Figure - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
Background: Distinguishing ICU patients with candidaemia can help with the precise prescription of antifungal drugs to create personalized guidelines. Previous prediction models of candidaemia have primarily used traditional logistic models and had some limitations. In this study, we developed a machine learning algorithm trained to predict candida...
Context in source publication
Citations
... An update of the search, employing the same combinations of key words, was performed on 28 November 2023. Overall, four original studies were eventually selected for inclusion in the present review [15][16][17][18]. The main characteristics of included studies are reported in Table 1. ...
... In 2021, Yuan and colleagues also assessed the performance of four different ML algorithms (XGBoost, support vector machine, RF and ExtraTrees) and logistic regression for the early diagnosis of candidemia in ICUs patients with systemic inflammatory response syndrome (SIRS) [15]. The study was multicenter and retrospective, and included a total of 137 patients with SIRS and candidemia and 7795 patients with SIRS and blood cultures negative or positive for a pathogen other than Candida species. ...
... The ML algorithm showing the best predictive performance in the study population was XGBoost, with 84% sensitivity and 89% specificity. The model also showed a very high negative predictive value (NPV) of 99.6%, partly connected to the low prevalence of candidemia in the study population (1.7%) [15]. ...
... He et al. used ML to establish predictive models for secondary candidemia in patients with systemic inflammatory response syndrome (SIRS) patients in the ICU. These models have a potential guiding role in the antifungal treatment of critically ill patients with SIRS (30). Researchers often focus on the pathogenic state and non-pathogenic state of fungi, which are known as "infection" and "colonization, " respectively (31, 32). ...
... The risk of fungal infection increased significantly after fungal colonization in ICU patients. One study found that 93 out of 137 (68%) patients with candidemia had Candida colonization (30). The preconception was that fungal infection is opportunistic. ...
Background
Fungal infections are associated with high morbidity and mortality in the intensive care unit (ICU), but their diagnosis is difficult. In this study, machine learning was applied to design and define the predictive model of ICU-acquired fungi (ICU-AF) in the early stage of fungal infections using Random Forest.
Objectives
This study aimed to provide evidence for the early warning and management of fungal infections.
Methods
We analyzed the data of patients with culture-positive fungi during their admission to seven ICUs of the First Affiliated Hospital of Chongqing Medical University from January 1, 2015, to December 31, 2019. Patients whose first culture was positive for fungi longer than 48 h after ICU admission were included in the ICU-AF cohort. A predictive model of ICU-AF was obtained using the Least Absolute Shrinkage and Selection Operator and machine learning, and the relationship between the features within the model and the disease severity and mortality of patients was analyzed. Finally, the relationships between the ICU-AF model, antifungal therapy and empirical antifungal therapy were analyzed.
Results
A total of 1,434 cases were included finally. We used lasso dimensionality reduction for all features and selected six features with importance ≥0.05 in the optimal model, namely, times of arterial catheter, enteral nutrition, corticosteroids, broadspectrum antibiotics, urinary catheter, and invasive mechanical ventilation. The area under the curve of the model for predicting ICU-AF was 0.981 in the test set, with a sensitivity of 0.960 and specificity of 0.990. The times of arterial catheter (p = 0.011, OR = 1.057, 95% CI = 1.053–1.104) and invasive mechanical ventilation (p = 0.007, OR = 1.056, 95%CI = 1.015–1.098) were independent risk factors for antifungal therapy in ICU-AF. The times of arterial catheter (p = 0.004, OR = 1.098, 95%CI = 0.855–0.970) were an independent risk factor for empirical antifungal therapy.
Conclusion
The most important risk factors for ICU-AF are the six time-related features of clinical parameters (arterial catheter, enteral nutrition, corticosteroids, broadspectrum antibiotics, urinary catheter, and invasive mechanical ventilation), which provide early warning for the occurrence of fungal infection. Furthermore, this model can help ICU physicians to assess whether empiric antifungal therapy should be administered to ICU patients who are susceptible to fungal infections.
... Most of the reviewed studies were conducted within a single institution; however, three studies utilized 188 datasets encompassing two hospital systems [20,22,44], and two studies expanded their analysis to incorporate 189 multi-center data [41,49]. External validation, which is critical for the generalizability of findings, was 190 performed in four studies [38,40,41,44]. ...
... Most studies reported imbalanced dataset with prevalence rates of BSI as given in the Table 265 1. To overcome challenges with data imbalance, Synthetic Minority Over-sampling Technique (SMOTE) was 266 widely used for augmenting the minority class in the dataset by generating synthetic samples [21,30,49,51]. compliance with data protection regulations researchers can implement effective deidentification of patient 279 records, involving elimination or alteration of direct identifiers, such as names, age, gender, or location, which 280 could be combined to identify an individual. ...
