Establishing and clinically validating a machine learning model for predicting unplanned reoperation risk in colorectal cancer.

BACKGROUND: Colorectal cancer significantly impacts global health, with unplanned reoperations post-surgery being key determinants of patient outcomes. Existing predictive models for these reoperations lack precision in integrating complex clinical data. AIM: To develop and validate a machine learning model for predicting unplanned reoperation risk in colorectal cancer patients. METHODS: Data of patients treated for colorectal cancer (n = 2044) at the First Affiliated Hospital of Wenzhou Medical University and Wenzhou Central Hospital from March 2020 to March 2022 were retrospectively collected. Patients were divided into an experimental group (n = 60) and a control group (n = 1984) according to unplanned reoperation occurrence. Patients were also divided into a training group and a validation group (7:3 ratio). We used three different machine learning methods to screen characteristic variables. A nomogram was created based on multifactor logistic regression, and the model performance was assessed using receiver operating characteristic curve, calibration curve, Hosmer-Lemeshow test, and decision curve analysis. The risk scores of the two groups were calculated and compared to validate the model. RESULTS: More patients in the experimental group were ≥ 60 years old, male, and had a history of hypertension, laparotomy, and hypoproteinemia, compared to the control group. Multiple logistic regression analysis confirmed the following as independent risk factors for unplanned reoperation (P < 0.05): Prognostic Nutritional Index value, history of laparotomy, hypertension, or stroke, hypoproteinemia, age, tumor-node-metastasis staging, surgical time, gender, and American Society of Anesthesiologists classification. Receiver operating characteristic curve analysis showed that the model had good discrimination and clinical utility. CONCLUSION: This study used a machine learning approach to build a model that accurately predicts the risk of postoperative unplanned reoperation in patients with colorectal cancer, which can improve treatment decisions and prognosis.

Analysis of Associated Factors, Construction, and Application of a Predictive Model for Lymph Node Metastasis in Colorectal Cancer.

Background: In colorectal cancer (CRC) , understanding lymph node metastasis (LNM) is critical for effective treatment. Better approaches are required for identifying and assessing the risk contributions of factors influencing lymph node metastasis in colorectal cancer. Objective: This study aims to analyze factors associated with LNM in CRC and develop a risk prediction model. Methods: A retrospective cohort study was conducted and a total of 181 CRC patients admitted between March 2020 and April 2023 were selected as research participants. Among them, 47 patients developed LNM, while the remaining 134 did not. Clinical data, including age, sex, pathological stages, were collected. Logistic regression was employed to identify factors influencing LNM in CRC, forming the basis for constructing a risk model. The diagnostic efficiency of this model was assessed through receiver operating characteristic (ROC) curves. Results: Tumor nodules and histological types showed no correlation with LNM in CRC (P > .05). However, pathological staging, vascular and neural invasion, use of VEGF inhibitors, and preoperative CEA were identified as independent risk factors for LNM in CRC (P < .05). The established model demonstrated a good fit with the observations. ROC curve analysis indicated an area under the curve (AUC) of 0.884 for predicting LNM in CRC, signifying excellent predictive performance. Conclusions: The risk model, formulated on factors associated with LNM in CRC, serves as a efficient tool in assessing the probability of LNM. It provides invaluable insights that can significantly enhance clinical approaches to the diagnosis and treatment of CRC in the future.

A New Nomogram Based on Early Postoperative NLR for Predicting Infectious Complications After Gastrectomy.

PURPOSE: Our study aimed to construct a visible model to evaluate the risk of infectious complications after gastrectomy. METHODS: The clinical data of 856 patients who underwent gastrectomy were used to retrieve medical records. Univariate and multivariate analyses were performed to correlate early postoperative NLR and operative variables with postoperative complications, and the construction of the nomogram was based on logistic regression. The concordance index and receiver operating characteristic curves were used to evaluate the model performance. RESULTS: The postoperative infectious and noninfectious complication rates after gastrectomy were 18.5% (158/856 cases) and 12.3% (105/856 cases) respectively. Postoperative NLR (within 24 h) independently predicted the development of postoperative infectious complication. Multivariate analysis revealed that age, diabetes, body mass index (BMI), intraoperative blood transfusion and postoperative NLR were independent risk factors. The nomogram model showed a good performance in terms of predicting infectious complications after gastrectomy (concordance index=0.718). CONCLUSION: Age, diabetes, BMI, intraoperative blood transfusion and postoperative NLR were independent risk factors of postoperative infectious complications after gastrectomy, and a novel nomogram based on these results can be used to predict postoperative infection and has the advantages of simple application and easy access.