scholarly journals Computed-Tomography-Based Radiomics Model for Predicting the Malignant Potential of Gastrointestinal Stromal Tumors Preoperatively: A Multi-Classifier and Multicenter Study

2021 ◽  
Vol 11 ◽  
Author(s):  
Minhong Wang ◽  
Zhan Feng ◽  
Lixiang Zhou ◽  
Liang Zhang ◽  
Xiaojun Hao ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Random Forest ◽  
Risk Stratification ◽  
Gastrointestinal Stromal Tumors ◽  
Validation Cohort ◽  
External Validation ◽  
Random Forest Model ◽  
Training Cohort ◽  
Stromal Tumors ◽  
Forest Model

Background: Our goal was to establish and verify a radiomics risk grading model for gastrointestinal stromal tumors (GISTs) and to identify the optimal algorithm for risk stratification.Methods: We conducted a retrospective analysis of 324 patients with GISTs, the presence of which was confirmed by surgical pathology. Patients were treated at three different hospitals. A training cohort of 180 patients was collected from the largest center, while an external validation cohort of 144 patients was collected from the other two centers. To extract radiomics features, regions of interest (ROIs) were outlined layer by layer along the edge of the tumor contour on CT images of the arterial and portal venous phases. The dimensionality of radiomic features was reduced, and the top 10 features with importance value above 5 were selected before modeling. The training cohort used three classifiers [logistic regression, support vector machine (SVM), and random forest] to establish three GIST risk stratification prediction models. The receiver operating characteristic curve (ROC) was used to compare model performance, which was validated by external data.Results: In the training cohort, the average area under the curve (AUC) was 0.84 ± 0.07 of the logistic regression, 0.88 ± 0.06 of the random forest, and 0.81 ± 0.08 of the SVM. In the external validation cohort, the AUC was 0.85 of the logistic regression, 0.90 of the random forest, and 0.80 of the SVM. The random forest model performed the best in both the training and the external validation cohorts and could be generalized.Conclusion: Based on CT radiomics, there are multiple machine-learning models that can predict the risk of GISTs. Among them, the random forest algorithm had the highest prediction efficiency and could be readily generalizable. Through external validation data, we assume that the random forest model may be used as an effective tool to guide preoperative clinical decision-making.

scholarly journals CT Radiomics Model for Discriminating the Risk Stratification of Gastrointestinal Stromal Tumors: A Multi-Class Classification and Multi-Center Study

2021 ◽  
Vol 11 ◽  
Author(s):  
Zhonghua Chen ◽  
Linyi Xu ◽  
Chuanmin Zhang ◽  
Chencui Huang ◽  
Minhong Wang ◽  
...  
Keyword(s):  
Risk Stratification ◽  
Prediction Model ◽  
Validation Cohort ◽  
External Validation ◽  
Low Risk ◽  
Class Prediction ◽  
Training Cohort ◽  
Stromal Tumors ◽  
Risk Patients ◽  
Multi Class Classification

ObjectiveTo establish and verify a computed tomography (CT)-based multi-class prediction model for discriminating the risk stratification of gastrointestinal stromal tumors (GISTs).Materials and MethodsA total of 381 patients with GISTs were confirmed by surgery and pathology. Information on 213 patients were obtained from one hospital and used as training cohort, whereas the details of 168 patients were collected from two other hospitals and used as independent validation cohort. Regions of interest on CT images of arterial and venous phases were drawn, radiomics features were extracted, and dimensionality reduction processing was performed. Using a one-vs-rest method, a Random Forest-based GISTs risk three-class prediction model was established, and the receiver operating characteristic curve (ROC) was used to evaluate the performance of the multi-class classification model, and the generalization ability was verified using external data.ResultsThe training cohort included 96 very low-risk and low-risk, 60 intermediate-risk and 57 high-risk patients. External validation cohort included 82 very low-risk and low-risk, 48 intermediate-risk and 38 high-risk patients. The GISTs risk three-class radiomics model had a macro/micro average area under the curve (AUC) of 0.84 and an accuracy of 0.78 in the training cohort. It had a stable performance in the external validation cohort, with a macro/micro average AUC of 0.83 and an accuracy of 0.80.ConclusionCT radiomics can discriminate GISTs risk stratification. The performance of the three-class radiomics prediction model is good, and its generalization ability has also been verified in the external validation cohort, indicating its potential to assist stratified and accurate treatment of GISTs in the clinic.

