Machine learning predicts lymph node metastasis of poorly differentiated-type intramucosal gastric cancer

Low Absolute Lymphocyte Count Correlates with Lymph Node Metastases and Worse Survival of Patients with Gastric Cancer

BACKGROUND

Several studies have found that the absolute lymphocyte (ALC) or neutrophil count predicts the survival of patients with solid tumors, and that the neutrophil-to-lymphocyte ratio and the prognostic nutritional index are useful markers of gastric cancer prognosis. However, it remains unclear whether the ALC is prognostic of lymph node (LN) metastasis in patients with gastric cancer. In this study, we aimed to explore the impact of ALC on prognosis and distinctive clinical characteristics in patients with gastric cancer.

PATIENTS AND METHODS

The medical records of patients with gastric adenocarcinomas who underwent radical gastrectomy with curative intent at Seoul St. Mary's Hospital and Yeouido St. Mary's Hospital between January 2010 and December 2017 were reviewed. Of these, 4149 patients for whom preoperative white blood cell, neutrophil, and lymphocyte counts were available were enrolled.

RESULTS

In all 4149 patients, ALC gradually decreased as the pN stage increased. Those with an ALC of less than 1360 cells/μL were defined as a low-ALC group, and advanced cT and cN stages were the strongest risk factors for LN metastasis in both univariate and multivariate analyses; undifferentiated tumor histology and a low ALC were also significant risk factors. Patients of all stages in the ALC-low group exhibited poorer prognoses. The ALC-low group also exhibited a higher recurrence rate in a greater proportion of LNs.

CONCLUSIONS

In patients with gastric cancer, as the preoperative ALC decreases, the incidence of LN metastasis increases. A low ALC is associated with a high recurrence rate, particularly in LNs.

Interpretable deep learning model to predict lymph node metastasis in early gastric cancer using whole slide images

In early gastric cancer (EGC), the presence of lymph node metastasis (LNM) is a crucial factor for determining the treatment options. Endoscopic resection is used for treatment of EGC with minimal risk of LNM. However, owing to the lack of definitive criteria for identifying patients who require additional surgery, some patients undergo unnecessary additional surgery. Considering that histopathologic patterns are significant factor for predicting lymph node metastasis in gastric cancer, we aimed to develop a machine learning algorithm which can predict LNM status using hematoxylin and eosin (H&E)-stained images. The images were obtained from several institutions. Our pipeline comprised two sequential approaches including a feature extractor and a risk classifier. For the feature extractor, a segmentation network (DeepLabV3+) was trained on 243 WSIs across three datasets to differentiate each histological subtype. The risk classifier was trained with XGBoost using 70 morphological features inferred from the trained feature extractor. The trained segmentation network, the feature extractor, achieved high performance, with pixel accuracies of 0.9348 and 0.8939 for the internal and external datasets in patch level, respectively. The risk classifier achieved an overall AUC of 0.75 in predicting LNM status. Remarkably, one of the datasets also showed a promising result with an AUC of 0.92. This is the first multi-institution study to develop machine learning algorithm for predicting LNM status in patients with EGC using H&E-stained histopathology images. Our findings have the potential to improve the selection of patients who require surgery among those with EGC showing high-risk histological features.

Analysis of lymph node metastasis and survival prognosis in early gastric cancer patients: A retrospective study

BACKGROUND

Early gastric cancer (EGC) is a common malignant tumor of the digestive system, and its lymph node metastasis and survival prognosis have been concerning. By retrospectively analyzing the clinical data of EGC patients, we can better understand the status of lymph node metastasis and its impact on survival and prognosis.

AIM

To evaluate the prognosis of EGC patients and the factors that affect lymph node metastasis.

