A predictive model for post-thoracoscopic surgery pulmonary complications based on the PBNN algorithm.

We constructed an early prediction model for postoperative pulmonary complications after thoracoscopic surgery using machine learning and deep learning algorithms. The artificial intelligence prediction models were built in Python, primarily using artificial intelligencealgorithms including both machine learning and deep learning algorithms. Correlation analysis showed that postoperative pulmonary complications were positively correlated with age and surgery duration, and negatively correlated with serum albumin. Using the light gradient boosting machine(LGBM) algorithm, weighted feature engineering revealed that single lung ventilation duration, history of smoking, surgery duration, ASA score, and blood glucose were the main factors associated with postoperative pulmonary complications. Results of artificial intelligence algorithms for predicting pulmonary complications after thoracoscopy in the test group: In terms of accuracy, the two best algorithms were Logistic Regression (0.831) and light gradient boosting machine(0.827); in terms of precision, the two best algorithms were Gradient Boosting (0.75) and light gradient boosting machine (0.742); in terms of recall, the three best algorithms were gaussian naive bayes (0.581), Logistic Regression (0.532), and pruning Bayesian neural network (0.516); in terms of F1 score, the two best algorithms were LogisticRegression (0.589) and pruning Bayesian neural network (0.566); and in terms of Area Under Curve(AUC), the two best algorithms were light gradient boosting machine(0.873) and pruning Bayesian neural network (0.869). The results of this study suggest that pruning Bayesian neural network (PBNN) can be used to assess the possibility of pulmonary complications after thoracoscopy, and to identify high-risk groups prior to surgery.

Artificial intelligence algorithms for predicting post-operative ileus after laparoscopic surgery.

Objective: By constructing a predictive model using machine learning and deep learning technologies, we aim to understand the risk factors for postoperative intestinal obstruction in laparoscopic colorectal cancer patients, and establish an effective artificial intelligence-based predictive model to guide individualized prevention and treatment, thus improving patient outcomes. Methods: We constructed a model of the artificial intelligence algorithm in Python. Subjects were randomly assigned to either a training set for variable identification and model construction, or a test set for testing model performance, at a ratio of 7:3. The model was trained with ten algorithms. We used the AUC values of the ROC curves, as well as accuracy, precision, recall rate and F1 scores. Results: The results of feature engineering composited with the GBDT algorithm showed that opioid use, anesthesia duration, and body weight were the top three factors in the development of POI. We used ten machine learning and deep learning algorithms to validate the model, and the results were as follows: the three algorithms with best accuracy were XGB (0.807), Decision Tree (0.807) and Neural DecisionTree (0.807); the two algorithms with best precision were XGB (0.500) and Decision Tree (0.500); the two algorithms with best recall rate were adab (0.243) and Decision Tree (0.135); the two algorithms with highest F1 score were adab (0.290) and Decision Tree (0.213); and the three algorithms with best AUC were Gradient Boosting (0.678), XGB (0.638) and LinearSVC (0.633). Conclusion: This study shows that XGB and Decision Tree are the two best algorithms for predicting the risk of developing ileus after laparoscopic colon cancer surgery. It provides new insight and approaches to the field of postoperative intestinal obstruction in colorectal cancer through the application of machine learning techniques, thereby improving our understanding of the disease and offering strong support for clinical decision-making.

Predicting early postoperative PONV using multiple machine-learning- and deep-learning-algorithms.

