Random Forest-Based Prediction of Outcome and Mortality in Patients with Traumatic Brain Injury Undergoing Primary Decompressive Craniectomy

Establishing a differential diagnosis model between primary membranous nephropathy and non-primary membranous nephropathy by machine learning algorithms

CONTEXT

Four algorithms with relatively balanced complexity and accuracy in deep learning classification algorithm were selected for differential diagnosis of primary membranous nephropathy (PMN).

OBJECTIVE

This study explored the most suitable classification algorithm for PMN identification, and to provide data reference for PMN diagnosis research.

METHODS

A total of 500 patients were referred to Luo-he Central Hospital from 2019 to 2021. All patients were diagnosed with primary glomerular disease confirmed by renal biopsy, contained 322 cases of PMN, the 178 cases of non-PMN. Using the decision tree, random forest, support vector machine, and extreme gradient boosting (Xgboost) to establish a differential diagnosis model for PMN and non-PMN. Based on the true positive rate, true negative rate, false-positive rate, false-negative rate, accuracy, feature work area under the curve (AUC) of subjects, the best performance of the model was chosen.

RESULTS

The efficiency of the Xgboost model based on the above evaluation indicators was the highest, which the diagnosis of PMN of the sensitivity and specificity, respectively 92% and 96%.

CONCLUSIONS

The differential diagnosis model for PMN was established successfully and the efficiency performance of the Xgboost model was the best. It could be used for the clinical diagnosis of PMN.

Artificial Intelligence-Enhanced Neurocritical Care for Traumatic Brain Injury : Past, Present and Future

In neurointensive care units (NICUs), particularly in cases involving traumatic brain injury (TBI), swift and accurate decision-making is critical because of rapidly changing patient conditions and the risk of secondary brain injury. The use of artificial intelligence (AI) in NICU can enhance clinical decision support and provide valuable assistance in these complex scenarios. This article aims to provide a comprehensive review of the current status and future prospects of AI utilization in the NICU, along with the challenges that must be overcome to realize this. Presently, the primary application of AI in NICU is outcome prediction through the analysis of preadmission and high-resolution data during admission. Recent applications include augmented neuromonitoring via signal quality control and real-time event prediction. In addition, AI can integrate data gathered from various measures and support minimally invasive neuromonitoring to increase patient safety. However, despite the recent surge in AI adoption within the NICU, the majority of AI applications have been limited to simple classification tasks, thus leaving the true potential of AI largely untapped. Emerging AI technologies, such as generalist medical AI and digital twins, harbor immense potential for enhancing advanced neurocritical care through broader AI applications. If challenges such as acquiring high-quality data and ethical issues are overcome, these new AI technologies can be clinically utilized in the actual NICU environment. Emphasizing the need for continuous research and development to maximize the potential of AI in the NICU, we anticipate that this will further enhance the efficiency and accuracy of TBI treatment within the NICU.

Predicting 30-day mortality in severely injured elderly patients with trauma in Korea using machine learning algorithms: a retrospective study

PURPOSE

The number of elderly patients with trauma is increasing; therefore, precise models are necessary to estimate the mortality risk of elderly patients with trauma for informed clinical decision-making. This study aimed to develop machine learning based predictive models that predict 30-day mortality in severely injured elderly patients with trauma and to compare the predictive performance of various machine learning models.

METHODS

This study targeted patients aged ≥65 years with an Injury Severity Score of ≥15 who visited the regional trauma center at Chungbuk National University Hospital between 2016 and 2022. Four machine learning models-logistic regression, decision tree, random forest, and eXtreme Gradient Boosting (XGBoost)-were developed to predict 30-day mortality. The models' performance was compared using metrics such as area under the receiver operating characteristic curve (AUC), accuracy, precision, recall, specificity, F1 score, as well as Shapley Additive Explanations (SHAP) values and learning curves.

RESULTS

The performance evaluation of the machine learning models for predicting mortality in severely injured elderly patients with trauma showed AUC values for logistic regression, decision tree, random forest, and XGBoost of 0.938, 0.863, 0.919, and 0.934, respectively. Among the four models, XGBoost demonstrated superior accuracy, precision, recall, specificity, and F1 score of 0.91, 0.72, 0.86, 0.92, and 0.78, respectively. Analysis of important features of XGBoost using SHAP revealed associations such as a high Glasgow Coma Scale negatively impacting mortality probability, while higher counts of transfused red blood cells were positively correlated with mortality probability. The learning curves indicated increased generalization and robustness as training examples increased.

CONCLUSIONS

We showed that machine learning models, especially XGBoost, can be used to predict 30-day mortality in severely injured elderly patients with trauma. Prognostic tools utilizing these models are helpful for physicians to evaluate the risk of mortality in elderly patients with severe trauma.

Machine learning prediction models for in-hospital postoperative functional outcome after moderate-to-severe traumatic brain injury

AIM

This study aims to utilize machine learning (ML) and logistic regression (LR) models to predict surgical outcomes among patients with traumatic brain injury (TBI) based on admission examination, assisting in making optimal surgical treatment decision for these patients.

METHOD

We conducted a retrospective review of patients hospitalized in our department for moderate-to-severe TBI. Patients admitted between October 2011 and October 2022 were assigned to the training set, while patients admitted between November 2022 and May 2023 were designated as the external validation set. Five ML algorithms and LR model were employed to predict the postoperative Glasgow Outcome Scale (GOS) status at discharge using clinical and routine blood data collected upon admission. The Shapley (SHAP) plot was utilized for interpreting the models.

RESULTS

A total of 416 patients were included in this study, and they were divided into the training set (n = 396) and the external validation set (n = 47). The ML models, using both clinical and routine blood data, were able to predict postoperative GOS outcomes with area under the curve (AUC) values ranging from 0.860 to 0.900 during the internal cross-validation and from 0.801 to 0.890 during the external validation. In contrast, the LR model had the lowest AUC values during the internal and external validation (0.844 and 0.567, respectively). When blood data was not available, the ML models achieved AUCs of 0.849 to 0.870 during the internal cross-validation and 0.714 to 0.861 during the external validation. Similarly, the LR model had the lowest AUC values (0.821 and 0.638, respectively). Through repeated cross-validation analysis, we found that routine blood data had a significant association with higher mean AUC values in all ML and LR models. The SHAP plot was used to visualize the contributions of all predictors and highlighted the significance of blood data in the lightGBM model.

CONCLUSION

The study concluded that ML models could provide rapid and accurate predictions for postoperative GOS outcomes at discharge following moderate-to-severe TBI. The study also highlighted the crucial role of routine blood tests in improving such predictions, and may contribute to the optimization of surgical treatment decision-making for patients with TBI.

A Machine Learning Application to Classify Patients at Differing Levels of Risk of Opioid Use Disorder: Clinician-Based Validation Study

Background

Despite restrictive opioid management guidelines, opioid use disorder (OUD) remains a major public health concern. Machine learning (ML) offers a promising avenue for identifying and alerting clinicians about OUD, thus supporting better clinical decision-making regarding treatment.

Objective

This study aimed to assess the clinical validity of an ML application designed to identify and alert clinicians of different levels of OUD risk by comparing it to a structured review of medical records by clinicians.