scholarly journals Establishment and Verification of Synchronous Metastatic Nomogram for Gastrointestinal Stromal Tumors (GISTs): A Population-Based Analysis

2020 ◽  
Vol 2020 ◽  
pp. 1-7 ◽  
Author(s):  
Yuqiang Li ◽  
Guangfeng Zhang ◽  
Xiangping Song ◽  
Lilan Zhao ◽  
Cenap Güngör ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Gastrointestinal Stromal Tumors ◽  
Logistic Regression Model ◽  
Validation Cohort ◽  
Tumor Location ◽  
Population Based ◽  
Brier Score ◽  
Validation Group ◽  
Synchronous Metastasis ◽  
Stromal Tumors

Aim. Assess the risk of synchronous metastasis and establish a nomogram in patients with GISTs. Methods. Surveillance, Epidemiology and End Results database (2004-2014) was accessed. With the logistic regression model as the basis, a nomogram was constructed. Results. 7,256 target patients were contained in our study. The nomogram discrimination for mGIST prediction revealed that tumor size contributed most to synchronous metastasis, followed by lymph nodes, extension, pathologic grade, tumor location, and mitotic count. C-index values of predictions were 0.821 (95% CI, 0.805-0.836) and 0.815 (95% CI, 0.800-0.831), and Brier score were 0.109 and 0.112 in training and validation group, respectively. The value of area under the ROCs were 0.813 (p<0.001) in the primary cohort and 0.819 (p<0.001) in the validation cohort. Through the calibration curves (as seen in the figures), nomogram prediction proved to have excellent agreement with actual metastatic diseases. Conclusion. A new nomogram was created that can evaluate synchronous metastatic diseases in patients with GISTs.

scholarly journals An Externally Validated Dynamic Nomogram for Predicting Unfavorable Prognosis in Patients With Aneurysmal Subarachnoid Hemorrhage

2021 ◽  
Vol 12 ◽  
Author(s):  
Ping Hu ◽  
Yang Xu ◽  
Yangfan Liu ◽  
Yuntao Li ◽  
Liguo Ye ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Prognostic Factors ◽  
Subarachnoid Hemorrhage ◽  
Validation Cohort ◽  
Aneurysmal Subarachnoid Hemorrhage ◽  
External Validation ◽  
Brier Score ◽  
Web Based ◽  
Training Cohort ◽  
Multivariable Logistic Regression

Background: Aneurysmal subarachnoid hemorrhage (aSAH) leads to severe disability and functional dependence. However, no reliable method exists to predict the clinical prognosis after aSAH. Thus, this study aimed to develop a web-based dynamic nomogram to precisely evaluate the risk of poor outcomes in patients with aSAH.Methods: Clinical patient data were retrospectively analyzed at two medical centers. One center with 126 patients was used to develop the model. Least absolute shrinkage and selection operator (LASSO) analysis was used to select the optimal variables. Multivariable logistic regression was applied to identify independent prognostic factors and construct a nomogram based on the selected variables. The C-index and Hosmer–Lemeshow p-value and Brier score was used to reflect the discrimination and calibration capacities of the model. Receiver operating characteristic curve and calibration curve (1,000 bootstrap resamples) were generated for internal validation, while another center with 84 patients was used to validate the model externally. Decision curve analysis (DCA) and clinical impact curves (CICs) were used to evaluate the clinical usefulness of the nomogram.Results: Unfavorable prognosis was observed in 46 (37%) patients in the training cohort and 24 (29%) patients in the external validation cohort. The independent prognostic factors of the nomogram, including neutrophil-to-lymphocyte ratio (NLR) (p = 0.005), World Federation of Neurosurgical Societies (WFNS) grade (p = 0.002), and delayed cerebral ischemia (DCI) (p = 0.0003), were identified using LASSO and multivariable logistic regression. A dynamic nomogram (https://hu-ping.shinyapps.io/DynNomapp/) was developed. The nomogram model demonstrated excellent discrimination, with a bias-corrected C-index of 0.85, and calibration capacities (Hosmer–Lemeshow p-value, 0.412; Brier score, 0.12) in the training cohort. Application of the model to the external validation cohort yielded a C-index of 0.84 and a Brier score of 0.13. Both DCA and CIC showed a superior overall net benefit over the entire range of threshold probabilities.Conclusion: This study identified that NLR on admission, WFNS grade, and DCI independently predicted unfavorable prognosis in patients with aSAH. These factors were used to develop a web-based dynamic nomogram application to calculate the precise probability of a poor patient outcome. This tool will benefit personalized treatment and patient management and help neurosurgeons make better clinical decisions.