OBJECTIVE: PONV reduces patient satisfaction and increases hospital costs as patients remain in the hospital for longer durations. In this study, we build a preliminary artificial intelligence algorithm model to predict early PONV in patients. METHODS: We use R for statistical analysis and Python for the machine learning prediction model. RESULTS: Average characteristic engineering results showed that haloperidol, sex, age, history of smoking, and history of PONV were the first 5 contributing factors in the occurrence of early PONV. Test group results for artificial intelligence prediction of early PONV: in terms of accuracy, the four best algorithms were CNNRNN (0.872), Decision Tree (0.868), SVC (0.866) and adab (0.865); in terms of precision, the three best algorithms were CNNRNN (1.000), adab (0.400) and adab (0.868); in terms of AUC, the top three algorithms were Logistic Regression (0.732), SVC (0.731) and adab (0.722). Finally, we built a website to predict early PONV online using the Streamlit app on the following website: ( https://zhouchengmao-streamlit-app-lsvc-ad-st-app-lsvc-adab-ponv-m9ynsb.streamlit.app/ ). CONCLUSION: Artificial intelligence algorithms can predict early PONV, whereas logistic regression, SVC and adab were the top three artificial intelligence algorithms in overall performance. Haloperidol, sex, age, smoking history, and PONV history were the first 5 contributing factors associated with early PONV.

Differentiation of Bone Metastasis in Elderly Patients With Lung Adenocarcinoma Using Multiple Machine Learning Algorithms.

OBJECTIVE: We tested the performance of general machine learning and joint machine learning algorithms in the classification of bone metastasis, in patients with lung adenocarcinoma. METHODS: We used R version 3.5.3 for statistical analysis of the general information, and Python to construct machine learning models. RESULTS: We first used the average classifiers of the 4 machine learning algorithms to rank the features and the results showed that race, sex, whether they had surgery and marriage were the first 4 factors affecting bone metastasis. Machine learning results in the training group: for area under the curve (AUC), except for RF and LR, the AUC values of all machine learning classifiers were greater than .8, but the joint algorithm did not improve the AUC for any single machine learning algorithm. Among the results related to accuracy and precision, the accuracy of other machine learning classifiers except the RF algorithm was higher than 70%, and only the precision of the LGBM algorithm was higher than 70%. Machine learning results in the test group: Similarly, for areas under the curve (AUC), except RF and LR, the AUC values for all machine learning classifiers were greater than .8, but the joint algorithm did not improve the AUC value for any single machine learning algorithm. For accuracy, except for the RF algorithm, the accuracy of other machine learning classifiers was higher than 70%. The highest precision for the LGBM algorithm was .675. CONCLUSION: The results of this concept verification study show that machine learning algorithm classifiers can distinguish the bone metastasis of patients with lung cancer. This will provide a new research idea for the future use of non-invasive technology to identify bone metastasis in lungcancer. However, more prospective multicenter cohort studies are needed.

Predicting postoperative gastric cancer prognosis based on inflammatory factors and machine learning technology.

OBJECTIVE: There is a strong association between gastric cancer and inflammatory factors. Many studies have shown that machine learning can predict cancer patients' prognosis. However, there has been no study on predicting gastric cancer death based on machine learning using related inflammatory factor variables. METHODS: Six machine learning algorithms are applied to predict total gastric cancer death after surgery. RESULTS: The Gradient Boosting Machine (GBM) algorithm factors accounting for the prognosis weight outcome show that the three most important factors are neutrophil-lymphocyte ratio (NLR), platelet lymphocyte ratio (PLR) and age. The total postoperative death model showed that among patients with gastric cancer from the predictive test group: The highest accuracy was LR (0.759), followed by the GBM algorithm (0.733). For the six algorithms, the AUC values, from high to low, were LR, GBM, GBDT, forest, Tr and Xgbc. Among the six algorithms, Logistic had the highest precision (precision = 0.736), followed by the GBM algorithm (precision = 0.660). Among the six algorithms, GBM had the highest recall rate (recall = 0.667). CONCLUSION: Postoperative mortality from gastric cancer can be predicted based on machine learning.

A long duration of intraoperative hypotension is associated with postoperative delirium occurrence following thoracic and orthopedic surgery in elderly.

