Publications by authors named "Wanliang Guo"

Background: Pulmonary hemorrhage (PH) in respiratory distress syndrome (RDS) in extremely preterm infants exhibits a high mortality rate and poor long-term outcomes. The aim of the present study was to develop a machine learning (ML) predictive model for RDS with PH in extremely preterm infants.

Methods: We performed a retrospective analysis of extremely preterm infants with RDS at the Children's Hospital of Soochow University between January 2015 and January 2021.

View Article and Find Full Text PDF

Background: Correctly diagnosing and accurately distinguishing mycoplasma pneumonia in children has consistently posed a challenge in clinical practice, as it can directly impact the prognosis of affected children. To address this issue, we analyzed chest X-rays (CXR) using various deep learning models to diagnose pediatric mycoplasma pneumonia.

Methods: We collected 578 cases of children with mycoplasma infection and 191 cases of children with virus infection, with available CXR sets.

View Article and Find Full Text PDF
Article Synopsis
  • The study aims to create a predictive model using clinical, radiomic, and deep learning features from X-ray and MRI to identify early risk factors for femoral head deformity in Legg-Calvé-Perthes disease (LCPD).
  • Involving 152 patients diagnosed with early unilateral LCPD, various machine learning methods (like XGBoost, which performed best) were used to develop predictive models, with the combined model showing the highest area under the ROC curve (AUC) at 0.853.
  • The results indicate that the integrated Clinic + Rad + DL model could help clinicians better assess the risk of early deformity in LCPD, leading to more personalized treatment strategies.
View Article and Find Full Text PDF
Article Synopsis
  • Children with atrial septal defects (ASD) and ventricular septal defects (VSD) are often assessed for respiratory symptoms, making it essential to distinguish between these two conditions due to differing treatments.
  • A study used deep learning analysis on chest radiographs to effectively differentiate between ASD and VSD; the research involved a large dataset and compared the performance of several algorithms.
  • The InceptionV3 model outperformed human radiologists with an accuracy of 87%, suggesting that deep learning could improve diagnostic efficiency for congenital heart defects.
View Article and Find Full Text PDF

Rationale And Objectives: Perineural invasion (PNI) is an important prognostic biomarker for prostate cancer (PCa). This study aimed to develop and validate a predictive model integrating biparametric MRI-based deep learning radiomics and clinical characteristics for the non-invasive prediction of PNI in patients with PCa.

Materials And Methods: In this prospective study, 557 PCa patients who underwent preoperative MRI and radical prostatectomy were recruited and randomly divided into the training and the validation cohorts at a ratio of 7:3.

View Article and Find Full Text PDF

Aim: To develop a decision-support tool for predicting extubation failure (EF) in neonates with bronchopulmonary dysplasia (BPD) using a set of machine-learning algorithms.

Methods: A dataset of 284 BPD neonates on mechanical ventilation was used to develop predictive models via machine-learning algorithms, including extreme gradient boosting (XGBoost), random forest, support vector machine, naïve Bayes, logistic regression, and k-nearest neighbor. The top three models were assessed by the area under the receiver operating characteristic curve (AUC), and their performance was tested by decision curve analysis (DCA).

View Article and Find Full Text PDF

Objective: Acute rupture and hemorrhage of pediatric brain arteriovenous malformations (AVMs) may lead to cerebral herniation or intractable intracranial hypertension, necessitating emerging surgical interventions to alleviate intracranial pressure. However, there is still controversy regarding the timing of treatment for ruptured AVMs. This study aimed to assess the feasibility of utilizing three-pillar expansive craniotomy (3PEC) at different times during the treatment of pediatric ruptured supratentorial AVMs.