P–016 Using artificial intelligence to predict semen upgrading after microsurgical varicocele repair

2021 ◽  
Vol 36 (Supplement_1) ◽  
Author(s):  
J Ory ◽  
M Tradewell ◽  
T Lima ◽  
U Blankstein ◽  
V Madhusoodanan ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Random Forest ◽  
External Validation ◽  
Sperm Concentration ◽  
Random Forest Model ◽  
Forest Model ◽  
Intelligence Models ◽  
Artificial Intelligence Models ◽  
Clinically Significant ◽  
Varicocele Repair

Abstract Study question Can we use artificial intelligence models to predict semen upgrading after microsurgical varicocele repair? Summary answer A machine learning model performed well in predicting clinically meaningful post-varicocelectomy semen upgrade using pre-operative hormonal, clinical, and semen analysis data. What is known already Varicocele repair is recommended in the presence of a clinical varicocele together with at least one abnormal semen parameter, and male infertility. Unfortunately, up to 50% of men who meet criteria for repair will not see meaningful benefit in outcomes despite successful surgery. Nomograms exist to help predict success, but these are based out of single-center databases, do not incorporate hormonal data, and are rarely designed to predict pre-defined, clinically meaningful improvements in semen parameters. Study design, size, duration Data were collected from an international, multi-center retrospective cohort. A total of 240 men were identified. Data from 160 men from Miami, USA and 80 men from Toronto, Canada were included. Data was collected from 2006 to 2020. Participants/materials, setting, methods We collected pre and postoperative clinical data following varicocele surgery. Clinical upgrading was defined as an increase in sperm concentration that would allow a couple to access new reproductive technologies/techniques. The tiers used for upgrading were 0–1million/cc (Intracytoplasmic Sperm Injection), 1–5 million (In Vitro Fertilization), 5–15 million (Intrauterine Insemination), and &gt;15 million (Natural conception). Artificial intelligence models were trained and tested using R to predict which patients upgraded after surgery. Main results and the role of chance 51% of men underwent bilateral varicocele repair. The majority of men had grade 2 varicocele on the left, and (when present) a grade 1 varicocele on the right. Overall, 47% of men experienced an upgrade following varicocele surgery, 47% did not change, and 6% downgraded. The data from Miami were used to create a random forest model for predicting clinically significant upgrade in sperm concentration. The most informative model parameters were preoperative FSH, sperm concentration, and surgical laterality. The model identified three clinical categories: men with unfavorable, intermediate, and favorable features to predict varicocele upgrade. On external validation using data from Toronto, the model accurately predicted upgrade in 87% of men with favorable features, and in 49% and 36% of men with intermediate and unfavorable features, respectively. Overall, the model performed well on external validation with an AUC of 0.72 and good calibration. Calibration plots, using cross-validation, define how well the predicted probabilities match the actual probability of sperm concentration upgrade. The random forest model was run twelve times. All model characteristics are the mean of ten model runs with the highest and lowest performing runs removed. The model was translated to an online calculator that can be used by clinicians. Limitations, reasons for caution One limitation to our study is that we were not able to predict total motile sperm count (TMSC), which has been shown to perform slightly better than concentration at predicting assisted reproduction outcomes. By focusing on clinically significant upgrading, this difference should be minimized. Wider implications of the findings: Predicting the chances of clinically significant semen upgrading after varicocele repair is essential for patients and clinicians to understand. Several men undergo surgery with no subsequent benefit, which may lead to a delay in definitive treatment with IVF/IUI. Understanding their chances will help couples make better informed decisions moving forward. Trial registration number Not applicable