Methods

The ML application generated OUD risk alerts on outpatient data for 649,504 patients from 2 medical centers between 2010 and 2013. A random sample of 60 patients was selected from 3 OUD risk level categories (n=180). An OUD risk classification scheme and standardized data extraction tool were developed to evaluate the validity of the alerts. Clinicians independently conducted a systematic and structured review of medical records and reached a consensus on a patient's OUD risk level, which was then compared to the ML application's risk assignments.

Results

A total of 78,587 patients without cancer with at least 1 opioid prescription were identified as follows: not high risk (n=50,405, 64.1%), high risk (n=16,636, 21.2%), and suspected OUD or OUD (n=11,546, 14.7%). The sample of 180 patients was representative of the total population in terms of age, sex, and race. The interrater reliability between the ML application and clinicians had a weighted kappa coefficient of 0.62 (95% CI 0.53-0.71), indicating good agreement. Combining the high risk and suspected OUD or OUD categories and using the review of medical records as a gold standard, the ML application had a corrected sensitivity of 56.6% (95% CI 48.7%-64.5%) and a corrected specificity of 94.2% (95% CI 90.3%-98.1%). The positive and negative predictive values were 93.3% (95% CI 88.2%-96.3%) and 60.0% (95% CI 50.4%-68.9%), respectively. Key themes for disagreements between the ML application and clinician reviews were identified.

Conclusions

A systematic comparison was conducted between an ML application and clinicians for identifying OUD risk. The ML application generated clinically valid and useful alerts about patients' different OUD risk levels. ML applications hold promise for identifying patients at differing levels of OUD risk and will likely complement traditional rule-based approaches to generating alerts about opioid safety issues.

Artificial Intelligence and Machine Learning Applications in Critically Ill Brain Injured Patients

The utilization of Artificial Intelligence (AI) and Machine Learning (ML) is paving the way for significant strides in patient diagnosis, treatment, and prognostication in neurocritical care. These technologies offer the potential to unravel complex patterns within vast datasets ranging from vast clinical data and EEG (electroencephalogram) readings to advanced cerebral imaging facilitating a more nuanced understanding of patient conditions. Despite their promise, the implementation of AI and ML faces substantial hurdles. Historical biases within training data, the challenge of interpreting multifaceted data streams, and the "black box" nature of ML algorithms present barriers to widespread clinical adoption. Moreover, ethical considerations around data privacy and the need for transparent, explainable models remain paramount to ensure trust and efficacy in clinical decision-making.This article reflects on the emergence of AI and ML as integral tools in neurocritical care, discussing their roles from the perspective of both their scientific promise and the associated challenges. We underscore the importance of extensive validation in diverse clinical settings to ensure the generalizability of ML models, particularly considering their potential to inform critical medical decisions such as withdrawal of life-sustaining therapies. Advancement in computational capabilities is essential for implementing ML in clinical settings, allowing for real-time analysis and decision support at the point of care. As AI and ML are poised to become commonplace in clinical practice, it is incumbent upon health care professionals to understand and oversee these technologies, ensuring they adhere to the highest safety standards and contribute to the realization of personalized medicine. This engagement will be pivotal in integrating AI and ML into patient care, optimizing outcomes in neurocritical care through informed and data-driven decision-making.

Establishment of a model to predict mortality after decompression craniotomy for traumatic brain injury

BACKGROUND

The mortality rate of patients with traumatic brain injury (TBI) is still high even while undergoing decompressive craniectomy (DC), and the expensive treatment costs bring huge economic burden to the families of patients.

OBJECTIVE

The aim of this study was to identify preoperative indicators that influence patient outcomes and to develop a risk model for predicting patient mortality by a retrospective analysis of TBI patients undergoing DC.

METHODS

A total of 288 TBI patients treated with DC, admitted to the First Affiliated Hospital of Shantou University Medical School from August 2015 to April 2021, were used for univariate and multivariate logistic regression analysis to determine the risk factors for death after DC in TBI patients. We also built a risk model for the identified risk factors and conducted internal verification and model evaluation.

RESULTS

Univariate and multivariate logistic regression analysis identified four risk factors: Glasgow Coma Scale, age, activated partial thrombin time, and mean CT value of the superior sagittal sinus. These risk factors can be obtained before DC. In addition, we also developed a 3-month mortality risk model and conducted a bootstrap 1000 resampling internal validation, with C-indices of 0.852 and 0.845, respectively.

CONCLUSIONS

We developed a risk model that has clinical significance for the early identification of patients who will still die after DC. Interestingly, we also identified a new early risk factor for TBI patients after DC, that is, preoperative mean CT value of the superior sagittal sinus (p < .05).

Comprehensive analysis of the interaction of antigen presentation during anti-tumour immunity and establishment of AIDPS systems in ovarian cancer

There are hundreds of prognostic models for ovarian cancer. These genes are based on different gene classes, and there are many ways to construct the models. Therefore, this paper aims to build the most stable prognostic evaluation system known to date through 101 machine learning strategies. We combined 101 algorithm combinations with 10 machine learning algorithms to create antigen presentation-associated genetic markers (AIDPS) with outstanding precision and steady performance. The inclusive set of algorithms comprises the elastic network (Enet), Ridge, stepwise Cox, Lasso, generalized enhanced regression model (GBM), random survival forest (RSF), supervised principal component (SuperPC), Cox partial least squares regression (plsRcox), survival support vector machine (Survival-SVM). Then, in the train cohort, the prediction model was fitted using a leave-one cross-validation (LOOCV) technique, which involved 101 different possible combinations of prognostic genes. Seven validation data sets (GSE26193, GSE26712, GSE30161, GSE63885, GSE9891, GSE140082 and ICGC_OV_AU) were compared and analysed, and the C-index was calculated. Finally, we collected 32 published ovarian cancer prognostic models (including mRNA and lncRNA). All data sets and prognostic models were subjected to a univariate Cox regression analysis, and the C-index was calculated to demonstrate that the antigen presentation process should be the core criterion for evaluating ovarian cancer prognosis. In a univariate Cox regression analysis, 22 prognostic genes were identified based on the expression profiles of 283 genes involved in antigen presentation and the intersection of genes (p < 0.05). AIDPS were developed by our machine learning-based integration method, which was applied to these 22 genes. One hundred and one prediction models are fitted using the LOOCV framework, and the C-index is calculated for each model across all validation sets. Interestingly, RSF + Lasso was the best model overall since it had the greatest average C-index and the highest C-index of any combination of models tested on the validated data sets. In comparing external cohorts, we found that the C-index correlated AIDPS method using the RSF + Lasso method in 101 prediction models was in contrast to other features. Notably, AIDPS outperformed the vast majority of models across all data sets. Antigen-presenting anti-tumour immune pathways can be used as a representative gene set of ovarian cancer to track the prognosis of patients with cancer. The antigen-presenting model obtained by the RSF + Lasso method has the best C-INDEX, which plays a key role in developing antigen-presenting targeted drugs in ovarian cancer and improving the treatment outcome of patients.