METHODS

The clinicopathological data of 1011 patients with EGC admitted to our hospital between January 2015 and December 2023 were collected in a retrospective cohort study. There were 561 males and 450 females. The mean age was 58 ± 11 years. The patient underwent radical gastrectomy. The status of lymph node metastasis in each group was determined according to the pathological examination results of surgical specimens. The outcomes were as follows: (1) Lymph node metastasis in EGC patients; (2) Analysis of influencing factors of lymph node metastasis in EGC; and (3) Analysis of prognostic factors in patients with EGC. Normally distributed measurement data are expressed as mean ± SD, and a t test was used for comparisons between groups. The data are expressed as absolute numbers or percentages, and the chi-square test was used for comparisons between groups. Rank data were compared using a nonparametric rank sum test. A log-rank test and a logistic regression model were used for univariate analysis. A logistic stepwise regression model and a Cox stepwise regression model were used for multivariate analysis. The Kaplan-Meier method was used to calculate the survival rate and construct survival curves. A log-rank test was used for survival analysis.

RESULTS

Analysis of influencing factors of lymph node metastasis in EGC. The results of the multifactor analysis showed that tumor length and diameter, tumor site, tumor invasion depth, vascular thrombus, and tumor differentiation degree were independent influencing factors for lymph node metastasis in patients with EGC (odds ratios = 1.80, 1.49, 2.65, 5.76, and 0.60; 95%CI: 1.29-2.50, 1.11-2.00, 1.81-3.88, 3.87-8.59, and 0.48-0.76, respectively; P < 0.05). Analysis of prognostic factors in patients with EGC. All 1011 patients with EGC were followed up for 43 (0-13) months. The 3-year overall survival rate was 97.32%. Multivariate analysis revealed that age > 60 years and lymph node metastasis were independent risk factors for prognosis in patients with EGC (hazard ratio = 9.50, 2.20; 95%CI: 3.31-27.29, 1.00-4.87; P < 0.05). Further analysis revealed that the 3-year overall survival rates of gastric cancer patients aged > 60 years and ≤ 60 years were 99.37% and 94.66%, respectively, and the difference was statistically significant (P < 0.05). The 3-year overall survival rates of patients with and without lymph node metastasis were 95.42% and 97.92%, respectively, and the difference was statistically significant (P < 0.05).

CONCLUSION

The lymph node metastasis rate of EGC patients was 23.64%. Tumor length, tumor site, tumor infiltration depth, vascular cancer thrombin, and tumor differentiation degree were found to be independent factors affecting lymph node metastasis in EGC patients. Age > 60 years and lymph node metastasis are independent risk factors for EGC prognosis.

Enhanced multi-class pathology lesion detection in gastric neoplasms using deep learning-based approach and validation

This study developed a new convolutional neural network model to detect and classify gastric lesions as malignant, premalignant, and benign. We used 10,181 white-light endoscopy images from 2606 patients in an 8:1:1 ratio. Lesions were categorized as early gastric cancer (EGC), advanced gastric cancer (AGC), gastric dysplasia, benign gastric ulcer (BGU), benign polyp, and benign erosion. We assessed the lesion detection and classification model using six-class, cancer versus non-cancer, and neoplasm versus non-neoplasm categories, as well as T-stage estimation in cancer lesions (T1, T2-T4). The lesion detection rate was 95.22% (219/230 patients) on a per-patient basis: 100% for EGC, 97.22% for AGC, 96.49% for dysplasia, 75.00% for BGU, 97.22% for benign polyps, and 80.49% for benign erosion. The six-class category exhibited an accuracy of 73.43%, sensitivity of 80.90%, specificity of 83.32%, positive predictive value (PPV) of 73.68%, and negative predictive value (NPV) of 88.53%. The sensitivity and NPV were 78.62% and 88.57% for the cancer versus non-cancer category, and 83.26% and 89.80% for the neoplasm versus non-neoplasm category, respectively. The T stage estimation model achieved an accuracy of 85.17%, sensitivity of 88.68%, specificity of 79.81%, PPV of 87.04%, and NPV of 82.18%. The novel CNN-based model remarkably detected and classified malignant, premalignant, and benign gastric lesions and accurately estimated gastric cancer T-stages.

Construction of machine learning models based on transrectal ultrasound combined with contrast-enhanced ultrasound to predict preoperative regional lymph node metastasis of rectal cancer

Purpose

Constructing a machine learning model based on transrectal ultrasound (TRUS) combined with contrast-enhanced ultrasound (CEUS) to predict preoperative regional lymph node metastasis (RLNM) of rectal cancer and provide new references for decision-making.