scholarly journals Preoperative CT-Based Deep Learning Model for Predicting Risk Stratification in Patients With Gastrointestinal Stromal Tumors

2021 ◽  
Vol 11 ◽  
Author(s):  
Bing Kang ◽  
Xianshun Yuan ◽  
Hexiang Wang ◽  
Songnan Qin ◽  
Xuelin Song ◽  
...  
Keyword(s):  
Deep Learning ◽  
Risk Stratification ◽  
Gastrointestinal Stromal Tumors ◽  
External Validation ◽  
Learning Model ◽  
Ct Images ◽  
Stromal Tumors ◽  
Contrast Enhanced Ct ◽  
Malignant Gists ◽  
Deep Learning Model

ObjectiveTo develop and evaluate a deep learning model (DLM) for predicting the risk stratification of gastrointestinal stromal tumors (GISTs).MethodsPreoperative contrast-enhanced CT images of 733 patients with GISTs were retrospectively obtained from two centers between January 2011 and June 2020. The datasets were split into training (n = 241), testing (n = 104), and external validation cohorts (n = 388). A DLM for predicting the risk stratification of GISTs was developed using a convolutional neural network and evaluated in the testing and external validation cohorts. The performance of the DLM was compared with that of radiomics model by using the area under the receiver operating characteristic curves (AUROCs) and the Obuchowski index. The attention area of the DLM was visualized as a heatmap by gradient-weighted class activation mapping.ResultsIn the testing cohort, the DLM had AUROCs of 0.90 (95% confidence interval [CI]: 0.84, 0.96), 0.80 (95% CI: 0.72, 0.88), and 0.89 (95% CI: 0.83, 0.95) for low-malignant, intermediate-malignant, and high-malignant GISTs, respectively. In the external validation cohort, the AUROCs of the DLM were 0.87 (95% CI: 0.83, 0.91), 0.64 (95% CI: 0.60, 0.68), and 0.85 (95% CI: 0.81, 0.89) for low-malignant, intermediate-malignant, and high-malignant GISTs, respectively. The DLM (Obuchowski index: training, 0.84; external validation, 0.79) outperformed the radiomics model (Obuchowski index: training, 0.77; external validation, 0.77) for predicting risk stratification of GISTs. The relevant subregions were successfully highlighted with attention heatmap on the CT images for further clinical review.ConclusionThe DLM showed good performance for predicting the risk stratification of GISTs using CT images and achieved better performance than that of radiomics model.

scholarly journals Machine Learning Approach Using Routine Immediate Postoperative Laboratory Values for Predicting Postoperative Mortality

2021 ◽  
Vol 11 (12) ◽  
pp. 1271
Author(s):  
Jaehyeong Cho ◽  
Jimyung Park ◽  
Eugene Jeong ◽  
Jihye Shin ◽  
Sangjeong Ahn ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Prediction Models ◽  
External Validation ◽  
Model Development ◽  
Postoperative Mortality ◽  
Random Forest Model ◽  
Forest Model ◽  
Laboratory Values ◽  
Increased Risk