An integrative multi-omics approach to characterize interactions between tuberculosis and diabetes mellitus

Tuberculosis-diabetes mellitus (TB-DM) is linked to a distinct inflammatory profile, which can be assessed using multi-omics analyses. Here, a machine learning algorithm was applied to multi-platform data, including cytokines and gene expression in peripheral blood and eicosanoids in urine, in a Brazilian multi-center TB cohort. There were four clinical groups: TB-DM(n = 24), TB only(n = 28), DM(HbA1c ≥ 6.5%) only(n = 11), and a control group of close TB contacts who did not have TB or DM(n = 13). After cross-validation, baseline expression or abundance of MMP-28, LTE-4, 11-dTxB2, PGDM, FBXO6, SECTM1, and LINCO2009 differentiated the four patient groups. A distinct multi-omic-derived, dimensionally reduced, signature was associated with TB, regardless of glycemic status. SECTM1 and FBXO6 mRNA levels were positively correlated with sputum acid-fast bacilli grade in TB-DM. Values of the biomarkers decreased during the course of anti-TB therapy. Our study identified several markers associated with the pathophysiology of TB-DM that could be evaluated in future mechanistic investigations.

Effect of topographic comparison of electroencephalographic microstates on the diagnosis and prognosis prediction of patients with prolonged disorders of consciousness

AIMS

The electroencephalography (EEG) microstates are indicative of fundamental information processing mechanisms, which are severely damaged in patients with prolonged disorders of consciousness (pDoC). We aimed to improve the topographic analysis of EEG microstates and explore indicators available for diagnosis and prognosis prediction of patients with pDoC, which were still lacking.

METHODS

We conducted EEG recordings on 59 patients with pDoC and 32 healthy controls. We refined the microstate method to accurately estimate topographical differences, and then classify and forecast the prognosis of patients with pDoC. An independent dataset was used to validate the conclusion.

RESULTS

Through optimized topographic analysis, the global explained variance (GEV) of microstate E increased significantly in groups with reduced levels of consciousness. However, its ability to classify the VS/UWS group was poor. In addition, the optimized GEV of microstate E exhibited a statistically significant decrease in the good prognosis group as opposed to the group with a poor prognosis. Furthermore, the optimized GEV of microstate E strongly predicted a patient's prognosis.

CONCLUSION

This technique harmonizes with the existing microstate analysis and exhibits precise and comprehensive differences in microstate topography between groups. Furthermore, this method has significant potential for evaluating the clinical prognosis of pDoC patients.

Prediction of Prostate Cancer Risk Stratification Based on A Nonlinear Transformation Stacking Learning Strategy

PURPOSE

Prostate cancer (PCa) is an epithelial malignancy that originates in the prostate gland and is generally categorized into low, intermediate, and high-risk groups. The primary diagnostic indicator for PCa is the measurement of serum prostate-specific antigen (PSA) values. However, reliance on PSA levels can result in false positives, leading to unnecessary biopsies and an increased risk of invasive injuries. Therefore, it is imperative to develop an efficient and accurate method for PCa risk stratification. Many recent studies on PCa risk stratification based on clinical data have employed a binary classification, distinguishing between low to intermediate and high risk. In this paper, we propose a novel machine learning (ML) approach utilizing a stacking learning strategy for predicting the tripartite risk stratification of PCa.

METHODS

Clinical records, featuring attributes selected using the lasso method, were utilized with 5 ML classifiers. The outputs of these classifiers underwent transformation by various nonlinear transformers and were then concatenated with the lasso-selected features, resulting in a set of new features. A stacking learning strategy, integrating different ML classifiers, was developed based on these new features.

RESULTS

Our proposed approach demonstrated superior performance, achieving an accuracy of 0.83 and an area under the receiver operating characteristic curve value of 0.88 in a dataset comprising 197 PCa patients with 42 clinical characteristics.

CONCLUSION

This study aimed to improve clinicians' ability to rapidly assess PCa risk stratification while reducing the burden on patients. This was achieved by using artificial intelligence-related technologies as an auxiliary method for diagnosing PCa.

Classification and Regression Tree Predictive Model for Acute Kidney Injury in Traumatic Brain Injury Patients

Background

Acute kidney injury (AKI) is prevalent in hospitalized patients with traumatic brain injury (TBI), and increases the risk of poor outcomes. We designed this study to develop a visual and convenient decision-tree-based model for predicting AKI in TBI patients.

Methods

A total of 376 patients admitted to the emergency department of the West China Hospital for TBI between January 2015 and June 2019 were included. Demographic information, vital signs on admission, laboratory test results, radiological signs, surgical options, and medications were recorded as variables. AKI was confirmed since the second day after admission, based on the Kidney Disease Improving Global Outcomes criteria. We constructed two predictive models for AKI using least absolute shrinkage and selection operator (LASSO) regression and classification and regression tree (CART), respectively. Receiver operating characteristic (ROC) curves of these two predictive models were drawn, and the area under the ROC curve (AUC) was calculated to compare their predictive accuracy.

Results

The incidence of AKI on the second day after admission was 10.4% among patients with TBI. Lasso regression identified five potent predictive factors for AKI: glucose, serum creatinine, cystatin C, serum uric acid, and fresh frozen plasma transfusions. The CART analysis showed that glucose, serum uric acid, and cystatin C ranked among the top three in terms of the feature importance of the decision tree model. The AUC value of the decision-tree predictive model was 0.892, which was higher than the 0.854 of the LASSO regression model, although the difference was not statistically significant.

Conclusion

The decision tree model is valuable for predicting AKI among patients with TBI. This tree-based flowchart is convenient for physicians to identify patients with TBI who are at high risk of AKI and prompts them to develop suitable therapeutic strategies.

MULTIPLE MACHINE LEARNING METHODS AND COMPARATIVE TRANSCRIPTOMICS IDENTIFY PIVOTAL GENES FOR ISCHEMIA-REPERFUSION INJURY IN HUMAN DONOR TISSUE UNDERGOING ORTHOTOPIC LIVER TRANSPLANTATION

ABSTRACT

Background: Hepatic ischemia-reperfusion injury (HIRI) is a major complication affecting patient prognosis during the period after orthotopic liver transplantation (OLT). Although an increasing number of scientists have investigated the molecular biology of ischemia-reperfusion injury (IRI) during OLT in animal and cellular models in recent years, studies using comprehensive and high-quality sequencing results from human specimens to screen for key molecules are still lacking. Aims: The objective of this study is to explore the molecular biological pathways and key molecules associated with HIRI during OLT through RNA sequencing and related bioinformatics analysis techniques. Methods: The study was done by performing mRNA sequencing on liver tissue samples obtained from 15 cases of in situ liver transplantation patients who experienced ischemia and reperfusion injury within 1 year at Guizhou Medical University, and combined with bioinformatics analysis and machine learning methods, we identified the genes and transcription factors that are closely associated with IRI during in situ liver transplantation surgery. Results: There were 877 differentially expressed genes (DEGs) identified in the included liver samples, of which 817 DEGs were upregulated and 60 were downregulated. Functional enrichment analysis revealed significant enrichment of immune-related terms, such as inflammation, defense responses, responses to cytokines, immune system processes, and cellular activation. In addition, core gene enrichment analysis after cytoHubba screening suggested that liver reperfusion injury might be associated with translation-related elements as a pathway together with protein translation processes. Machine learning with the weighted correlation network analysis screening method identified PTGS2, IRF1, and CDKN1A as key genes in the reperfusion injury process. Conclusions: This study demonstrated that the pathways and genomes whose expression is altered throughout the reperfusion process might be critical for the progression of HIRI during OLT.