... (which was not certified by peer review) patient groups. ICU-focused research numbered six studies[44][45][46][47][48][49], again with two concentrating on specific 176 patient subsets. Bacteremia was the primary condition investigated in 24 studies, alongside candidemia and 177 central line-associated bloodstream infection (CLABSI) in others. ...
Purpose: Bloodstream infections (BSIs) present significant public health challenges. With the advent of machine learning (ML), promising predictive models have been developed. This study evaluates their performance through a systematic review and meta-analysis.
Methods: We performed a comprehensive systematic review across multiple databases, including PubMed, IEEE Xplore, ScienceDirect, ACM Digital Library, SpringerLink, Web of Science, Scopus, and Google Scholar. Eligible studies focused on BSIs within any hospital setting, employing ML models as the diagnostic test. We evaluated the risk of bias with the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) checklist and assessed the quality of evidence using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach. Models reporting the area under the receiver operating characteristic curve (AUROC) were included in the meta-analysis to identify key performance drivers.
Results: After screening, a total of 30 studies were eligible for synthesis, from which 41 models and 8 data types were extracted. Most of the studies were carried out in the inpatient settings (n=17; 56%), followed by the emergency department (ED) settings (n=7; 23%), and followed by the ICU settings (n=6; 20%). The reported AUROCs in the hospital inpatients settings, ranged from 0.51-0.866, in the ICU settings AUROCs ranged from 0.668-0.970, and in the emergency department (ED) settings the AUROCs of the models ranged from 0.728-0.844. One study reported prospective cohort study, while two prospectively validated their models. In the meta-analysis, laboratory tests, Complete Blood Count/Differential Count (CBC/DC), and ML model type contributed the most to model performance.
Conclusion: This systematic review and meta-analysis show that on retrospective data, individual ML models can accurately predict BSIs at different stages of patient trajectory. Although they enable early prediction of BSI, a comprehensive approach to integrate data types and models is necessary. Systematic reporting, externally validated, and clinical implementation studies are needed to establish clinical confidence.
... The insulin features of the private dataset were predicted using a semi-supervised model with heavy gradient boosting. The SMOTE and ADASYN methods [13] were used to the problem of class disproportion. With an accuracy of 81%, an F1 coefficient of 0.81, and an area under the curve (AUC) of 0.84, the suggested system outperformed the XGBoost classifier using the ADASYN method. ...
Detection and management of diabetes at an early stage is essential since it is rapidly becoming a global health crisis in many countries. Predictions of diabetes using machine learning algorithms have been promising. In this work, we use data collected from the Pima Indians to assess the performance of multiple machine-learning approaches to diabetes prediction. Ages, body mass indexes, and glucose levels for 768 patients are included in the data set. The methods evaluated are Logistic Regression, Decision Tree, Random Forest, k-Nearest Neighbors, Naive Bayes, Support Vector Machine, Gradient Boosting, and Neural Network. The findings indicate that the Logistic Regression and Neural Network models perform the best on most criteria when considering all classes together. The SVM, Random Forest, and Naive Bayes models also receive moderate to high scores, suggesting their strength as classification models. However, the kNN and Tree models show poorer scores on most criteria across all classes, making them less favorable choices for this dataset. The SGD, AdaBoost, and CN2 rule inducer models perform the poorest when comparing all models using a weighted average of class scores. The results of the study suggest that machine learning algorithms may help predict the onset of diabetes and for detecting the disease at an early stage.
... Machine learning techniques for risk prediction and factor identification have only recently started to be used in AFF-IFIs Mayer et al., 2022;Yan et al., 2022;Yuan et al., 2021). Only Potter et al. (2019) developed a decision support system based on probabilistic graphical models for combat-related AFF-IFI patients. ...
Machine learning models are increasingly used in the medical domain to study the association between risk factors and diseases to support practitioners in predicting health outcomes. In this paper, we showcase the use of machine-learned staged tree models for investigating complex asymmetric dependence structures in health data. Staged trees are a specific class of generative, probabilistic graphical models that formally model asymmetric conditional independence and non-regular sample spaces. An investigation of the risk factors in invasive fungal infections demonstrates the insights staged trees provide to support medical decision-making.
... Gu, J et al. developed nomogram model to predict the occurrence of SIRS after PCNL and related factors [11]. Previous studies have shown that machine learning algorithms can be utilized to predict the occurrence of SIRS in intensive care unit (ICU) or emergency department (ED) patients [12]. ...