Background: Several prediction models have been proposed for preoperative risk stratification for mortality. However, few studies have investigated postoperative risk factors, which have a significant influence on survival after surgery. This study aimed to develop prediction models using routine immediate postoperative laboratory values for predicting postoperative mortality. Methods: Two tertiary hospital databases were used in this research: one for model development and another for external validation of the resulting models. The following algorithms were utilized for model development: LASSO logistic regression, random forest, deep neural network, and XGBoost. We built the models on the lab values from immediate postoperative blood tests and compared them with the SASA scoring system to demonstrate their efficacy. Results: There were 3817 patients who had immediate postoperative blood test values. All models trained on immediate postoperative lab values outperformed the SASA model. Furthermore, the developed random forest model had the best AUROC of 0.82 and AUPRC of 0.13, and the phosphorus level contributed the most to the random forest model. Conclusions: Machine learning models trained on routine immediate postoperative laboratory values outperformed previously published approaches in predicting 30-day postoperative mortality, indicating that they may be beneficial in identifying patients at increased risk of postoperative death.

P1511MODELO PREDICTIVO DE SUPERVIVENCIA DE INTELIGENCIA ARTIFICIAL (BOSQUE ALEATORIO) EN HEMODIALISIS. DATOS DEL REGISTRO ANDALUSINA DE ENFERMEDADES RENALES. SICATA

2020 ◽  
Vol 35 (Supplement_3) ◽  
Author(s):  
Manuel Benítez Sánchez ◽  
Guillermo Martín ◽  
Luis Gil Sacaluga ◽  
Maria Jose Garcia Cortes ◽  
Sergio García Marcos ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Random Forest ◽  
Roc Curve ◽  
Explanatory Models ◽  
Multivariate Model ◽  
Random Forest Model ◽  
Predictor Variables ◽  
Multivariate Logistic Regression ◽  
Analytical Technique ◽  
Forest Model

Abstract Background and Aims Random Forest (RF) is an analytical technique of Artificial Intelligence (AI) that consists of an assembly of trees built by bootstrapping (resampling with replacement). In each node a subset of predictor variables is selected and for them the best cut point is determined. Each division of the tree is based on a random sample of the predictors. The trees are as long as possible. In the construction of each RF tree a part of the observations is not used (37% approx.). It is called an out-of-bag (OOB) sample and is used to obtain an honest estimate of the predictive capacity of the model. So it does not require validation. In each analysis, a few hundred Regression or classification trees are carried out, depending on whether the response variable is numerical or qualitative respectively. The result is an average of the repeated predictions of the model (Bagging). RF allows to calculate the importance of the predictor variables, which can be used later to be included in a multivariate regression model. Method We analyzed 14750 records between 2011 and 2014 contained in Information System of the Autonomous Transplant Coordination of Andalusia (SICATA) a system that includes clinical-epidemiological variables, about anemia, bone bone metabolism, adequacy of dialysis and vascular access. 1911 patients presented the event of interest (exitus). Three predictive and explanatory models of survival are developed: 1-RF. 2-.Multivariate Logistic Regression. 3- Multivariate Logistic Regression that includes the important variables of the previous RF model. We compare them in terms of accuracy (AUC of the ROC curve). Results AUC of the ROC curve of the multivariate model without prior RF was: 0.75 AUC of the ROC curve of the multivariate model with previous RF was: 0.81. AUC of the ROC curve of the Random Forest model: 0.98 Conclusion The Random Forest model has a 98% discrimination in the mortality of patients on Hemodialysis, far superior to the classic multivariate analyzes. The Multivariate Logistic Regression performed with the important RF variables improves the AUC of the previous model 0.81 vs. 0.75.

scholarly journals Identifying the Risk Factors Associated with Nursing Home Residents’ Pressure Ulcers Using Machine Learning Methods

2021 ◽  
Vol 18 (6) ◽  
pp. 2954
Author(s):  
Soo-Kyoung Lee ◽  
Juh Hyun Shin ◽  
Jinhyun Ahn ◽  
Ji Yeon Lee ◽  
Dong Eun Jang
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Random Forest ◽  
Pressure Ulcers ◽  
Nursing Home Residents ◽  
Random Forest Model ◽  
Support Vector ◽  
Predictive Values ◽  
Forest Model ◽  
Linear Svm