Long-term survival after primary decompressive craniectomy for severe traumatic brain injury: an observational study from 1 to 17 years

Primary decompressive craniectomy (DC) is carried out to prevent intracranial hypertension after removal of mass lesions resulting from traumatic brain injury (TBI). While primary DC can be a life-saving intervention, significant mortality risks persist during the follow-up period. This study was undertaken to investigate the long-term survival rate and ascertain the risk factors of mortality in TBI patients who underwent primary DC. We enrolled 162 head-injured patients undergoing primary DC in this retrospective study. The primary focus was on long-term mortality, which was monitored over a range of 12 to 209 months post-TBI. We compared the clinical parameters of survivors and non-survivors, and used a multivariate logistic regression model to adjust for independent risk factors of long-term mortality. For the TBI patients who survived the initial hospitalization period following surgery, the average duration of follow-up was 106.58 ± 65.45 months. The recorded long-term survival rate of all patients was 56.2% (91/162). Multivariate logistic regression analysis revealed that age (odds ratio, 95% confidence interval = 1.12, 1.07-1.18; p < 0.01) and the status of basal cisterns (absent versus normal; odds ratio, 95% confidence interval = 9.32, 2.05-42.40; p < 0.01) were the two independent risk factors linked to long-term mortality. In conclusion, this study indicated a survival rate of 56.2% for patients subjected to primary DC for TBI, with at least a one-year follow-up. Key risk factors associated with long-term mortality were advanced age and absent basal cisterns, critical considerations for developing effective TBI management strategies.

Identification of novel potential homologous repair deficiency-associated genes in pancreatic adenocarcinoma via WGCNA coexpression network analysis and machine learning

Homologous repair deficiency (HRD) impedes double-strand break repair, which is a common driver of carcinogenesis. Positive HRD status can be used as theranostic markers of response to platinum- and PARP inhibitor-based chemotherapies. Here, we aimed to fully investigate the therapeutic and prognostic potential of HRD in pancreatic adenocarcinoma (PAAD) and identify effective biomarkers related to HRD using comprehensive bioinformatics analysis. The HRD score was defined as the unweighted sum of the LOH, TAI, and LST scores, and it was obtained based on the previous literature. To characterize PAAD immune infiltration subtypes, the "ConsensusClusterPlus" package in R was used to conduct unsupervised clustering. A WGCNA was conducted to elucidate the gene coexpression modules and hub genes in the HRD-related gene module of PAAD. The functional enrichment study was performed using Metascape. LASSO analysis was performed using the "glmnet" package in R, while the random forest algorithm was realized using the "randomForest" package in R. The prognostic variables were evaluated using univariate Cox analysis. The prognostic risk model was built using the LASSO approach. ROC curve and KM survival analyses were performed to assess the prognostic potential of the risk model. The half-maximal inhibitory concentration (IC50) of the PARP inhibitors was estimated using the "pRRophetic" package in R and the Genomics of Drug Sensitivity in Cancer database. The "rms" package in R was used to create the nomogram. A high HRD score indicated a poor prognosis and an advanced clinical process in PAAD patients. PAAD tumors with high HRD levels revealed significant T helper lymphocyte depletion, upregulated levels of cancer stem cells, and increased sensitivity to rucaparib, Olaparib, and veliparib. Using WGCNA, 11 coexpression modules were obtained. The red module and 122 hub genes were identified as the most correlated with HRD in PAAD. Functional enrichment analysis revealed that the 122 hub genes were mainly concentrated in cell cycle pathways. One novel HRD-related gene signature consisting of CKS1B, HJURP, and TPX2 were screened via LASSO analysis and a random forest algorithm, and they were validated using independent validation sets. No direct association between HRD and CKS1B, HJURP, or TPX2 has not been reported in the literature so far. Thus, these findings indicated that CKS1B, HJURP, and TPX2 have potential as diagnostic and prognostic biomarkers for PAAD. We constructed a novel HRD-related prognostic model that provides new insights into PAAD prognosis and immunotherapy. Based on bioinformatics analysis, we comprehensively explored the therapeutic and prognostic potential of HRD in PAAD. One novel HRD-related gene signature consisting of CKS1B, HJURP, and TPX2 were identified through the combination of WGCNA, LASSO analysis and a random forest algorithm. A novel HRD-related risk model that can predict clinical prognosis and immunotherapeutic response in PAAD patients was constructed.

Novel Molecular Subtyping Scheme Based on In Silico Analysis of Cuproptosis Regulator Gene Patterns Optimizes Survival Prediction and Treatment of Hepatocellular Carcinoma

BACKGROUND

The liver plays an important role in maintaining copper homeostasis. Copper ion accumulation was elevated in HCC tissue samples. Copper homeostasis is implicated in cancer cell proliferation and angiogenesis. The potential of copper homeostasis as a new theranostic biomarker for molecular imaging and the targeted therapy of HCC has been demonstrated. Recent studies have reported a novel copper-dependent nonapoptotic form of cell death called cuproptosis, strikingly different from other known forms of cell death. The correlation between cuproptosis and hepatocellular carcinoma (HCC) is not fully understood.

MATERIALS AND METHODS

The transcriptomic data of patients with HCC were retrieved from the Cancer Genome Atlas-Liver Hepatocellular Carcinoma (TCGA-LIHC) and were used as a discovery cohort to construct the prognosis model. The gene expression data of patients with HCC retrieved from the International Cancer Genome Consortium (ICGC) and Gene Expression Omnibus (GEO) databases were used as the validation cohort. The Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis was used to construct the prognosis model. A principal component analysis (PCA) was used to evaluate the overall characteristics of cuproptosis regulator genes and obtain the PC1 and PC2 scores. Unsupervised clustering was performed using the ConsensusClusterPlus R package to identify the molecular subtypes of HCC. Cox regression analysis was performed to identify cuproptosis regulator genes that could predict the prognosis of patients with HCC. The receiver operating characteristics curve and Kaplan-Meier survival analysis were used to understand the role of hub genes in predicting the diagnosis and prognosis of patients, as well as the prognosis risk model. A weighted gene co-expression network analysis (WGCNA) was used for screening the cuproptosis subtype-related hub genes. The functional enrichment analysis was performed using Metascape. The 'glmnet' R package was used to perform the LASSO regression analysis, and the randomForest algorithm was performed using the 'randomForest' R package. The 'pRRophetic' R package was used to estimate the anticancer drug sensitivity based on the data retrieved from the Genomics of Drug Sensitivity in Cancer database. The nomogram was constructed using the 'rms' R package. Pearson's correlation analysis was used to analyze the correlations.