View Article and Find Full Text PDF
Article Synopsis
  • This study explored using machine learning to identify risk factors and predict seizures in children with COVID-19.
  • A total of 519 children were analyzed using various machine learning models, with the random forest (RF) model showing the best performance in predicting seizures based on specific variables like neutrophil percentage, cough, and fever duration.
  • The findings suggest that the RF model and a newly created nomogram can aid clinicians in making informed decisions to prevent and manage seizures in affected children.
View Article and Find Full Text PDF
Article Synopsis
  • - The study focused on creating a quantitative radiomics-based nomogram to better diagnose acute pancreatitis (AP) in children with pancreaticobiliary maljunction (PBM), especially when CT scans show nonobvious signs of the condition.
  • - Researchers reviewed patient data to identify key clinical and radiological features associated with AP, ultimately developing various models that showed strong validity in both training and testing datasets (AUCs ranging from 0.757 to 0.938).
  • - The findings indicate that this new nomogram is a reliable and noninvasive way to diagnose AP in pediatric patients with PBM, even when typical CT signs are absent.
View Article and Find Full Text PDF

Background: Pancreaticobiliary maljunction (PBM) is a congenital defect, with risk of developing various pancreaticobiliary and hepatic complications. The presentations of PBM in children and adults are believed to be different, but studies on PBM children of different age groups are limited. This study was to evaluate clinicopathologic characteristics and outcomes in PBM children of different ages.

View Article and Find Full Text PDF
Article Synopsis
  • Researchers aimed to find radiomic features that could predict the pathological type of neuroblastic tumors in children by analyzing data from 104 cases.
  • They used advanced algorithms to classify tumors into two main comparisons: ganglioneuroma versus the other tumor types, and ganglioneuroblastoma versus neuroblastoma.
  • The study found that the classifier achieved high sensitivity and specificity, indicating that radiomic features are effective in predicting the types of neuroblastic tumors in children.
View Article and Find Full Text PDF
Article Synopsis
  • The study aimed to create machine learning models that can predict surgical risks in children with pancreaticobiliary maljunction (PBM) and biliary dilatation.
  • Conducted on 157 pediatric patients, four ML models (logistic regression, random forest, support vector machine, and extreme gradient boosting) were utilized, with their effectiveness measured using the area under the receiver operator characteristic curve (AUC).
  • The XGBoost model showed the best performance (AUC = 0.822), identifying key predictive factors such as choledochal cyst characteristics and bile duct variations, potentially helping surgeons prevent injuries during surgery.
View Article and Find Full Text PDF
Article Synopsis
  • The study aimed to create a model predicting post-operative acute pancreatitis (POAP) risk in children with pancreaticobiliary maljunction (PBM) using pre-operative data.
  • Various machine learning techniques, including logistic regression, were utilized to analyze patient data collected from August 2015 to August 2022 at a children's hospital.
  • The logistic regression model emerged as the most effective, highlighting key risk factors such as protein plugs, age, white blood cell count, and common bile duct diameter for predicting POAP.
View Article and Find Full Text PDF
Article Synopsis
  • This study aimed to create a diagnostic model using clinical and MRI radiomics features for detecting chronic cholangitis in children with pancreaticobiliary maljunction (PBM).
  • A total of 144 subjects' clinical characteristics and MRI features were analyzed, and a radiomics score was calculated to develop a combined model.
  • The combined model outperformed the clinical model alone in predictive accuracy, demonstrating significant potential for improving diagnosis in affected children.
View Article and Find Full Text PDF
Article Synopsis
  • The study aims to develop and validate a nomogram using MR-based radiomics and clinical factors to diagnose liver fibrosis in children with pancreaticobiliary maljunction (PBM), enhancing clinical decision-making and outcomes.
  • A total of 136 PBM patients from two centers were analyzed, with data divided into training and validation sets for accurate assessment of liver fibrosis determined via histopathological examination.
  • The results indicated that two clinical factors and four radiomics features effectively predicted liver fibrosis, with the nomogram demonstrating strong performance across different validation sets, showing clinical promise for patient management.
View Article and Find Full Text PDF

Background: To develop an end-to-end deep learning method for automated quantitative assessment of pediatric blunt hepatic trauma based on contrast-enhanced computed tomography (CT).

Methods: This retrospective study included 170 children with blunt hepatic trauma between May 1, 2015, and August 30, 2021, who had undergone contrast-enhanced CT. Both liver parenchyma and liver trauma regions were manually segmented from CT images.