Background: Machine learning (ML) can keep improving predictions and generating automated knowledge via data-driven predictors or decisions. Objective: The purpose of this study was to compare different ML methods including random forest, logistics regression, linear support vector machine (SVM), polynomial SVM, radial SVM, and sigmoid SVM in terms of their accuracy, sensitivity, specificity, negative predictor values, and positive predictive values by validating real datasets to predict factors for pressure ulcers (PUs). Methods: We applied representative ML algorithms (random forest, logistic regression, linear SVM, polynomial SVM, radial SVM, and sigmoid SVM) to develop a prediction model (N = 60). Results: The random forest model showed the greatest accuracy (0.814), followed by logistic regression (0.782), polynomial SVM (0.779), radial SVM (0.770), linear SVM (0.767), and sigmoid SVM (0.674). Conclusions: The random forest model showed the greatest accuracy for predicting PUs in nursing homes (NHs). Diverse factors that predict PUs in NHs including NH characteristics and residents’ characteristics were identified according to diverse ML methods. These factors should be considered to decrease PUs in NH residents.

scholarly journals P122 Machine learning for prediction of intra-abdominal abscesses in patients with Crohn’s disease visiting the emergency department

2021 ◽  
Vol 15 (Supplement_1) ◽  
pp. S214-S214
Author(s):  
A Levartovsky ◽  
Y Barash ◽  
S Ben-Horin ◽  
B Ungar ◽  
E Klang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Emergency Department ◽  
Logistic Regression ◽  
Random Forest ◽  
Medical Treatment ◽  
Abdominal Imaging ◽  
Random Forest Model ◽  
Data Repository ◽  
Abdominal Abscess ◽  
Forest Model

Abstract Background Intra-abdominal abscess is an important clinical complication of Crohn’s disease (CD), which can be diagnosed using computed tomography (CT) or magnetic resonance imaging (MRI). However, a high index of clinical suspicion is needed to diagnose an abscess as abdominal imaging is not routinely used during hospital admission. This study aimed to identify clinical predictors of an intra-abdominal abscess among hospitalized patients with CD. Methods We created an electronic data repository of all patients with CD who visited the emergency department (ED) of our tertiary medical center between 2012 and 2018. Data included tabular demographic and clinical variables, as well as CT and MRI imaging outcomes. We searched the data repository for the presence of an abscess on abdominal imaging within seven days from the ED visit. Machine learning models were trained to predict the presence of an abscess. A logistic regression model was compared to a random forest model. The area under the receiver operator curve (AUC) was used as a metric. To establish statistical significance, bootstrapping of 100 experiments with random 80/20 training/testing splits was performed. We included only patients who were hospitalized due to complaints that can be attributed to CD exacerbation. Patients presenting within 30 days from an abdominal surgery were excluded. Results Overall, 1556 patients with CD visited the ED, of those 555 patients with a CD exacerbation. Of them, 339 patients were hospitalized and underwent abdominal imaging within 7 days from the ED visit. Forty-two patients (12.1%) were diagnosed with an abscess on abdominal imaging. The average length of the abscess was 32 mm (IQR 21.5, 43.5), mainly in the mesentery adjacent to the small bowel (38.1%). On multivariate analysis, high CRP values (64.97 mg/L, aOR 14.42 [95% CI 4.93–42.13]), high platelet count (322.5 K/microL, aOR 4.01 [95% CI 1.97–8.15]), leukocytosis (10.55 K/microL, aOR 3.83 [95% CI 1.71–8.56]) and higher heart rate (over 87.5 beats per minute, aOR 2.58 [95% CI 1.22–5.46]) were independently associated with an intra-abdominal abscess. Overall, random forest and logistic regression showed similar performance. The random forest model showed an AUC of 0.824±0.065 with eight features (CRP, Hemoglobin, WBC, age, current biologic medical treatment, BUN, current immunomodulatory medical treatment, gender). Conclusion In our large tertiary center cohort, the machine-learning model identified features associated with the presentation of an intra-abdominal abscess. Such a decision support tool may assist in triaging CD patients for imaging to exclude this potentially life-threatening complication.

Sign in / Sign up

Export Citation Format

Share Document