BACKGROUND: Postoperative delirium (POD) is a common surgical complication associated with increased morbidity and mortality in elderly. Although the underlying mechanisms remain elusive, perioperative risk factors were reported to be closely related to its development. This study was designed to investigate the association between the duration of intraoperative hypotension and POD incidence following thoracic and orthopedic surgery in elderly. METHOD: The perioperative data from 605 elderly undergoing thoracic and orthopedic surgery from January 2021 to July 2022 were analyzed. The primary exposure was a cumulative duration of mean arterial pressure (MAP) ≤ 65 mmHg. The primary end-point was the POD incidence assessed with confusion assessment method (CAM) or CAM-ICU for three days after surgery. Restricted cubic spline (RCS) was conducted to examine the continuous relationship between the duration of intraoperative hypotension and POD incidence adjusted with patients' demographics and surgery related factors. Then the duration of intraoperative hypotension was categorized into three groups: no hypotension, short (< 5 mins) or long duration (≥ 5 mins) of hypotension for further analysis. RESULT: The incidence of POD was 14.7% (89 cases out of 605) within three days after surgery. The duration of hypotension presented a non-linear and "inverted L-shaped" effect on POD development. Compared to no hypotension, long duration (adjusted OR 3.93; 95% CI: 2.07-7.45; P < 0.001) rather than short duration of MAP ≤65 mmHg (adjusted OR 1.18; 95% CI: 0.56-2.50; P = 0.671) was closely related to the POD incidence. CONCLUSION: Intraoperative hypotension (MAP ≤65 mmHg) for ≥5 mins was associated with an increased incidence of POD after thoracic and orthopedic surgery in elderly.

Predicting difficult airway intubation in thyroid surgery using multiple machine learning and deep learning algorithms.

Background: In this paper, we examine whether machine learning and deep learning can be used to predict difficult airway intubation in patients undergoing thyroid surgery. Methods: We used 10 machine learning and deep learning algorithms to establish a corresponding model through a training group, and then verify the results in a test group. We used R for the statistical analysis and constructed the machine learning prediction model in Python. Results: The top 5 weighting factors for difficult airways identified by the average algorithm in machine learning were age, sex, weight, height, and BMI. In the training group, the AUC values and accuracy and the Gradient Boosting precision were 0.932, 0.929, and 100%, respectively. As for the modeled effects of predicting difficult airways in test groups, among the models constructed by the 10 algorithms, the three algorithms with the highest AUC values were Gradient Boosting, CNN, and LGBM, with values of 0.848, 0.836, and 0.812, respectively; In addition, among the algorithms, Gradient Boosting had the highest accuracy with a value of 0.913; Additionally, among the algorithms, the Gradient Boosting algorithm had the highest precision with a value of 100%. Conclusion: According to our results, Gradient Boosting performed best overall, with an AUC >0.8, an accuracy >90%, and a precision of 100%. Besides, the top 5 weighting factors identified by the average algorithm in machine learning for difficult airways were age, sex, weight, height, and BMI.

Quadratus Lumborum Block Spares Postoperative Opioid Usage but Does Not appear to Prevent the Development of Chronic Pain After Gastrointestinal Surgery.

BACKGROUND: Regional anesthesia has been used to reduce acute postsurgical pain and to  prevent chronic pain. The best technique, however, remains controversial. OBJECTIVES: The aim of this study was to assess the short- and long-term postoperative analgesic efficacy of ultrasound-guided quadratus lumborum block (QLB) in open gastrointestinal surgery. STUDY DESIGN: A randomized, double-blinded, controlled trial. SETTING: Operating room; postoperative recovery room and ward. METHODS: One hundred eighteen patients underwent elective gastrointestinal surgery randomly assigned into 2 groups (QLB group or control group). Before anesthetic induction, QLB was performed bilaterally under ultrasound guidance using 20 mL of 0.375% ropivacaine or saline solution at each abdominal wall. The primary outcome was cumulative oxycodone consumption within 24 h after surgery. The secondary outcomes were acute pain intensity, incidence of chronic pain, and incidence of postoperative nausea or vomiting (PONV), dizziness, and pruritus. RESULTS: The cumulative oxycodone consumption was significantly lower in the QLB group during the first 6, 6-24, 24, and 48 h postoperatively when compared to the control group. At rest or during coughing, the numeric rating scale scores were significantly lower at 1, 3, 6, and 12 h postoperatively in the QLB group compared to the control group. There were no significant differences between the 2 groups regarding the incidence of chronic postoperative pain at 3 or 6 months after surgery. Significant differences were found in the incidence of PONV between the two groups, but other complications, such as dizziness and pruritus, did not differ significantly. LIMITATIONS: We did not confirm the QLB effectiveness with sensory level testing after local anesthetic injection. Cumulative oxycodone consumption could have been affected by the patients' use of oxycodone for nonsurgical pain. CONCLUSIONS: Ultrasound-guided QLB provided superior short-term analgesia and reduced oxycodone consumption and the incidence of PONV after gastrointestinal surgery. However, the incidence of chronic pain was not significantly affected by this anesthetic technique.