RESULTS

We constructed a six-gene signature prognosis model and a nomogram to predict the prognosis of patients with HCC. The Kaplan-Meier survival analysis revealed that patients with a high-risk score, which was predicted by the six-gene signature model, had poor prognoses (log-rank test p < 0.001; HR = 1.83). The patients with HCC were grouped into three distinct cuproptosis subtypes (Cu-clusters A, B, and C) based on the expression pattern of cuproptosis regulator genes. The patients in Cu-cluster B had poor prognosis (log-rank test p < 0.001), high genomic instability, and were not sensitive to conventional chemotherapeutic treatment compared to the patients in the other subtypes. Cancer cells in Cu-cluster B exhibited a higher degree of the senescence-associated secretory phenotype (SASP), a marker of cellular senescence. Three representative genes, CDCA8, MCM6, and NCAPG2, were identified in patients in Cu-cluster B using WGCNA and the "randomForest" algorithm. A nomogram was constructed to screen patients in the Cu-cluster B subtype based on three genes: CDCA8, MCM6, and NCAPG2.

CONCLUSION

Publicly available databases and various bioinformatics tools were used to study the heterogeneity of cuproptosis in patients with HCC. Three HCC subtypes were identified, with differences in the survival outcomes, genomic instability, senescence environment, and response to anticancer drugs. Further, three cuproptosis-related genes were identified, which could be used to design personalized therapeutic strategies for HCC.

Machine Learning Approach for the Prediction of In-Hospital Mortality in Traumatic Brain Injury Using Bio-Clinical Markers at Presentation to the Emergency Department

BACKGROUND

Accurate prediction of in-hospital mortality is essential for better management of patients with traumatic brain injury (TBI). Machine learning (ML) algorithms have been shown to be effective in predicting clinical outcomes. This study aimed to identify predictors of in-hospital mortality in TBI patients using ML algorithms.

MATERIALS AND METHOD

A retrospective study was performed using data from both the trauma registry and electronic medical records among TBI patients admitted to the Hamad Trauma Center in Qatar between June 2016 and May 2021. Thirteen features were selected for four ML models including a Support Vector Machine (SVM), Logistic Regression (LR), Random Forest (RF), and Extreme Gradient Boosting (XgBoost), to predict the in-hospital mortality.

RESULTS

A dataset of 922 patients was analyzed, of which 78% survived and 22% died. The AUC scores for SVM, LR, XgBoost, and RF models were 0.86, 0.84, 0.85, and 0.86, respectively. XgBoost and RF had good AUC scores but exhibited significant differences in log loss between the training and testing sets (% difference in logloss of 79.5 and 41.8, respectively), indicating overfitting compared to the other models. The feature importance trend across all models indicates that aPTT, INR, ISS, prothrombin time, and lactic acid are the most important features in prediction. Magnesium also displayed significant importance in the prediction of mortality among serum electrolytes.

CONCLUSIONS

SVM was found to be the best-performing ML model in predicting the mortality of TBI patients. It had the highest AUC score and did not show overfitting, making it a more reliable model compared to LR, XgBoost, and RF.

Molecular characterization of colorectal adenoma and colorectal cancer via integrated genomic transcriptomic analysis

Introduction

Colorectal adenoma can develop into colorectal cancer. Determining the risk of tumorigenesis in colorectal adenoma would be critical for avoiding the development of colorectal cancer; however, genomic features that could help predict the risk of tumorigenesis remain uncertain.

Methods

In this work, DNA and RNA parallel capture sequencing data covering 519 genes from colorectal adenoma and colorectal cancer samples were collected. The somatic mutation profiles were obtained from DNA sequencing data, and the expression profiles were obtained from RNA sequencing data.

Results

Despite some similarities between the adenoma samples and the cancer samples, different mutation frequencies, co-occurrences, and mutually exclusive patterns were detected in the mutation profiles of patients with colorectal adenoma and colorectal cancer. Differentially expressed genes were also detected between the two patient groups using RNA sequencing. Finally, two random forest classification models were built, one based on mutation profiles and one based on expression profiles. The models distinguished adenoma and cancer samples with accuracy levels of 81.48% and 100.00%, respectively, showing the potential of the 519-gene panel for monitoring adenoma patients in clinical practice.

Conclusion

This study revealed molecular characteristics and correlations between colorectal adenoma and colorectal cancer, and it demonstrated that the 519-gene panel may be used for early monitoring of the progression of colorectal adenoma to cancer.

Interpretable prediction of cardiopulmonary complications after non-small cell lung cancer surgery based on machine learning and SHapley additive exPlanations

Introduction

Lung cancer is a prevalent malignancy globally, with approximately 20% of patients developing cardiopulmonary complications after lobectomy. In order to prevent complications, an accurate and personalized method based on machine learning (ML) is required.

Methods

During the period of 2017-2021, a retrospective analysis was conducted on the medical records of patients who had undergone lobectomy for non-small cell lung cancer (NSCLC). We performed logical regression, decision tree (DT), random forest (RF), gradient boost DT, and eXtreme gradient boosting analyses to establish an ML model. The ten-fold cross-validation was used to evaluate the performance of multiple ML models based on various evaluation metrics, including accuracy, precision, recall, F1 score, and area under the receiver operating (AUC). Additionally, we also calculated the Kappa value of these model. Each model used grid search to optimize hyper-parameters and then used the interpretability method to provide explanations for the model's Decisions.

Results

The study included 718 eligible patients, among whom the incidence of postoperative cardiopulmonary complications was 20.89%. The RF model showed the best comprehensive performance among all models, and its ten-fold cross-validation accuracy, precision, recall, F1 score, and AUC were (OR and 95% confidence interval [CI]) 0.786 (0.738-0.834), 0.803 (0.735-0.872), 0.738 (0.678-0.797), 0.766 (0.714-0.818), 0.856 (0.815-0.898), respectively. The kappa value of the RF model was 0.696 (0.617-0.768). The SHAP method showed that gender, age, and intraoperative blood loss were closely associated with postoperative cardiopulmonary complications.

Conclusion

The application of ML methods for predicting postoperative cardiopulmonary complications based on clinical data of patients with NSCLC showed a good performance. The results indicate that ML combined with the SHAP individualized interpretation method has practical clinical value.

Machine learning and network analysis for diagnosis and prediction in disorders of consciousness

BACKGROUND

Prolonged Disorders of Consciousness (PDOC) resulting from severe acquired brain injury can lead to complex disabilities that make diagnosis challenging. The role of machine learning (ML) in diagnosing PDOC states and identifying intervention strategies is relatively under-explored, having focused on predicting mortality and poor outcome. This study aims to: (a) apply ML techniques to predict PDOC diagnostic states from variables obtained from two non-invasive neurobehavior assessment tools; and (b) apply network analysis for guiding possible intervention strategies.

METHODS

The Coma Recovery Scale-Revised (CRS-R) is a well-established tool for assessing patients with PDOC. More recently, music has been found to be a useful medium for assessment of coma patients, leading to the standardization of a music-based assessment of awareness: Music Therapy Assessment Tool for Awareness in Disorders of Consciousness (MATADOC). CRS-R and MATADOC data were collected from 74 PDOC patients aged 16-70 years at three specialist centers in the USA, UK and Ireland. The data were analyzed by three ML techniques (neural networks, decision trees and cluster analysis) as well as modelled through system-level network analysis.

RESULTS

PDOC diagnostic state can be predicted to a relatively high level of accuracy that sets a benchmark for future ML analysis using neurobehavioral data only. The outcomes of this study may also have implications for understanding the role of music therapy in interdisciplinary rehabilitation to help patients move from one coma state to another.