Materials and methods

233 patients with rectal cancer were enrolled and underwent TRUS and CEUS prior to surgery. Clinicopathological and ultrasound data were collected to analyze the correlation of RLNM status, clinical features and ultrasound parameters. A 75% training set and 25% test set were utilized to construct seven machine learning algorithms. The DeLong test was used to assess the model's diagnostic performance, then chose the best one to predict RLNM of rectal cancer.

Results

The diagnostic performance was most dependent on the following: MMT difference (36), length (30), location (29), AUC ratio (27), and PI ratio (24). The prediction accuracy, sensitivity, specificity, precision, and F1 score range of KNN, Bayes, MLP, LR, SVM, RF, and LightGBM were (0.553-0.857), (0.000-0.935), (0.600-1.000), (0.557-0.952), and (0.617-0.852), respectively. The LightGBM model exhibited the optimal accuracy (0.857) and F1 score (0.852). The AUC for machine learning analytics were (0.517-0.941, 95% CI: 0.380-0.986). The LightGBM model exhibited the highest AUC (0.941, 95% CI: 0.843-0.986), though no statistic significant showed in comparison with the SVM, LR, RF, and MLP models (P > 0.05), it was significantly higher than that of the KNN and Bayes models (P < 0.05).

Conclusion

The LightGBM machine learning model based on TRUS combined with CEUS may help predict RLNM prior to surgery and provide new references for clinical treatment in rectal cancer.

Machine learning-based classifiers to predict metastasis in colorectal cancer patients

Background

The increasing prevalence of colorectal cancer (CRC) in Iran over the past three decades has made it a key public health burden. This study aimed to predict metastasis in CRC patients using machine learning (ML) approaches in terms of demographic and clinical factors.

Methods

This study focuses on 1,127 CRC patients who underwent appropriate treatments at Taleghani Hospital, a tertiary care facility. The patients were divided into training and test datasets in an 80:20 ratio. Various ML methods, including Naive Bayes (NB), random rorest (RF), support vector machine (SVM), neural network (NN), decision tree (DT), and logistic regression (LR), were used for predicting metastasis in CRC patients. Model performance was evaluated using 5-fold cross-validation, reporting sensitivity, specificity, the area under the curve (AUC), and other indexes.

Results

Among the 1,127 patients, 183 (16%) had experienced metastasis. In the predictionof metastasis, both the NN and RF algorithms had the highest AUC, while SVM ranked third in both the original and balanced datasets. The NN and RF algorithms achieved the highest AUC (100%), sensitivity (100% and 100%, respectively), and accuracy (99.2% and 99.3%, respectively) on the balanced dataset, followed by the SVM with an AUC of 98.8%, a sensitivity of 97.5%, and an accuracy of 97%. Moreover, lower false negative rate (FNR), false positive rate (FPR), and higher negative predictive value (NPV) can be confirmed by these two methods. The results also showed that all methods exhibited good performance in the test datasets, and the balanced dataset improved the performance of most ML methods. The most important variables for predicting metastasis were the tumor stage, the number of involved lymph nodes, and the treatment type. In a separate analysis of patients with tumor stages I-III, it was identified that tumor grade, tumor size, and tumor stage are the most important features.

Conclusion

This study indicated that NN and RF were the best among ML-based approaches for predicting metastasis in CRC patients. Both the tumor stage and the number of involved lymph nodes were considered the most important features.

A discrimination model by machine learning to avoid gastrectomy for early gastric cancer

Aim

Gastrectomy is recommended for patients with early gastric cancer (EGC) because the possibility of lymph node metastasis (LNM) cannot be completely denied. The aim of this study was to develop a discrimination model to select patients who do not require surgery using machine learning.

Methods

Data from 382 patients who received gastrectomy for gastric cancer and who were diagnosed with pT1b were extracted for developing a discrimination model. For the validation of this discrimination model, data from 140 consecutive patients who underwent endoscopic resection followed by gastrectomy, with a diagnosis of pT1b EGC, were extracted. We applied XGBoost to develop a discrimination model for clinical and pathological variables. The performance of the discrimination model was evaluated based on the number of cases classified as true negatives for LNM, with no false negatives for LNM allowed.