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.

Machine Learning Can Predict Total Death After Radiofrequency Ablation in Liver Cancer Patients.

OBJECTIVE: Over 1 million new cases of hepatocellular carcinoma (HCC) are diagnosed worldwide every year. Its prognosis remains poor, and the 5-year survival rate in all disease stages is estimated to be between 10% and 20%. Radiofrequency ablation (RFA) has become an important local treatment for liver cancer, and machine learning (ML) can provide many shortcuts for liver cancer medical research. Therefore, we explore the role of ML in predicting the total mortality of liver cancer patients undergoing RFA. METHODS: This study is a secondary analysis of public database data from 578 liver cancer patients. We used Python for ML to establish the prognosis model. RESULTS: The results showed that the 5 most important factors were platelet count (PLT), Alpha-fetoprotein (AFP), age, tumor size, and total bilirubin, respectively. Results of the total death model for liver cancer patients in test group: among the 5 algorithm models, the highest accuracy rate was that of gbm (0.681), followed by the Logistic algorithm (0.672); among the 5 algorithms, area under the curve (AUC) values, from high to low, were Logistic (0.738), DecisionTree (0.723), gbm (0.717), GradientBoosting (0.714), and Forest (0.693); Among the 5 algorithms, gbm had the highest precision rate (0.721), followed by the Logistic algorithm (0.714). Among the 5 algorithms, DecisionTree had the highest recall rate (0.642), followed by the GradientBoosting algorithm (0.571). CONCLUSION: Machine learning can predict total death after RFA in liver cancer patients. Therefore, ML research has great potential for both personalized treatment and prognosis of liver cancer.

Predicting postoperative liver cancer death outcomes with machine learning.

OBJECTIVE: To investigate the effect of 5 machine learning algorithms in predicting total hepatocellular carcinoma (HCC) postoperative death outcomes. METHODS: This study was a secondary analysis. A prognosis model was established using machine learning with python. RESULTS: The results from the machine learning gbm algorithm showed that the most important factors, ranked from first to fifth, were: preoperative aspartate aminotransferase (GOT), preoperative AFP, preoperative cereal third transaminase (GPT), preoperative total bilirubin, and LC3. Postoperative death model results for liver cancer patients in the test group: of the 5 algorithm models, the highest accuracy rate was that of forest (0.739), followed by the gbm algorithm (0.714); of the 5 algorithms, the AUC values, from high to low, were forest (0.803), GradientBoosting (0.746), gbm (0.724), Logistic (0.660) and DecisionTree (0.578). CONCLUSION: Machine learning can predict total hepatocellular carcinoma postoperative death outcomes.

Predictors of Acute Postsurgical Pain following Gastrointestinal Surgery: A Prospective Cohort Study.