CONCLUSIONS

This study has shown how ML can derive rules for diagnosis of PDOC with data from two neurobehavioral tools without the need to harvest large clinical and imaging datasets. Network analysis using the measures obtained from these two non-invasive tools provides novel, system-level ways of interpreting possible transitions between PDOC states, leading to possible use in novel, next-generation decision-support systems for PDOC.

Use of random forest machine learning algorithm to predict short term outcomes following posterior cervical decompression with instrumented fusion

Random Forest (RF) is a widely used machine learning algorithm that can be utilized for identification of patient characteristics important for outcome prediction. Posterior cervical decompression with instrumented fusion (PCDF) is a procedure for the management of cervical spondylosis, cervical spinal stenosis, and degenerative disorders that can cause cervical myelopathy or radiculopathy. An RF algorithm was employed to predict and describe length of stay (LOS), readmission, reoperation, transfusion, and infection rates following elective PCDF using The American College of Surgeons National Quality Improvement Program (ACS-NSQIP) database 2008 through 2018. The RF algorithm was tasked with determining the importance of independent clinical variables in predicting our outcomes of interest and importance of each variable based on the reduction in the Gini index. Application of an RF algorithm to the ACS-NSQIP database yielded a highly predictive set of patient characteristics and perioperative events for five outcomes of interest related to elective PCDF. These variables included postoperative infection, increased age, BMI, operative time, and LOS, and decreased preoperative hematocrit and white blood cell count. Risk factors that were predictive for rate of reoperation, readmission, hospital length of stay, transfusion requirement, and post-operative infection were identified with AUC values of 0.781, 0.791, 0.781, 0.902, and 0.724 respectively. Utilization of these findings may assist in risk analysis during the perioperative period and may influence clinical or surgical decision-making.

MARCKS is a New Prognostic Biomarker in Hepatocellular Carcinoma

Background

Hepatocellular carcinoma (HCC) is one of the most common type of cancers, but there is still a lack of known biomarkers for the effective diagnosis or prognosis of HCC. Myristoylated alanine-rich C-kinase substrate (MARCKS) is a substrate of protein kinase C, which was located in the cell plasma membrane. The purpose of our study was to evaluate the role of MARCKS in HCC.

Methods

The role of MARCKS in HCC was explored by bioinformatics and experiment.

Results

We demonstrated that MARCKS expression was significantly elevated in HCC datasets of TCGA. MARCKS was up-regulated in tumor sample in HCC. Functional enrichment indicated that MARCKS-related differentially expressed genes (DEGs) were mainly enriched in cell junction tissue, response to growth factors and cell population proliferation. Tumor and ECM-receptor interactions related pathways were enriched by the KEGG. MARCKS expression in HCC patients was higher in females, younger individuals, and those at worse clinical stages. Cox regression analysis showed that MARCKS expression was a risk factor for overall survival and disease-specific survival of patients.

Conclusion

MARCKS was up-regulated in HCC, may play a crucial role in HCCs, and has prognostic value for clinical outcomes.

Efficacy Analysis of Team-Based Nursing Compliance in Young and Middle-Aged Diabetes Mellitus Patients Based on Random Forest Algorithm and Logistic Regression

Objective

Long-term hyperglycemia in young and middle-aged diabetic patients can be complicated with diabetic ketoacidosis, stroke, myocardial infarction, infection, and other complications. The objective was to explore the application value of machine learning in predicting the recurrence risk of young and middle-aged diabetes patients with team-based nursing intervention.

Methods

Clinical data of 80 patients with diabetes treated in the Department of Endocrinology from 2019 to 2020 were retrospectively collected. The data set was divided into 70% training set (n =56) and 30% test set (n =24). All the selected research cases were intervened by the team-based management mode involving family and clinical doctors and nurses. The degree of diabetes knowledge learning, the level of blood glucose changes, and the psychological state of the patients were evaluated. The random forest (RF) algorithm and logistic regression prediction model were constructed to predict the risk factors of diabetes recurrence.

Results

There was no significant difference in the degree of diabetes knowledge learning, the level of blood glucose changes, and the psychological state between the training set and the test set (P > 0.05). The FPG, HbA1c, and 2hPG of recurrence group patients were significantly higher than those of nonrecurrence group patients, and the difference was statistically significant (P < 0.05). In descending order of importance based on the RF algorithm prediction model were glucose, BMI, age, insulin, pedigree function, skin thickness, and blood diastolic pressure. The accuracy of RF and logistic regression prediction models is 81.46% and 80.21%, respectively.

Conclusion

The team-based nursing model has a good effect on the blood glucose control level of middle-aged and young diabetic patients. Age, BMI, and glucose values are risk factors for diabetes. The SF algorithm has a good effect on predicting the risk of diabetes, which is worthy of further clinical application.

Comparing human milk macronutrients measured using analyzers based on mid-infrared spectroscopy and ultrasound and the application of machine learning in data fitting

OBJECTIVE

Fat, carbohydrates (mainly lactose) and protein in breast milk all provide indispensable benefits for the growth of newborns. The only source of nutrition in early infancy is breast milk, so the energy of breast milk is also crucial to the growth of infants. Some macronutrients composition in human breast milk varies greatly, which could affect its nutritional fulfillment to preterm infant needs. Therefore, rapid analysis of macronutrients (including lactose, fat and protein) and milk energy in breast milk is of clinical importance. This study compared the macronutrients results of a mid-infrared (MIR) analyzer and an ultrasound-based breast milk analyzer and unified the results by machine learning.

METHODS

This cross-sectional study included breastfeeding mothers aged 22-40 enrolled between November 2019 and February 2021. Breast milk samples (n = 546) were collected from 244 mothers (from Day 1 to Day 1086 postpartum). A MIR milk analyzer (BETTERREN Co., HMIR-05, SH, CHINA) and an ultrasonic milk analyzer (Honɡyanɡ Co,. HMA 3000, Hebei, CHINA) were used to determine the human milk macronutrient composition. A total of 465 samples completed the tests in both analyzers. The results of the ultrasonic method were mathematically converted using machine learning, while the Bland-Altman method was used to determine the limits of agreement (LOA) between the adjusted results of the ultrasonic method and MIR results.

RESULTS

The MIR and ultrasonic milk analyzer results were significantly different. The protein, fat, and energy determined using the MIR method were higher than those determined by the ultrasonic method, while lactose determined by the MIR method were lower (all p < 0.05). The consistency between the measured MIR and the adjusted ultrasound values was evaluated using the Bland-Altman analysis and the scatter diagram was generated to calculate the 95% LOA. After adjustments, 93.96% protein points (436 out of 465), 94.41% fat points (439 out of 465), 95.91% lactose points (446 out of 465) and 94.62% energy points (440 out of 465) were within the LOA range. The 95% LOA of protein, fat, lactose and energy were - 0.6 to 0.6 g/dl, -0.92 to 0.92 g/dl, -0.88 to 0.88 g/dl and - 40.2 to 40.4 kj/dl, respectively and clinically acceptable. The adjusted ultrasonic results were consistent with the MIR results, and LOA results were high (close to 95%).

