Ultrasound-based radiomics XGBoost model to assess the risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual application of SHAP
Yan ShiYing ZouJihua LiuYuanyuan WangYingbin ChenFang SunZhi YangGuanghe CuiXijun ZhuXu CuiFei‐Fei Liu
24
Citation
43
Reference
10
Related Paper
Citation Trend
Abstract:
A radiomics-based explainable eXtreme Gradient Boosting (XGBoost) model was developed to predict central cervical lymph node metastasis (CCLNM) in patients with papillary thyroid carcinoma (PTC), including positive and negative effects.A total of 587 PTC patients admitted at Binzhou Medical University Hospital from 2017 to 2021 were analyzed retrospectively. The patients were randomized into the training and test cohorts with an 8:2 ratio. Radiomics features were extracted from ultrasound images of the primary PTC lesions. The minimum redundancy maximum relevance algorithm and the least absolute shrinkage and selection operator regression were used to select CCLNM positively-related features and radiomics scores were constructed. Clinical features, ultrasound features, and radiomics score were screened out by the Boruta algorithm, and the XGBoost model was constructed from these characteristics. SHapley Additive exPlanations (SHAP) was used for individualized and visualized interpretation. SHAP addressed the cognitive opacity of machine learning models.Eleven radiomics features were used to calculate the radiomics score. Five critical elements were used to build the XGBoost model: capsular invasion, radiomics score, diameter, age, and calcification. The area under the curve was 91.53% and 90.88% in the training and test cohorts, respectively. SHAP plots showed the influence of each parameter on the XGBoost model, including positive (i.e., capsular invasion, radiomics score, diameter, and calcification) and negative (i.e., age) impacts. The XGBoost model outperformed the radiologist, increasing the AUC by 44%.The radiomics-based XGBoost model predicted CCLNM in PTC patients. Visual interpretation using SHAP made the model an effective tool for preoperative guidance of clinical procedures, including positive and negative impacts.Keywords:
Nomogram
Nomogram
Cite
Citations (235)
Nomogram
Cite
Citations (102)
A radiomics nomogram for preoperatively predicting prognosis of patients in hepatocellular carcinoma
Background: Increasing studies have indicated that biomarkers based on quantitative radiomics features are related to clinical prognosis across a range of cancer types, but the association between radiomics and prognosis in hepatocellular carcinoma (HCC) is unclear. We aimed to develop and validate a radiomics nomogram for the preoperative prediction of prognosis for patients with HCC undergoing partial hepatectomy. Methods: In total, 177 patients were randomly divided into training (n=113) and validation (n=64) cohorts. A total number of 980 radiomics features were extracted from computed tomography images. And the least absolute shrinkage and selection operator algorithm was used to select the optimal features and build a radiomics signature in the training set. Besides, a radiomics nomogram was developed using multivariate regression analysis. The performance of the radiomics nomogram was estimated regarding its discrimination and calibration abilities, and clinical usefulness. Results: The radiomics signature was significantly associated with disease-free survival (DFS) (P Conclusions: The proposed radiomics nomogram showed excellent performance for the individualized and non-invasive estimation of DFS, which may help clinicians better identify patients with HBV-related HCC who can benefit from the surgery.
Nomogram
Cite
Citations (3)
Nomogram
Cite
Citations (18)
Nomogram
Biochemical recurrence
Cite
Citations (80)
Abstract This 1:5 case‐control study aimed to identify the risk factors of hospital‐acquired pressure injuries (HAPIs) and to develop a mathematical model of nomogram for the risk prediction of HAPIs. Data for 370 patients with HAPIs and 1971 patients without HAPIs were extracted from the adverse events and the electronic medical systems. They were randomly divided into two sets: training (n = 1951) and validation (n = 390). Significant risk factors were identified by univariate and multivariate analyses in the training set, followed by a nomogram constructed. Age, independent movement, sensory perception and response, moisture, perfusion, use of medical devices, compulsive position, hypoalbuminaemia, an existing pressure injury or scarring from a previous pressure injury, and surgery sufferings were considered significant risk factors and were included to construct a nomogram. In both of the training and validation sets, the areas of 0.90 under the receiver operating characteristic curves showed excellent discrimination of the nomogram; calibration plots demonstrated a good consistency between the observed probability and the nomogram's prediction; decision curve analyses exhibited preferable net benefit along with the threshold probability in the nomogram. The excellent performance of the nomogram makes it a convenient and reliable tool for the risk prediction of HAPIs.
Nomogram
Univariate
Pressure injury
Cite
Citations (13)
Nomogram
Cite
Citations (9)
Objective: To compare the diagnostic accuracy of various transcutaneous bilirubin (TcB) nomograms for predischarge screening. Methods: The paired total serum bilirubin (TSB) and TcB measurements collected in neonates ≥35 weeks and ≥2000 g birth weight were analyzed. BiliCare™ bilirubinometer was used for TcB measurement. We chose the following nomograms for the study: Bhutani nomogram, Maisel's nomogram, Agarwal nomogram, Thakkar nomogram, American Academy of Pediatrics (AAP) nomogram within 3 mg/dl of phototherapy cutoff, AAP nomogram >70% of phototherapy cutoff and if TcB value is above 13 mg/dl. The diagnostic accuracy of these nomograms for TcB was compared with TSB plotted in the Bhutani nomogram. Results: TcB showed a positive correlation with TSB (Pearson correlation coefficient = 0.783). Bhutani nomogram, Maisel's nomogram and AAP (using within 3 mg/dL cutoff) nomogram showed good sensitivity and low false-negative rate while avoiding blood draws in most neonates. Conclusion: Bhutani nomogram, Maisel's nomogram, and AAP (using within 3 mg/dL of phototherapy cutoff) nomograms have comparable diagnostic accuracy for predischarge bilirubin screening in neonates.
Nomogram
Cut-off
Cite
Citations (0)
Nomogram
Predictive modelling
Cite
Citations (96)
Abstract Background: The aim of the study was to establish and validate nomograms to predict the mortality risk of patients with COVID-19 using routine clinical indicators. Method: This retrospective study included a development cohort enrolled 2119 hospitalized COVID-19 patients and a validation cohort included 1504 COVID-19 patients. The demographics, clinical manifestations, vital signs and laboratory test results of the patients at admission and outcome of in-hospital death were recorded. The independent factors associated with death were identified by a forward stepwise multivariate logistic regression analysis and used to construct two prognostic nomograms. The models were then tested in an external dataset. Results: Nomogram 1 is a full model included nine factors identified in the multivariate logistic regression and nomogram 2 is built by selecting four factors from nine to perform as a reduced model. Nomogram 1 and nomogram 2 established showed better performance in discrimination and calibration than the MuLBSTA score in training. In validation, Nomogram 1 performed better than nomogram 2 for calibration. Conclusion: Nomograms we established performed better than the MuLBSTA score. We recommend the application of nomogram 1 in general hospital which provide robust prognostic performance but more cumbersome; nomogram 2 in mobile cabin hospitals which depend on less laboratory examinations and more convenient. Both nomograms can help clinicians in identifying patients at risk of death with routine clinical indicators at admission, which may reduce the overall mortality of COVID-19.
Nomogram
Cite
Citations (1)