypothesized moderators and mediators were independently assessed in PROCESS macro models.ResultsA total of 731 patients were included, of whom 185 (25.31%) suffered from IVH growth, 130 (17.78%) had eIVH, and 55 (7.52%) had dIVH. Irregular shape was significantly associated with IVH growth (OR 1.68; 95%CI [1.16-2.44]; p = 0.006). In the subgroup analysis stratified by the IVH growth type, hypodensities were significantly associated with eIVH (OR 2.06; 95%CI [1.48-2.64]; p = 0.015), whereas irregular shape (OR 2.72; 95%CI [1.91-3.53]; p = 0.016) in dIVH. The association between NCCT markers and IVH growth was not mediated by parenchymal hematoma expansion.ConclusionsNCCT features identified ICH patients at a high risk of IVH growth. Our findings suggest the possibility to stratify the risk of IVH growth with baseline NCCT and might inform ongoing and future studies.Clinical relevance statementNon-contrast CT features identified ICH patients at a high risk of intraventricular hemorrhage growth with subtype-specific differences. Our findings may assist in the risk stratification of intraventricular hemorrhage growth with baseline CT and might inform ongoing and future clinical studies.Key points• NCCT features identified ICH patients at a high risk of IVH growth with subtype-specific differences. • The effect of NCCT features was not moderated by time and location or indirectly mediated by hematoma expansion. • Our findings may assist in the risk stratification of IVH growth with baseline NCCT and might inform ongoing and future studies.
| S-EPMC10598100 | biostudies-literature
Project description:Recent advancements in deep learning for automated image processing and classification have accelerated many new applications for medical image analysis. However, most deep learning algorithms have been developed using reconstructed, human-interpretable medical images. While image reconstruction from raw sensor data is required for the creation of medical images, the reconstruction process only uses a partial representation of all the data acquired. Here, we report the development of a system to directly process raw computed tomography (CT) data in sinogram-space, bypassing the intermediary step of image reconstruction. Two classification tasks were evaluated for their feasibility of sinogram-space machine learning: body region identification and intracranial hemorrhage (ICH) detection. Our proposed SinoNet, a convolutional neural network optimized for interpreting sinograms, performed favorably compared to conventional reconstructed image-space-based systems for both tasks, regardless of scanning geometries in terms of projections or detectors. Further, SinoNet performed significantly better when using sparsely sampled sinograms than conventional networks operating in image-space. As a result, sinogram-space algorithms could be used in field settings for triage (presence of ICH), especially where low radiation dose is desired. These findings also demonstrate another strength of deep learning where it can analyze and interpret sinograms that are virtually impossible for human experts.
| S-EPMC6820559 | biostudies-literature
Project description:BackgroundSepsis is a leading cause of morbidity and mortality in hospitalized patients. Up to now, there are no well-established longitudinal networks from molecular mechanisms to clinical phenotypes in sepsis. Adding to the problem, about one of the five patients presented with diabetes. For this subgroup, management is difficult, and prognosis is difficult to evaluate.MethodsFrom the three databases, a total of 7,001 patients were enrolled on the basis of sepsis-3 standard and diabetes diagnosis. Input variable selection is based on the result of correlation analysis in a handpicking way, and 53 variables were left. A total of 5,727 records were collected from Medical Information Mart for Intensive Care database and randomly split into a training set and an internal validation set at a ratio of 7:3. Then, logistic regression with lasso regularization, Bayes logistic regression, decision tree, random forest, and XGBoost were conducted to build the predictive model by using training set. Then, the models were tested by the internal validation set. The data from eICU Collaborative Research Database (n = 815) and dtChina critical care database (n = 459) were used to test the model performance as the external validation set.ResultsIn the internal validation set, the accuracy values of logistic regression with lasso regularization, Bayes logistic regression, decision tree, random forest, and XGBoost were 0.878, 0.883, 0.865, 0.883, and 0.882, respectively. Likewise, in the external validation set 1, lasso regularization = 0.879, Bayes logistic regression = 0.877, decision tree = 0.865, random forest = 0.886, and XGBoost = 0.875. In the external validation set 2, lasso regularization = 0.715, Bayes logistic regression = 0.745, decision tree = 0.763, random forest = 0.760, and XGBoost = 0.699.ConclusionThe top three models for internal validation set were Bayes logistic regression, random forest, and XGBoost, whereas the top three models for external validation set 1 were random forest, logistic regression, and Bayes logistic regression. In addition, the top three models for the external validation set 2 were decision tree, random forest, and Bayes logistic regression. Random forest model performed well with the training and three validation sets. The most important features are age, albumin, and lactate.