View Article and Find Full Text PDF

To assess the impact of intrawound vancomycin on surgical site wound infections in non-spinal neurosurgical operations, we conducted a meta-analysis. A thorough review of the literature up to September 2022 showed that 4286 participants had a non-spinal neurosurgical operation at the start of the investigations; 1975 of them used intrawound vancomycin, while 2311 were control. Using dichotomous or contentious methods and a random or fixed-effect model, odds ratios (OR) and mean difference (MD) with 95% confidence intervals (CIs) were estimated to evaluate the impact of intrawound vancomycin on surgical site wound infections in non-spinal neurosurgical operation.

View Article and Find Full Text PDF

According to relevant data, the morbidity and mortality of strokes in China remain high. Without effective treatment, stroke morbidity and mortality will continue to rise, which may become the second leading disease in the world. With the nonstop advancement and improvement of clinical innovation in China, the death pace of stroke patients has dropped altogether.

View Article and Find Full Text PDF
Article Synopsis
  • The purpose of the study was to create a model that identifies risk factors and predicts acute pancreatitis in children with pancreaticobiliary maljunction (PBM).
  • The researchers used machine learning techniques, dividing their patient data into training and validation sets to compare the effectiveness of three models: logistic regression, support vector machine, and extreme gradient boosting.
  • The findings revealed that both XGBoost and SVM models performed similarly well (AUC around 0.814), outperforming the logistic regression model, and identified age, protein plugs, and white blood cell count as key predictors of acute pancreatitis.
View Article and Find Full Text PDF

Objective: This study was designed to investigate the predictors of bronchopulmonary dysplasia in neonates with respiratory distress syndrome.

Methods: This was a single-center retrospective cohort study conducted between 1 January 2015 and 31 December 2020. A total of 625 neonates with respiratory distress syndrome (RDS) were enrolled.

View Article and Find Full Text PDF

Objectives: The aim of this study was to establish an automatic classification model for chronic inflammation of the choledoch wall using deep learning with CT images in patients with pancreaticobiliary maljunction (PBM).

Methods: CT images were obtained from 76 PBM patients, including 61 cases assigned to the training set and 15 cases assigned to the testing set. The region of interest (ROI) containing the choledochal lesion was extracted and segmented using the UNet++ network.

View Article and Find Full Text PDF
Article Synopsis
  • Hirschsprung's disease (HSCR) is a condition where certain nerve cells are missing in the intestine, and researchers discovered various circular RNAs (circRNAs) related to this disease that could play a role in its development.* -
  • The study involved comparing circRNA levels in diseased and normal intestinal tissues, identifying 17 circRNAs that were upregulated and 10 that were downregulated in HSCR; the top five upregulated circRNAs were validated.* -
  • The research created a circRNA-miRNA-mRNA interaction network, suggesting that these circRNAs might function as molecular sponges, offering insights that could enhance HSCR diagnosis and treatment.*
View Article and Find Full Text PDF
Article Synopsis
  • The study investigates long non-coding RNAs (lncRNAs) as potential biomarkers for pancreaticobiliary maljunction (PBM) in children, highlighting the health complications linked to this condition.
  • Researchers compared the expression of lncRNAs and mRNAs between pediatric patients with PBM and control subjects, finding significant upregulation and downregulation of specific genetic markers.
  • The findings suggest that lncRNAs play a role in the pathogenesis of PBM, with one lncRNA identified as a promising biomarker for the disease.
View Article and Find Full Text PDF

Background: Choledochal cyst (CC) is a congenital bile duct malformation, with a higher incidence in minors. Patients with CCs are at risk of pancreatitis and ascending cholangitis. The main forms of treatments aim to avoid any possible hepatic, pancreatic, or biliary complications.

View Article and Find Full Text PDF

Introduction: Pediatric intussusception is one of the most common causes of bowel obstruction in the pediatric population. Affected children have one section of the intestine sliding into the adjacent section. Intestinal ischemia-reperfusion injury (I/R) can occur during pediatric intussusception, and any delay in diagnosis or treatment can lead to loss of intestinal viability that requires bowel resection.

View Article and Find Full Text PDF