CONCLUSIONS

While the results of the breast milk rapid analyzers using the two methods varied significantly, they could still be considered comparable after data adjustments using linear regression algorithm in machine learning. Machine learning methods can play a role in data fitting using different analyzers.

Multiple Machine Learning Methods Reveal Key Biomarkers of Obstructive Sleep Apnea and Continuous Positive Airway Pressure Treatment

Obstructive sleep apnea (OSA) is a worldwide health issue that affects more than 400 million people. Given the limitations inherent in the current conventional diagnosis of OSA based on symptoms report, novel diagnostic approaches are required to complement existing techniques. Recent advances in gene sequencing technology have made it possible to identify a greater number of genes linked to OSA. We identified key genes in OSA and CPAP treatment by screening differentially expressed genes (DEGs) using the Gene Expression Omnibus (GEO) database and employing machine learning algorithms. None of these genes had previously been implicated in OSA. Moreover, a new diagnostic model of OSA was developed, and its diagnostic accuracy was verified in independent datasets. By performing Single Sample Gene Set Enrichment Analysis (ssGSEA) and Counting Relative Subsets of RNA Transcripts (CIBERSORT), we identified possible immunologic mechanisms, which led us to conclude that patients with high OSA risk tend to have elevated inflammation levels that can be brought down by CPAP treatment.

Learning Models for Traumatic Brain Injury Mortality Prediction on Pediatric Electronic Health Records

Background

Traumatic Brain Injury (TBI) is one of the leading causes of injury related mortality in the world, with severe cases reaching mortality rates of 30-40%. It is highly heterogeneous both in causes and consequences, complicating medical interpretation and prognosis. Gathering clinical, demographic, and laboratory data to perform a prognosis requires time and skill in several clinical specialties. Machine learning (ML) methods can take advantage of the data and guide physicians toward a better prognosis and, consequently, better healthcare. The objective of this study was to develop and test a wide range of machine learning models and evaluate their capability of predicting mortality of TBI, at hospital discharge, while assessing the similarity between the predictive value of the data and clinical significance.

Methods

The used dataset is the Hackathon Pediatric Traumatic Brain Injury (HPTBI) dataset, composed of electronic health records containing clinical annotations and demographic data of 300 patients. Four different classification models were tested, either with or without feature selection. For each combination of the classification model and feature selection method, the area under the receiver operator curve (ROC-AUC), balanced accuracy, precision, and recall were calculated.

Results

Methods based on decision trees perform better when using all features (Random Forest, AUC = 0.86 and XGBoost, AUC = 0.91) but other models require prior feature selection to obtain the best results (k-Nearest Neighbors, AUC = 0.90 and Artificial Neural Networks, AUC = 0.84). Additionally, Random Forest and XGBoost allow assessing the feature's importance, which could give insights for future strategies on the clinical routine.

Conclusion

Predictive capability depends greatly on the combination of model and feature selection methods used but, overall, ML models showed a very good performance in mortality prediction for TBI. The feature importance results indicate that predictive value is not directly related to clinical significance.

XGBoost machine learning algorism performed better than regression models in predicting mortality of moderate to severe traumatic brain injury

BACKGROUND

Traumatic brain injury (TBI) brings severe mortality and morbidity risk to patients. Predicting outcome of these patients is necessary for physicians to make suitable treatments to improve prognosis. The aim of this study is to develop a mortality prediction approach using the XGBoost (extreme gradient boosting) in moderate to severe TBI.

METHODS

368 patients hospitalized in West China hospital for TBI with GCS below 13 were identified. To construct XGBoost prediction approach, patients were divided into training set and test set with ratio of 7:3. Logistic regression prediction model was also constructed and compared with XGBoost model. Area under the receiver operating characteristic curve (AUC), accuracy, sensitivity and specificity were calculated to compare the prognostic value between XGBoost and logistic regression.

RESULTS

205 patients suffered poor outcome with mortality of 55.7%. Non-survivors had lower Glasgow Coma Scale (GCS) (5 vs 7, p<0.001) and higher Injury Severit Score (ISS) than survivors (25 vs 16, p<0.001). Platelet (p<0.001), albumin (p<0.001), hemoglobin (p<0.001) were significantly lower in non-survivors while glucose (p<0.001) and prothrombin time (PT) (p<0.001)was significantly higher in non-survivors. Among the XGBoost approach, GCS, PT and glucose had the most significant feature importance. The AUC (0.955 vs 0.805) and accuracy (0.955 vs 0.70) of XGBoost were both higher than logistic regression.

CONCLUSION

Predicting mortality of moderate to severe TBI patients using XGBoost algorism is more effective and precise than logistic regression. The XGBoost prediction approach is beneficial for physicians to evaluate TBI patients at high risk of poor outcome.

A machine learning approach for identification of gastrointestinal predictors for the risk of COVID-19 related hospitalization

Background and aim

COVID-19 can be presented with various gastrointestinal symptoms. Shortly after the pandemic outbreak, several machine learning algorithms were implemented to assess new diagnostic and therapeutic methods for this disease. The aim of this study is to assess gastrointestinal and liver-related predictive factors for SARS-CoV-2 associated risk of hospitalization.

Methods

Data collection was based on a questionnaire from the COVID-19 outpatient test center and from the emergency department at the University Hospital in combination with the data from internal hospital information system and from a mobile application used for telemedicine follow-up of patients. For statistical analysis SARS-CoV-2 negative patients were considered as controls in three different SARS-CoV-2 positive patient groups (divided based on severity of the disease). The data were visualized and analyzed in R version 4.0.5. The Chi-squared or Fisher test was applied to test the null hypothesis of independence between the factors followed, where appropriate, by the multiple comparisons with the Benjamini Hochberg adjustment. The null hypothesis of the equality of the population medians of a continuous variable was tested by the Kruskal Wallis test, followed by the Dunn multiple comparisons test. In order to assess predictive power of the gastrointestinal parameters and other measured variables for predicting an outcome of the patient group the Random Forest machine learning algorithm was trained on the data. The predictive ability was quantified by the ROC curve, constructed from the Out-of-Bag data. Matthews correlation coefficient was used as a one-number summary of the quality of binary classification. The importance of the predictors was measured using the Variable Importance. A 2D representation of the data was obtained by means of Principal Component Analysis for mixed type of data. Findings with the p-value below 0.05 were considered statistically significant.

Results

A total of 710 patients were enrolled in the study. The presence of diarrhea and nausea was significantly higher in the emergency department group than in the COVID-19 outpatient test center. Among liver enzymes only aspartate transaminase (AST) has been significantly elevated in the hospitalized group compared to patients discharged home. Based on the Random Forest algorithm, AST has been identified as the most important predictor followed by age or diabetes mellitus. Diarrhea and bloating have also predictive importance, although much lower than AST.

Conclusion

SARS-CoV-2 positivity is connected with isolated AST elevation and the level is linked with the severity of the disease. Furthermore, using the machine learning Random Forest algorithm, we have identified the elevated AST as the most important predictor for COVID-19 related hospitalizations.

Predicting Outcome of Traumatic Brain Injury: Is Machine Learning the Best Way?