| S-EPMC9709414 | biostudies-literature
Project description:BackgroundA new clinical guideline for idiopathic pulmonary fibrosis (IPF) uses high-resolution computed tomography (HRCT) patterns for diagnostic purposes. However, it is unknown how they relate to the IPF clinical course. We aimed to investigate whether HRCT patterns could be used to predict lung function changes and survival in patients with IPF.MethodsClinical data were retrospectively reviewed in 337 patients with IPF (all biopsy-proven cases). HRCT patterns were classified according to the 2018 IPF diagnostic criteria.ResultsThe median follow-up was 46.9 months. The mean age was 62.5 years, and 74.2% were men. Among the HRCT patterns, usual interstitial pneumonia (UIP), probable UIP, indeterminate for UIP, and an alternative diagnosis were identified in 163 (48.4%), 110 (32.6%), 33 (9.8%), and 31 (9.2%) patients, respectively. The indeterminate for UIP group showed higher lung function and exercise capacity and better prognosis than the other groups. They also had a lesser decline in lung function than the other groups during follow-up. In the multivariate Cox analysis, which was adjusted by age, smoking status, lung function, exercise capacity, and use of antifibrotic agents, indeterminate for UIP pattern was found to be an independent prognostic factor (hazard ratio 0.559, 95% confidence interval 0.335-0.933, P = 0.026). However, the probable UIP group had similar lung function changes and prognosis when compared the UIP group.ConclusionsOur results suggest that indeterminate for UIP pattern on HRCT may predict a more favorable clinical course in patients with IPF, supporting the validity of the new IPF diagnostic guidelines.
| S-EPMC7653759 | biostudies-literature
Project description:BackgroundThe accurate preoperative identification of decompression levels is crucial for the success of surgery in patients with multi-level lumbar spinal stenosis (LSS). The objective of this study was to develop machine learning (ML) classifiers that can predict decompression levels using computed tomography myelography (CTM) data from LSS patients.MethodsA total of 1095 lumbar levels from 219 patients were included in this study. The bony spinal canal in CTM images was manually delineated, and radiomic features were extracted. The extracted data were randomly divided into training and testing datasets (8:2). Six feature selection methods combined with 12 ML algorithms were employed, resulting in a total of 72 ML classifiers. The main evaluation indicator for all classifiers was the area under the curve of the receiver operating characteristic (ROC-AUC), with the precision-recall AUC (PR-AUC) serving as the secondary indicator. The prediction outcome of ML classifiers was decompression level or not.ResultsThe embedding linear support vector (embeddingLSVC) was the optimal feature selection method. The feature importance analysis revealed the top 5 important features of the 15 radiomic predictors, which included 2 texture features, 2 first-order intensity features, and 1 shape feature. Except for shape features, these features might be eye-discernible but hardly quantified. The top two ML classifiers were embeddingLSVC combined with support vector machine (EmbeddingLSVC_SVM) and embeddingLSVC combined with gradient boosting (EmbeddingLSVC_GradientBoost). These classifiers achieved ROC-AUCs over 0.90 and PR-AUCs over 0.80 in independent testing among the 72 classifiers. Further comparisons indicated that EmbeddingLSVC_SVM appeared to be the optimal classifier, demonstrating superior discrimination ability, slight advantages in the Brier scores on the calibration curve, and Net benefits on the Decision Curve Analysis.ConclusionsML successfully extracted valuable and interpretable radiomic features from the spinal canal using CTM images, and accurately predicted decompression levels for LSS patients. The EmbeddingLSVC_SVM classifier has the potential to assist surgical decision making in clinical practice, as it showed high discrimination, advantageous calibration, and competitive utility in selecting decompression levels in LSS patients using canal radiomic features from CTM.
| S-EPMC10795799 | biostudies-literature
Project description:Increasingly more patients exposed to radiation from computed axial tomography (CT) will have a greater risk of developing tumors or cancer that are caused by cell mutation in the future. A minor dose level would decrease the number of these possible cases. However, this framework can result in medical specialists (radiologists) not being able to detect anomalies or lesions. This work explores a way of addressing these concerns, achieving the reduction of unnecessary radiation without compromising the diagnosis. We contribute with a novel methodology in the CT area to predict the precise radiation that a patient should be given to accomplish this goal. Specifically, from a real dataset composed of the dose data of over fifty thousand patients that have been classified into standardized protocols (skull, abdomen, thorax, pelvis, etc.), we eliminate atypical information (outliers), to later generate regression curves employing diverse well-known Machine Learning techniques. As a result, we have chosen the best analytical technique per protocol; a selection that was thoroughly carried out according to traditional dosimetry parameters to accurately quantify the dose level that the radiologist should apply in each CT test.