Results

Lymph node metastasis was observed in 95 patients (25%) in the development cohort and 11 patients (8%) in the validation cohort. The discrimination model was developed to identify 27 (7%) patients with no indications for additional surgery due to the prediction of an LNM-negative status with no false negatives. In the validation cohort, 13 (9%) patients were identified as having no indications for additional surgery and no patients with LNM were classified into this group.

Conclusion

The discrimination model using XGBoost algorithms could select patients with no risk of LNM from patients with pT1b EGC. This discrimination model was considered promising for clinical decision-making in relation to patients with EGC.

Application of Machine Learning Based on Structured Medical Data in Gastroenterology

The era of big data has led to the necessity of artificial intelligence models to effectively handle the vast amount of clinical data available. These data have become indispensable resources for machine learning. Among the artificial intelligence models, deep learning has gained prominence and is widely used for analyzing unstructured data. Despite the recent advancement in deep learning, traditional machine learning models still hold significant potential for enhancing healthcare efficiency, especially for structured data. In the field of medicine, machine learning models have been applied to predict diagnoses and prognoses for various diseases. However, the adoption of machine learning models in gastroenterology has been relatively limited compared to traditional statistical models or deep learning approaches. This narrative review provides an overview of the current status of machine learning adoption in gastroenterology and discusses future directions. Additionally, it briefly summarizes recent advances in large language models.

Development and Validation of Models to Predict Lymph Node Metastasis in Early Gastric Cancer Using Logistic Regression and Gradient Boosting Machine Methods

PURPOSE

To identify important features of lymph node metastasis (LNM) and develop a prediction model for early gastric cancer (EGC) using a gradient boosting machine (GBM) method.

MATERIALS AND METHODS

The clinicopathologic data of 2556 patients with EGC who underwent gastrectomy were used as training set and the internal validation set (set 1) at a ratio of 8:2. Additionally, 548 patients with EGC who underwent endoscopic submucosal dissection (ESD) as the initial treatment were included in the external validation set (set 2). The GBM model was constructed, and its performance was compared with that of the Japanese guidelines.

RESULTS

LNM was identified in 12.6% (321/2556) of the gastrectomy group (training set & set 1) and 4.3% (24/548) of the ESD group (set 2). In the GBM analysis, the top five features that most affected LNM were lymphovascular invasion, depth, differentiation, size, and location. The accuracy, sensitivity, specificity, and the area under the receiver operating characteristics of set 1 were 0.566, 0.922, 0.516, and 0.867, while those of set 2 were 0.810, 0.958, 0.803, and 0.944, respectively. When the sensitivity of GBM was adjusted to that of Japanese guidelines (beyond the expanded criteria in set 1 [0.922] and eCuraC-2 in set 2 [0.958]), the specificities of GBM in sets 1 and 2 were 0.516 (95% confidence interval, 0.502-0.523) and 0.803 (0.795-0.805), while those of the Japanese guidelines were 0.502 (0.488-0.509) and 0.788 (0.780-0.790), respectively.

CONCLUSION

The GBM model showed good performance comparable with the eCura system in predicting LNM risk in EGCs.

A Nomogram Based on Nutrition-Related Indicators and Computed Tomography Imaging Features for Predicting Preoperative Lymph Node Metastasis in Curatively Resected Esophagogastric Junction Adenocarcinoma

BACKGROUNDS

Preoperative noninvasive tools to predict pretreatment lymph node metastasis (PLNM) status accurately for esophagogastric junction adenocarcinoma (EJA) are few. Thus, the authors aimed to construct a nomogram for predicting PLNM in curatively resected EJA.

METHODS

This study enrolled 638 EJA patients who received curative surgery resection and divided them randomly (7:3) into training and validation groups. For nomogram construction, 26 candidate parameters involving 21 preoperative clinical laboratory blood nutrition-related indicators, computed tomography (CT)-reported tumor size, CT-reported PLNM, gender, age, and body mass index were screened.