One of the main challenges in traumatic brain injury (TBI) patients is to achieve an early and definite prognosis. Despite the recent development of algorithms based on artificial intelligence for the identification of these prognostic factors relevant for clinical practice, the literature lacks a rigorous comparison among classical regression and machine learning (ML) models. This study aims at providing this comparison on a sample of TBI patients evaluated at baseline (T0), after 3 months from the event (T1), and at discharge (T2). A Classical Linear Regression Model (LM) was compared with independent performances of Support Vector Machine (SVM), k-Nearest Neighbors (k-NN), Naïve Bayes (NB) and Decision Tree (DT) algorithms, together with an ensemble ML approach. The accuracy was similar among LM and ML algorithms on the analyzed sample when two classes of outcome (Positive vs. Negative) approach was used, whereas the NB algorithm showed the worst performance. This study highlights the utility of comparing traditional regression modeling to ML, particularly when using a small number of reliable predictor variables after TBI. The dataset of clinical data used to train ML algorithms will be publicly available to other researchers for future comparisons.

MRI Radiomics Features From Infarction and Cerebrospinal Fluid for Prediction of Cerebral Edema After Acute Ischemic Stroke

Neuroimaging biomarkers that predict the edema after acute stroke may help clinicians provide targeted therapies and minimize the risk of secondary injury. In this study, we applied pretherapy MRI radiomics features from infarction and cerebrospinal fluid (CSF) to predict edema after acute ischemic stroke. MRI data were obtained from a prospective, endovascular thrombectomy (EVT) cohort that included 389 patients with acute stroke from two centers (dataset 1, n = 292; dataset 2, n = 97), respectively. Patients were divided into edema group (brain swelling and midline shift) and non-edema group according to CT within 36 h after therapy. We extracted the imaging features of infarct area on diffusion weighted imaging (DWI) (abbreviated as DWI), CSF on fluid-attenuated inversion recovery (FLAIR) (CSF_FLAIR) and CSF on DWI (CSF_DWI), and selected the optimum features associated with edema for developing models in two forms of feature sets (DWI + CSF_FLAIR and DWI + CSF_DWI) respectively. We developed seven ML models based on dataset 1 and identified the most stable model. External validations (dataset 2) of the developed stable model were performed. Prediction model performance was assessed using the area under the receiver operating characteristic curve (AUC). The Bayes model based on DWI + CSF_FLAIR and the RF model based on DWI + CSF_DWI had the best performances (DWI + CSF_FLAIR: AUC, 0.86; accuracy, 0.85; recall, 0.88; DWI + CSF_DWI: AUC, 0.86; accuracy, 0.84; recall, 0.84) and the most stability (RSD% in DWI + CSF_FLAIR AUC: 0.07, RSD% in DWI + CSF_DWI AUC: 0.09), respectively. External validation showed that the AUC of the Bayes model based on DWI + CSF_FLAIR was 0.84 with accuracy of 0.77 and area under precision-recall curve (auPRC) of 0.75, and the AUC of the RF model based on DWI + CSF_DWI was 0.83 with accuracy of 0.81 and the auPRC of 0.76. The MRI radiomics features from infarction and CSF may offer an effective imaging biomarker for predicting edema.

Metabolomic Profiling Reveals That 5-Hydroxylysine and 1-Methylnicotinamide Are Metabolic Indicators of Keloid Severity

Background: Keloid is a skin fibroproliferative disease with unknown pathogenesis. Metabolomics provides a new perspective for revealing biomarkers related to metabolites and their metabolic mechanisms. Method: Metabolomics and transcriptomics were used for data analysis. Quality control of the data was performed to standardize the data. Principal component analysis (PCA), PLS-DA, OPLS-DA, univariate analysis, CIBERSORT, neural network model, and machine learning correlation analysis were used to calculate differential metabolites. The molecular mechanisms of characteristic metabolites and differentially expressed genes were identified through enrichment analysis and topological analysis. Result: Compared with normal tissue, lipids have a tendency to decrease in keloids, while peptides have a tendency to increase in keloids. Significantly different metabolites between the two groups were identified by random forest analysis, including 1-methylnicotinamide, 4-hydroxyproline, 5-hydroxylysine, and l-prolinamide. The metabolic pathways which play important roles in the pathogenesis of keloids included arachidonic acid metabolism and d-arginine and d-ornithine metabolism. Metabolomic profiling reveals that 5-hydroxylysine and 1-methylnicotinamide are metabolic indicators of keloid severity. The high-risk early warning index for 5-hydroxylysine is 4 × 10⁸-6.3×10⁸ (p = 0.0008), and the high-risk predictive index for 1-methylnicotinamide is 0.95 × 10⁷-1.6×10⁷ (p = 0.0022). Conclusion: This study was the first to reveal the metabolome profile and transcriptome of keloids. Differential metabolites and metabolic pathways were calculated by machine learning. Metabolomic profiling reveals that 5-hydroxylysine and 1-methylnicotinamide may be metabolic indicators of keloid severity.

Different Routes or Methods of Application for Dimensionality Reduction in Multicenter Studies Databases

Technological progress and digital transformation, which began with Big Data and Artificial Intelligence (AI), are currently transforming ways of working in all fields, to support decision-making, particularly in multicenter research. This study analyzed a sample of 5178 hospital patients, suffering from exacerbation of chronic obstructive pulmonary disease (eCOPD). Because of differences in disease stages and progression, the clinical pathologies and characteristics of the patients were extremely diverse. Our objective was thus to reduce dimensionality by projecting the data onto a lower dimensional subspace. The results obtained show that principal component analysis (PCA) is the most effective linear technique for dimensionality reduction. Four patient profile groups are generated with similar affinity and characteristics. In conclusion, dimensionality reduction is found to be an effective technique that permits the visualization of early indications of clinical patterns with similar characteristics. This is valuable since the development of other pathologies (chronic diseases) over any given time period influences clinical parameters. If healthcare professionals can have access to such information beforehand, this can significantly improve the quality of patient care, since this type of study is based on a multitude of data-variables that can be used to evaluate and monitor the clinical status of the patient.

Artificial intelligence, machine learning, and deep learning for clinical outcome prediction

AI is a broad concept, grouping initiatives that use a computer to perform tasks that would usually require a human to complete. AI methods are well suited to predict clinical outcomes. In practice, AI methods can be thought of as functions that learn the outcomes accompanying standardized input data to produce accurate outcome predictions when trialed with new data. Current methods for cleaning, creating, accessing, extracting, augmenting, and representing data for training AI clinical prediction models are well defined. The use of AI to predict clinical outcomes is a dynamic and rapidly evolving arena, with new methods and applications emerging. Extraction or accession of electronic health care records and combining these with patient genetic data is an area of present attention, with tremendous potential for future growth. Machine learning approaches, including decision tree methods of Random Forest and XGBoost, and deep learning techniques including deep multi-layer and recurrent neural networks, afford unique capabilities to accurately create predictions from high dimensional, multimodal data. Furthermore, AI methods are increasing our ability to accurately predict clinical outcomes that previously were difficult to model, including time-dependent and multi-class outcomes. Barriers to robust AI-based clinical outcome model deployment include changing AI product development interfaces, the specificity of regulation requirements, and limitations in ensuring model interpretability, generalizability, and adaptability over time.