The objective of this study was to develop and compare the performance of nomogram model and machine learning models for predicting the possibility of systemic inflammatory response syndrome (SIRS) following percutaneous nephrolithotomy (PCNL). We retrospectively reviewed the clinical data of 337 patients who received PCNL between May 2020 and June 2022. Eighty percent of the data were used as the training set, and the remaining data were used as the testing set. The nomogram and machine learning (ML) models were created using the training set and were validated using the testing set. Based on the areas under the receiver operating characteristic curve (AUC) and the calibration curve, we evaluated the predictive ability of the nomogram. The predictive performance of six machine learning models was determined by the AUC and accuracy. Multivariate logistic regression analysis revealed four independent risk factors associated with SIRS, including preoperative monocyte, serum fibrinogen, serum prealbumin, and preoperative SII. The above independent related factors were used as variables to construct the nomogram model. Among the six machine learning algorithms, the support vector machine (SVM) delivered the best performance with accuracy of 0.926, AUC of 0.952 [95% Confidence Interval (CI): 0.906–0.999], while the nomogram showed an AUC of 0.818. Compared with the nomogram model, the SVM model can provide more reliable prognostic information about the possibility of SIRS after PCNL, which can assist surgeons in clinical decision-making.
... As a subset of machine learning algorithms, random forest algorithm can build a mathematical model based on sample data and be used to make predictions or decisions (11)(12)(13). The previous studies demonstrated that the prediction model based on random forest algorithm exhibited a high accuracy in predicting the development of end-stage renal disease (14). ...
Background
Post-operative mortality risk assessment for geriatric patients with hip fractures (HF) is a challenge for clinicians. Early identification of geriatric HF patients with a high risk of post-operative death is helpful for early intervention and improving clinical prognosis. However, a single significant risk factor of post-operative death cannot accurately predict the prognosis of geriatric HF patients. Therefore, our study aims to utilize a machine learning approach, random forest algorithm, to fabricate a prediction model for post-operative death of geriatric HF patients.
Methods
This retrospective study enrolled consecutive geriatric HF patients who underwent treatment for surgery. The study cohort was divided into training and testing datasets at a 70:30 ratio. The random forest algorithm selected or excluded variables according to the feature importance. Least absolute shrinkage and selection operator (Lasso) was utilized to compare feature selection results of random forest. The confirmed variables were used to create a simplified model instead of a full model with all variables. The prediction model was then verified in the training dataset and testing dataset. Additionally, a prediction model constructed by logistic regression was used as a control to evaluate the efficiency of the new prediction model.
Results
Feature selection by random forest algorithm and Lasso regression demonstrated that seven variables, including age, time from injury to surgery, chronic obstructive pulmonary disease (COPD), albumin, hemoglobin, history of malignancy, and perioperative blood transfusion, could be used to predict the 1-year post-operative mortality. The area under the curve (AUC) of the random forest algorithm-based prediction model in training and testing datasets were 1.000, and 0.813, respectively. While the prediction tool constructed by logistic regression in training and testing datasets were 0.895, and 0.797, respectively.
Conclusions
Compared with logistic regression, the random forest algorithm-based prediction model exhibits better predictive ability for geriatric HF patients with a high risk of death within post-operative 1 year.
... It has been widely used in the medical field, from radiology to surgery, from oncology to intensive care [5][6][7][8][9]. Supervised machine learning-based systems have been employed to predict patient deterioration risk [10,11], heart failure onset [12,13], acute kidney injury [14], delirium [15], sepsis [16][17][18][19] and mortality [20,21]. Unsupervised ML, on the other hand, has been used to analyze, cluster and manage large amounts of data that lie beyond clinicians' ability to handle them. ...
Machine learning and cluster analysis applied to the clinical setting of an intensive care unit can be a valuable aid for clinical management, especially with the increasing complexity of clinical monitoring. Providing a method to measure clinical experience, a proxy for that automatic gestalt evaluation that an experienced clinician sometimes effortlessly, but often only after long, hard consideration and consultation with colleagues, relies upon for decision making, is what we wanted to achieve with the application of machine learning to antibiotic therapy and clinical monitoring in the present work. This is a single-center retrospective analysis proposing methods for evaluation of vitals and antimicrobial therapy in intensive care patients. For each patient included in the present study, duration of antibiotic therapy, consecutive days of treatment and type and combination of antimicrobial agents have been assessed and considered as single unique daily record for analysis. Each parameter, composing a record was normalized using a fuzzy logic approach and assigned to five descriptive categories (fuzzy domain sub-sets ranging from “very low” to “very high”). Clustering of these normalized therapy records was performed, and each patient/day was considered to be a pertaining cluster. The same methodology was used for hourly bed-side monitoring. Changes in patient conditions (monitoring) can lead to a shift of clusters. This can provide an additional tool for assessing progress of complex patients. We used Fuzzy logic normalization to descriptive categories of parameters as a form nearer to human language than raw numbers.
Machine learning models are increasingly used in the medical domain to study the association between risk factors and diseases to support practitioners in understanding health outcomes. In this paper, we showcase the use of machine-learned staged tree models for investigating complex asymmetric dependence structures in health data. Staged trees are a specific class of generative, probabilistic graphical models that formally model asymmetric conditional independence and non-regular sample spaces. An investigation of the risk factors in invasive fungal infections demonstrates the insights staged trees provide to support medical decision-making.
