Preoperative Prediction Model Research Articles

High-resolution MRI-based radiomics analysis to predict lymph node metastasis and tumor deposits respectively in rectal cancer.

To establish and validate two predictive radiomics models for preoperative prediction of lymph node metastases (LNMs) and tumor deposits (TDs) respectively in rectal cancer (RC) patients. A total of 139 RC patients (98 in the training cohort and 41 in the validation cohort) were enrolled in the present study. High-resolution magnetic resonance images (HRMRI) were retrieved for tumor segmentation and feature extraction. HRMRI findings of RC were assessed by three experienced radiologists. Two radiomics nomograms were established by integrating the clinical risk factors, HRMRI findings and radiomics signature. The predictive nomogram of LNMs showed good predictive performance (area under the curve [AUC], 0.90; 95% confidence interval [CI] 0.83-0.96) which was better than clinico-radiological (AUC, 0.83; 95% CI 0.74-0.93; Delong test, p=0.017) or radiomics signature-only model (AUC, 0.77; 95% CI 0.67-0.86; Delong test, p=0.003) in training cohort. Application of the nomogram in the validation cohort still exhibited good performance (AUC, 0.87; 95% CI 0.76-0.98). The accuracy, sensitivity and specificity of the combined model in predicting LNMs was 0.86,0.79 and 0.91 in training cohort and 0.83,0.85 and 0.82 in validation cohort. As for TDs, the predictive efficacy of the nomogram (AUC, 0.82; 95% CI 0.71-0.93) was not significantly higher than radiomics signature-only model (AUC, 0.80; 95% CI 0.69-0.92; Delong test, p = 0.71). Radiomics signature-only model was adopted to predict TDs with accuracy=0.76, sensitivity=0.72 and specificity=0.94 in training cohort and 0.68, 0.62 and 0.97 in validation cohort. HRMRI-based radiomics models could be helpful for the prediction of LNMs and TDs preoperatively in RC patients.

Establishment of Decision Rules and Risk Assessment Model for Preoperative Prediction of Lymph Node Metastasis in Gastric Cancer.

Background: Preoperative accurate prediction of lymph node status is especially important for the formulation of treatment plans for patients with gastric cancer (GC). The purpose of this study was to establish decision rules and a risk assessment model for lymph node metastasis (LNM) in GC using preoperative indicators.Methods: The clinical data of 554 patients who underwent gastrectomy with D2 lymphadenectomy were collected. A 1:1 propensity score matching (PSM) system was used, and the clinical data of the matched 466 patients were further analyzed. The important risk factors for LNM were extracted by the random forest algorithm, and decision rules and nomogram models for LNM were constructed with a classification tree and the “rms” package of R software, respectively.Results: Tumor size (OR: 2.058; P = 0.000), computed tomography (CT) findings (OR: 1.969; P = 0.001), grade (OR: 0.479; P = 0.000), hemoglobin (Hb) (OR: 1.211; P = 0.005), CEA (OR: 1.111; P = 0.017), and CA19-9 (OR: 1.040; P = 0.033) were independent risk factors for LNM in GC. Tumor size did rank first in the ranking of important factors for LNM in GC and was the first-level segmentation of the two initial branches of the classification tree. The accuracy, sensitivity, specificity, and positive predictive value of the decision rules in diagnosing preoperative LNM in GC were 75.6, 85.7, 73.9, 73.5, and 79.3%, respectively. The accuracy, sensitivity, and specificity of the risk assessment model in predicting preoperative LNM in GC were 79.3, 80.3, and 79.4%, respectively.Conclusion: Tumor size was the most important factor for evaluating LNM in GC. This decision rules and nomogram model constructed to take into account tumor size, CT findings, grade, hemoglobin, CEA, and CA19-9 effectively predicted the incidence of LNM in preoperative GC.

Ultrasound-based risk model for preoperative prediction of lymph-node metastases in women with endometrial cancer: model-development study.

To develop a preoperative risk model, using endometrial biopsy results and clinical and ultrasound variables, to predict the individual risk of lymph-node metastases in women with endometrial cancer. A mixed-effects logistic regression model for prediction of lymph-node metastases was developed in 1501 prospectively included women with endometrial cancer undergoing transvaginal ultrasound examination before surgery, from 16 European centers. Missing data, including missing lymph-node status, were imputed. Discrimination, calibration and clinical utility of the model were evaluated using leave-center-out cross validation. The predictive performance of the model was compared with that of risk classification from endometrial biopsy alone (high-risk defined as endometrioid cancer Grade 3/non-endometrioid cancer) or combined endometrial biopsy and ultrasound (high-risk defined as endometrioid cancer Grade 3/non-endometrioid cancer/deep myometrial invasion/cervical stromal invasion/extrauterine spread). Lymphadenectomy was performed in 691 women, of whom 127 had lymph-node metastases. The model for prediction of lymph-node metastases included the predictors age, duration of abnormal bleeding, endometrial biopsy result, tumor extension and tumor size according to ultrasound and undefined tumor with an unmeasurable endometrium. The model's area under the curve was 0.73 (95% CI, 0.68-0.78), the calibration slope was 1.06 (95% CI, 0.79-1.34) and the calibration intercept was 0.06 (95% CI, -0.15 to 0.27). Using a risk threshold for lymph-node metastases of 5% compared with 20%, the model had, respectively, a sensitivity of 98% vs 48% and specificity of 11% vs 80%. The model had higher sensitivity and specificity than did classification as high-risk, according to endometrial biopsy alone (50% vs 35% and 80% vs 77%, respectively) or combined endometrial biopsy and ultrasound (80% vs 75% and 53% vs 52%, respectively). The model's clinical utility was higher than that of endometrial biopsy alone or combined endometrial biopsy and ultrasound at any given risk threshold. Based on endometrial biopsy results and clinical and ultrasound characteristics, the individual risk of lymph-node metastases in women with endometrial cancer can be estimated reliably before surgery. The model is superior to risk classification by endometrial biopsy alone or in combination with ultrasound. Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd.