Background: Several predictors have been shown to be independently associated with chronic postsurgical pain for gastrointestinal surgery, but few studies have investigated the factors associated with acute postsurgical pain (APSP). The aim of this study was to identify the predictors of APSP intensity and severity through investigating demographic, psychological, and clinical variables. Methods: We performed a prospective cohort study of 282 patients undergoing gastrointestinal surgery to analyze the predictors of APSP. Psychological questionnaires were assessed 1 day before surgery. Meanwhile, demographic characteristics and perioperative data were collected. The primary outcomes are APSP intensity assessed by numeric rating scale (NRS) and APSP severity defined as a clinically meaningful pain when NRS ≥4. The predictors for APSP intensity and severity were determined using multiple linear regression and multivariate logistic regression, respectively. Results: 112 patients (39.7%) reported a clinically meaningful pain during the first 24 hours postoperatively. Oral morphine milligram equivalent (MME) consumption (ß 0.05, 95% CI 0.03-0.07, p < 0.001), preoperative anxiety (ß 0.12, 95% CI 0.08-0.15, p < 0.001), and expected postsurgical pain intensity (ß 0.12, 95% CI 0.06-0.18, p < 0.001) were positively associated with APSP intensity. Furthermore, MME consumption (OR 1.15, 95% CI 1.10-1.21, p < 0.001), preoperative anxiety (OR 1.33, 95% CI 1.21-1.46, p < 0.001), and expected postsurgical pain intensity (OR 1.36, 95% CI 1.17-1.57, p < 0.001) were independently associated with APSP severity. Conclusion: These results suggested that the predictors for APSP intensity following gastrointestinal surgery included analgesic consumption, preoperative anxiety, and expected postsurgical pain, which were also the risk factors for APSP severity.

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

To construct a machine learning algorithm model of lymph node metastasis (LNM) in patients with poorly differentiated-type intramucosal gastric cancer. 1169 patients with postoperative gastric cancer were divided into a training group and a test group at a ratio of 7:3. The model for lymph node metastasis was established with python machine learning. The Gbdt algorithm in the machine learning results finds that number of resected nodes, lymphovascular invasion and tumor size are the primary 3 factors that account for the weight of LNM. Effect of the LNM model of PDC gastric cancer patients in the training group: Among the 7 algorithm models, the highest accuracy rate was that of GBDT (0.955); The AUC values for the 7 algorithms were, from high to low, XGB (0.881), RF (0.802), GBDT (0.798), LR (0.778), XGB + LR (0.739), RF + LR (0.691) and GBDT + LR (0.626). Results of the LNM model of PDC gastric cancer patients in test group : Among the 7 algorithmic models, XGB had the highest accuracy rate (0.952); Among the 7 algorithms, the AUC values, from high to low, were GBDT (0.788), RF (0.765), XGB (0.762), LR (0.750), RF + LR (0.678), GBDT + LR (0.650) and XGB + LR (0.619). Single machine learning algorithm can predict LNM in poorly differentiated-type intramucosal gastric cancer, but fusion algorithm can not improve the effect of machine learning in predicting LNM.

Using Nomograms to Predict the PPCs of Patients With Diffuse Peritonitis Undergoing Emergency Gastrointestinal Surgery.

Objective: To develop and validate a nomogram model for predicting postoperative pulmonary complications (PPCs) in patients with diffuse peritonitis undergoing emergency gastrointestinal surgery. Methods: We used the least absolute shrinkage and selection operator (LASSO) regression model to analyze the independent risk factors for PPCs in patients with diffuse peritonitis who underwent emergency gastrointestinal surgery. Using R, we developed and validated a nomogram model for predicting PPCs in patients with diffuse peritonitis undergoing emergency gastrointestinal surgery. Results: The LASSO regression analysis showed that AGE, American Society of Anesthesiologists physical status classification (ASA), DIAGNOSIS, platelets (on the 3rd day after surgery), cholesterol (on the 3rd day after surgery), ALBUMIN (on the first day after surgery), and preoperative ALBUMIN were independent risk factors for PPCs in patients with diffuse peritonitis undergoing emergency gastrointestinal surgery. The area under the curve (AUC) value of the nomogram model in the training group was 0.8240; its accuracy was 0.7000, and its sensitivity was 0.8658. This demonstrates that the nomogram has a high prediction value. Also in the test group, the AUC value of the model established by the variables AGE, ASA, and platelets (on the 3rd day after surgery), cholesterol (on the 3rd day after surgery), ALBUMIN (on the first day after surgery), and preoperative ALBUMIN was 0.8240; its accuracy was 0.8000; and its specificity was 0.8986. In the validation group, the same results were obtained. The results of the clinical decision curve show that the benefit rate was also high. Conclusion: Based on the risk factors AGE, ASA, DIAGNOSIS, platelets (on the 3rd day after surgery), cholesterol (on the 3rd day after surgery), ALBUMIN (on the first day after surgery), and preoperative ALBUMIN, the nomogram model established in this study for predicting PPCs in patients with diffuse peritonitis undergoing emergency gastrointestinal surgery has high accuracy and discrimination.