RESULTS

In the training group, Lasso regression included nine nutrition-related blood indicators in the PLNM-prediction nomogram. The PLNM prediction nomogram yielded an area under the receiver operating characteristic (ROC) curve of 0.741 (95 % confidence interval [CI], 0.697-0.781), which was better than that of the CT-reported PLNM (0.635; 95% CI 0.588-0.680; p < 0.0001). Application of the nomogram in the validation cohort still gave good discrimination (0.725 [95% CI 0.658-0.785] vs 0.634 [95% CI 0.563-0.700]; p = 0.0042). Good calibration and a net benefit were observed in both groups.

CONCLUSIONS

This study presented a nomogram incorporating preoperative nutrition-related blood indicators and CT imaging features that might be used as a convenient tool to facilitate the preoperative individualized prediction of PLNM for patients with curatively resected EJA.

Identification of survival-associated biomarkers based on three datasets by bioinformatics analysis in gastric cancer

BACKGROUND

Gastric cancer (GC) is one of the most common malignant tumors with poor prognosis in terms of advanced stage. However, the survival-associated biomarkers for GC remains unclear.

AIM

To investigate the potential biomarkers of the prognosis of patients with GC, so as to provide new methods and strategies for the treatment of GC.

METHODS

RNA sequencing data from The Cancer Genome Atlas (TCGA) database of STAD tumors, and microarray data from Gene Expression Omnibus (GEO) database (GSE19826, GSE79973 and GSE29998) were obtained. The differentially expressed genes (DEGs) between GC patients and health people were picked out using R software (x64 4.1.3). The intersections were underwent between the above obtained co-expression of differential genes (co-DEGs) and the DEGs of GC from Gene Expression Profiling Interactive Analysis database, and Gene Ontology (GO) analysis, Kyoto Encyclopedia of Gene and Genome (KEGG) pathway analysis, Gene Set Enrichment Analysis (GSEA), Protein-protein Interaction (PPI) analysis and Kaplan-Meier Plotter survival analysis were performed on these DEGs. Using Immunohistochemistry (IHC) database of Human Protein Atlas (HPA), we verified the candidate Hub genes.

RESULTS

With DEGs analysis, there were 334 co-DEGs, including 133 up-regulated genes and 201 down-regulated genes. GO enrichment analysis showed that the co-DEGs were involved in biological process, cell composition and molecular function pathways. KEGG enrichment analysis suggested the co-DEGs pathways were mainly enriched in ECM-receptor interaction, protein digestion and absorption pathways, etc. GSEA pathway analysis showed that co-DEGs mainly concentrated in cell cycle progression, mitotic cell cycle and cell cycle pathways, etc. PPI analysis showed 84 nodes and 654 edges for the co-DEGs. The survival analysis illustrated 11 Hub genes with notable significance for prognosis of patients were screened. Furtherly, using IHC database of HPA, we confirmed the above candidate Hub genes, and 10 Hub genes that associated with prognosis of GC were identified, namely BGN, CEP55, COL1A2, COL4A1, FZD2, MAOA, PDGFRB, SPARC, TIMP1 and VCAN.

CONCLUSION

The 10 Hub genes may be the potential biomarkers for predicting the prognosis of GC, which can provide new strategies and methods for the diagnosis and treatment of GC.

Machine-learning based investigation of prognostic indicators for oncological outcome of pancreatic ductal adenocarcinoma

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive malignancy with a poor prognosis. Surgical resection remains the only potential curative treatment option for early-stage resectable PDAC. Patients with locally advanced or micrometastatic disease should ideally undergo neoadjuvant therapy prior to surgical resection for an optimal treatment outcome. Computerized tomography (CT) scan is the most common imaging modality obtained prior to surgery. However, the ability of CT scans to assess the nodal status and resectability remains suboptimal and depends heavily on physician experience. Improved preoperative radiographic tumor staging with the prediction of postoperative margin and the lymph node status could have important implications in treatment sequencing. This paper proposes a novel machine learning predictive model, utilizing a three-dimensional convoluted neural network (3D-CNN), to reliably predict the presence of lymph node metastasis and the postoperative positive margin status based on preoperative CT scans.