| S-EPMC6928694 | biostudies-literature
Project description:Coronary computed tomography angiography (CCTA) derived machine learning fractional flow reserve (ML-FFRCT) can assess the hemodynamic significance of coronary artery stenoses. We aimed to assess sex differences in the association of ML-FFRCT and incident cardiovascular outcomes. We studied a retrospective cohort of consecutive patients who underwent clinically indicated CCTA and single photon emission computed tomography (SPECT). Obstructive stenosis was defined as ≥ 70% stenosis severity in non-left main vessels or ≥ 50% in the left main coronary. ML-FFRCT was computed using a machine learning algorithm with significant stenosis defined as ML-FFRCT < 0.8. The primary outcome was a composite of death or non-fatal myocardial infarction (D/MI). Our study population consisted of 471 patients with mean (SD) age 65 (13) years, 53% men, and multiple comorbidities (78% hypertension, 66% diabetes, 81% dyslipidemia). Compared to men, women were less likely to have obstructive stenosis by CCTA (9% vs. 18%; p = 0.006), less multivessel CAD (4% vs. 6%; p = 0.25), lower prevalence of ML-FFRCT < 0.8 (39% vs. 44%; p = 0.23) and higher median (IQR) ML-FFRCT (0.76 (0.53-0.86) vs. 0.71 (0.47-0.84); p = 0.047). In multivariable adjusted models, there was no significant association between ML-FFRCT < 0.8 and D/MI [Hazard Ratio 0.82, 95% confidence interval (0.30, 2.20); p = 0.25 for interaction with sex.]. In a high-risk cohort of symptomatic patients who underwent CCTA and SPECT testing, ML-FFRCT was higher in women than men. There was no significant association between ML-FFRCT and incident mortality or MI and no evidence that the prognostic value of ML-FFRCT differs by sex.
| S-EPMC9381799 | biostudies-literature
Project description:BackgroundExposure to cigarette smoke has been shown to lead to vascular remodelling. Computed tomography (CT) imaging measures of vascular pruning have been associated with pulmonary vascular disease, an important morbidity associated with smoking. In this study we compare CT-based measures of distal vessel loss to histological vascular and parenchymal changes.MethodsA retrospective review of 80 patients who had undergone lung resection identified patients with imaging appropriate for three-dimensional (3D) vascular reconstruction (n=18) and a second group for two-dimensional (2D) analysis (n=19). Measurements of the volume of the small vessels (3D) and the cross-sectional area of the small vessels (<5?mm2 cross-section) were computed. Histological measures of cross-sectional area of the vasculature and loss of alveoli septa were obtained for all subjects.ResultsThe 2D cross-sectional area of the vasculature on CT imaging was associated with the histological vascular cross-sectional area (r=0.69; p=0.001). The arterial small vessel volume assessed by CT correlated with the histological vascular cross-sectional area (r=0.50; p=0.04), a relationship that persisted even when adjusted for CT-derived measures of emphysema in a regression model.ConclusionsLoss of small vessel volume in CT imaging of smokers is associated with histological loss of vascular cross-sectional area. Imaging-based quantification of pulmonary vasculature provides a noninvasive method to study the multiscale effects of smoking on the pulmonary circulation.
| S-EPMC7007984 | biostudies-literature
Project description:Pulmonary hypertension is characterized histologically by intimal and medial thickening in the small pulmonary arteries, eventually resulting in vascular "pruning." Computed tomography (CT)-based quantification of pruning is associated with clinical measures of pulmonary hypertension, but it is not established whether CT-based pruning correlates with histologic arterial remodeling. Our sample consisted of 138 patients who underwent resection for early-stage lung adenocarcinoma. From histologic sections, we identified small pulmonary arteries and measured the relative area comprising the intima and media (VWA%), with higher VWA% representing greater histologic remodeling. From pre-operative CTs, we used image analysis algorithms to calculate the small vessel volume fraction (BV5/TBV) as a CT-based indicator of pruning (lower BV5/TBV represents greater pruning). We investigated relationships of CT pruning and histologic remodeling using Pearson correlation, simple linear regression, and multivariable regression with adjustment for age, sex, height, weight, smoking status, and total pack-years. We also tested for effect modification by sex and smoking status. In primary models, more severe CT pruning was associated with greater histologic remodeling. The Pearson correlation coefficient between BV5/TBV and VWA% was -0.41, and in linear regression models, VWA% was 3.13% higher (95% CI: 1.95-4.31%, p < 0.0001) per standard deviation lower BV5/TBV. This association persisted after multivariable adjustment. We found no evidence that these relationships differed by sex or smoking status. Among individuals who underwent resection for lung adenocarcinoma, more severe CT-based vascular pruning was associated with greater histologic arterial remodeling. These findings suggest CT imaging may be a non-invasive indicator of pulmonary vascular pathology.
| S-EPMC8647266 | biostudies-literature