Machine learning to predict the cancer-specific mortality of patients with primary non-metastatic invasive breast cancer.

PURPOSE: We used five machine-learning algorithms to predict cancer-specific mortality after surgical resection of primary non-metastatic invasive breast cancer. METHODS: This study was a secondary analysis of data for 1661 women with primary non-metastatic invasive breast cancer. The overall patient population was divided into a training group and a test group at a ratio of 8:2 and python was used for machine learning to establish the prognosis model. RESULTS: The machine-learning Gbdt algorithm for cancer-specific death caused by various factors showed the five most important factors, ranked from high to low as follows: the number of regional lymph node metastases, LDH, triglyceride, plasma fibrinogen, and cholesterol. Among the five algorithm models in the test group, the highest accuracy rate was by DecisionTree (0.841), followed by the gbm algorithm (0.838). Among the five algorithms, the AUC values from high to low were GradientBoosting (0.755), gbm (0.755), Logistic (0.733), Forest (0.715), and DecisionTree (0.677). CONCLUSION: Machine learning can predict cancer-specific mortality after surgery for patients with primary non-metastatic invasive breast.

Predicting postoperative delirium after microvascular decompression surgery with machine learning.

OBJECTIVE: The aim of this study was to predict early delirium after microvascular decompression using machine learning. DESIGN: Retrospective cohort study. SETTING: Second Hospital of Lanzhou University. PATIENTS: This study involved 912 patients with primary cranial nerve disease who had undergone microvascular decompression surgery between July 2007 and June 2018. INTERVENTIONS: None. MEASUREMENTS: We collected data on preoperative, intraoperative, and postoperative variables. Statistical analysis was conducted in R, and the model was constructed with python. The machine learning model was run using the following models: decision tree, logistic regression, random forest, gbm, and GBDT models. RESULTS: 912 patients were enrolled in this study, 221 of which (24.2%) had postoperative delirium. The machine learning Gbm algorithm finds that the first five factors accounting for the weight of postoperative delirium are CBZ use duration, hgb, serum CBZ level measured 24 h before surgery, preoperative CBZ dose, and BUN. Through machine learning five algorithms to build prediction models, we found the following values for the training group: Logistic algorithm (AUC value = 0.925, accuracy = 0.900); Forest algorithm (AUC value = 0.994, accuracy = 0.948); GradientBoosting algorithm (AUC value = 0.994, accuracy = 0.970) and DecisionTree algorithm (aucvalue = 0.902, accuracy = 0.861); Gbm algorithm (AUC value = 0.979, accuracy = 0.944). The test group had the following values: Logistic algorithm (aucvalue = 0.920, accuracy = 0.901); DecisionTree algorithm (aucvalue = 0.888, accuracy = 0.883); Forest algorithm (aucvalue = 0.963, accuracy = 0.909); GradientBoostingc algorithm (aucvalue = 0.962, accuracy = 0.923); Gbm algorithm (AUC value = 0.956, accuracy = 0.920). CONCLUSION: Machine learning algorithms predict the occurrence of delirium after microvascular decompression with an accuracy rate of 96.7%. And the major risk factors for the development of post-cardiac delirium are carbamazepine, hgb, and BUN.