Methods

A total of 881 CT scans were obtained from 110 patients with PDAC. Patients and images were separated into training and validation groups for both lymph node and margin prediction studies. Per-scan analysis and per-patient analysis (utilizing majority voting method) were performed.

Results

For a lymph node prediction 3D-CNN model, accuracy was 90% for per-patient analysis and 75% for per-scan analysis. For a postoperative margin prediction 3D-CNN model, accuracy was 81% for per-patient analysis and 76% for per-scan analysis.

Discussion

This paper provides a proof of concept that utilizing radiomics and the 3D-CNN deep learning framework may be used preoperatively to improve the prediction of positive resection margins as well as the presence of lymph node metastatic disease. Further investigations should be performed with larger cohorts to increase the generalizability of this model; however, there is a great promise in the use of convoluted neural networks to assist clinicians with treatment selection for patients with PDAC.

Machine learning for lymph node metastasis prediction of in patients with gastric cancer: A systematic review and meta-analysis

Objective

To evaluate the diagnostic performance of machine learning (ML) in predicting lymph node metastasis (LNM) in patients with gastric cancer (GC) and to identify predictors applicable to the models.

Methods

PubMed, EMBASE, Web of Science, and Cochrane Library were searched from inception to March 16, 2022. The pooled c-index and accuracy were used to assess the diagnostic accuracy. Subgroup analysis was performed based on ML types. Meta-analyses were performed using random-effect models. Risk of bias assessment was conducted using PROBAST tool.

Results

A total of 41 studies (56182 patients) were included, and 33 of the studies divided the participants into a training set and a test set, while the rest of the studies only had a training set. The c-index of ML for LNM prediction in training set and test set was 0.837 [95%CI (0.814, 0.859)] and 0.811 [95%CI (0.785-0.838)], respectively. The pooled accuracy was 0.781 [(95%CI (0.756-0.805)] in training set and 0.753 [95%CI (0.721-0.783)] in test set. Subgroup analysis for different ML algorithms and staging of GC showed no significant difference. In contrast, in the subgroup analysis for predictors, in the training set, the model that included radiomics had better accuracy than the model with only clinical predictors (F = 3.546, p = 0.037). Additionally, cancer size, depth of cancer invasion and histological differentiation were the three most commonly used features in models built for prediction.

Conclusion

ML has shown to be of excellent diagnostic performance in predicting the LNM of GC. One of the models covering radiomics and its ML algorithms showed good accuracy for the risk of LNM in GC. However, the results revealed some methodological limitations in the development process. Future studies should focus on refining and improving existing models to improve the accuracy of LNM prediction.

Systematic Review Registration

https://www.crd.york.ac.uk/PROSPERO/, identifier CRD42022320752

Comparison of Four Machine Learning Techniques for Prediction of Intensive Care Unit Length of Stay in Heart Transplantation Patients

Background

Post-operative heart transplantation patients often require admission to an intensive care unit (ICU). Early prediction of the ICU length of stay (ICU-LOS) of these patients is of great significance and can guide treatment while reducing the mortality rate among patients. However, conventional linear models have tended to perform worse than non-linear models.

Materials and Methods

We collected the clinical data of 365 patients from Wuhan Union Hospital who underwent heart transplantation surgery between April 2017 and August 2020. The patients were randomly divided into training data (N = 256) and test data (N = 109) groups. 84 clinical features were collected for each patient. Features were validated using the Least Absolute Shrinkage and Selection Operator (LASSO) regression’s fivefold cross-validation method. We obtained Shapley Additive explanations (SHAP) values by executing package “shap” to interpret model predictions. Four machine learning models and logistic regression algorithms were developed. The area under the receiver operating characteristic curve (AUC-ROC) was used to compare the prediction performance of different models. Finally, for the convenience of clinicians, an online web-server was established and can be freely accessed via the website https://wuhanunion.shinyapps.io/PredictICUStay/.

Results

In this study, 365 consecutive patients undergoing heart transplantation surgery for moderate (NYHA grade 3) or severe (NYHA grade 4) heart failure were collected in Wuhan Union Hospital from 2017 to 2020. The median age of the recipient patients was 47.2 years, while the median age of the donors was 35.58 years. 330 (90.4%) of the donor patients were men, and the average surgery duration was 260.06 min. Among this cohort, 47 (12.9%) had renal complications, 25 (6.8%) had hepatic complications, 11 (3%) had undergone chest re-exploration and 19 (5.2%) had undergone extracorporeal membrane oxygenation (ECMO). The following six important clinical features were selected using LASSO regression, and according to the result of SHAP, the rank of importance was (1) the use of extracorporeal membrane oxygenation (ECMO); (2) donor age; (3) the use of an intra-aortic balloon pump (IABP); (4) length of surgery; (5) high creatinine (Cr); and (6) the use of continuous renal replacement therapy (CRRT). The eXtreme Gradient Boosting (XGBoost) algorithm presented significantly better predictive performance (AUC-ROC = 0.88) than other models [Accuracy: 0.87; sensitivity: 0.98; specificity: 0.51; positive predictive value (PPV): 0.86; negative predictive value (NPV): 0.93].

Conclusion

Using the XGBoost classifier with heart transplantation patients can provide an accurate prediction of ICU-LOS, which will not only improve the accuracy of clinical decision-making but also contribute to the allocation and management of medical resources; it is also a real-world example of precision medicine in hospitals.

Cost-Sensitive Uncertainty Hypergraph Learning for Identification of Lymph Node Involvement With CT Imaging

Lung adenocarcinoma (LUAD) is the most common type of lung cancer. Accurate identification of lymph node (LN) involvement in patients with LUAD is crucial for prognosis and making decisions of the treatment strategy. CT imaging has been used as a tool to identify lymph node involvement. To tackle the shortage of high-quality data and improve the sensitivity of diagnosis, we propose a Cost-Sensitive Uncertainty Hypergraph Learning (CSUHL) model to identify the lymph node based on the CT images. We design a step named "Multi-Uncertainty Measurement" to measure the epistemic and the aleatoric uncertainty, respectively. Given the two types of attentional uncertainty weights, we further propose a cost-sensitive hypergraph learning to boost the sensitivity of diagnosing, targeting task-driven optimization of the clinical scenarios. Extensive qualitative and quantitative experiments on the real clinical dataset demonstrate our method is capable of accurately identifying the lymph node and outperforming state-of-the-art methods across the board.

Developing Machine Learning Algorithms to Predict Pulmonary Complications After Emergency Gastrointestinal Surgery

Objective: Investigate whether machine learning can predict pulmonary complications (PPCs) after emergency gastrointestinal surgery in patients with acute diffuse peritonitis.

Methods: This is a secondary data analysis study. We use five machine learning algorithms (Logistic regression, DecisionTree, GradientBoosting, Xgbc, and gbm) to predict postoperative pulmonary complications.

Results: Nine hundred and twenty-six cases were included in this study; 187 cases (20.19%) had PPCs. The five most important variables for the postoperative weight were preoperative albumin, cholesterol on the 3rd day after surgery, albumin on the day of surgery, platelet count on the 1st day after surgery and cholesterol count on the 1st day after surgery for pulmonary complications. In the test group: the logistic regression model shows AUC = 0.808, accuracy = 0.824 and precision = 0.621; Decision tree shows AUC = 0.702, accuracy = 0.795 and precision = 0.486; The GradientBoosting model shows AUC = 0.788, accuracy = 0.827 and precision = 1.000; The Xgbc model shows AUC = 0.784, accuracy = 0.806 and precision = 0.583. The Gbm model shows AUC = 0.814, accuracy = 0.806 and precision = 0.750.

Conclusion: Machine learning algorithms can predict patients' PPCs with acute diffuse peritonitis. Moreover, the results of the importance matrix for the Gbdt algorithm model show that albumin, cholesterol, age, and platelets are the main variables that account for the highest pulmonary complication weights.