A Novel RNA-Seq-Based Model for Preoperative Prediction of Lymph Node Metastasis in Oral Squamous Cell Carcinoma.

Objective To develop and validate a novel RNA-seq-based nomogram for preoperative prediction of lymph node metastasis (LNM) for patients with oral squamous cell carcinoma (OSCC). Methods RNA-seq data for 276 OSCC patients (including 157 samples with LNM and 119 without LNM) were downloaded from TCGA database. Differential expression analysis was performed between LNM and non-LNM of OSCC. These samples were divided into a training set and a test set by a ratio of 9 : 1 while the relative proportion of the non-LNM and LNM groups was kept balanced within each dataset. Based on clinical features and seven candidate RNAs, we established a prediction model of LNM for OSCC using logistic regression analysis. Tenfold crossvalidation was utilized to examine the accuracy of the nomogram. Decision curve analysis was performed to evaluate the clinical utility of the nomogram. Results A total of 139 differentially expressed RNAs were identified between LNM and non-LNM of OSCC. Seven candidate RNAs were screened based on FPKM values, including NEURL1, AL162581.1 (miscRNA), AP002336.2 (lncRNA), CCBE1, CORO6, RDH12, and AC129492.6 (pseudogene). Logistic regression analysis revealed that the clinical N stage (p < 0.001) was an important factor to predict LNM. Moreover, three RNAs including RDH12 (p value < 0.05), CCBE1 (p value < 0.01), and AL162581.1 (p value < 0.05) could be predictive biomarkers for LNM in OSCC patients. The average accuracy rate of the model was 0.7661, indicating a good performance of the model. Conclusion Our findings constructed an RNA-seq-based nomogram combined with clinicopathology, which could potentially provide clinicians with a useful tool for preoperative prediction of LNM and be tailored for individualized therapy in patients with OSCC.

NCV INDEX IS A POTENTIAL PREDICTOR OF MVI IN PATIENTS WITH HCC WHO UNDERGOING LIVER TRANSPLANTATION

Background: MVI is a potential screening indicator for liver cancer liver transplantation prognosis and neoadjuvant therapy. Currently, there is no effective way to accurately predict MVI before surgery. Objective: To construct a preoperative MVI model for liver cancer patients undergoing liver transplantation. Methods: We retrospectively analyzed the clinical and follow-up data of 168 patients with hepatocellular carcinoma treated in the Department of General Surgery, Huashan Hospital, Fudan University from June 25, 2014 to October 31, 2018. (Randomly divided into experimental group 112 cases, verified Group of 56 patients). Inclusion criteria: pathological diagnosis of hepatocellular carcinoma (HCC); exclusion criteria: (1) major blood vessel invasion; (2) extrahepatic lymph node metastasis; (3) distant metastasis; (4) death within one month after surgery. We used Logistic regression model to analyze the risk factors of MVI, and constructed a preoperative risk prediction model of MVI. The area under the receiver curve (ROC) was used to test the predictive power of the risk model, and a calibration analysis was used to test the accuracy of the predictive model. Finally, K-M analysis was used to analyze the relationship between high and low MVI risk scores and prognosis. Results: Two factors (number of tumors and cumulative tumor volume) were found to be independent risk factors for predicting MVI by multivariate logistic regression analysis. Based on this, they were included in the nomogram analysis to establish the number of tumors (Number)-cumulative volume (Cumulative Volume) index (NCV index). The area under the NCV index of the experimental group and the validation cohort reached 0.933 (95% CI: 0.875-0.992) and 0.861 (95% CI: 0.733-0.988), respectively, which are significantly better than the similar models reported in the literature (P <0.05). The prediction consistency of the NCV model reached 0.949 and 0.252 in the experimental group and the validation group, respectively. The optimal threshold for NCV index according to Yoden index is 0.81. The accuracy of NCV index prediction is as follows: experimental group: sensitivity 89%, specificity 90%, positive predictive value 96.1%, negative predictive value 75%; validation group: sensitivity 79.5%, specificity 91.7%, positive predictive value 97.2%, negative The predicted value is 55%. K-M analysis showed that MVI score was significantly correlated with recurrence and survival after liver cancer liver transplantation (P <0.05). Conclusion: The NCV index based on the number of tumors and the cumulative volume can effectively predict the MVI of liver cancer patients undergoing liver transplantation and is expected to become an important evaluation tool for clinical treatment decision-making. China National Natural Science Foundation (81672365).

A novel MRI-based predictive index can identify patients suitable for preservation of the nipple-areola complex in breast reconstructive surgery

IntroductionAccurately predicting nipple-areola complex (NAC) involvement in breast cancer is necessary for identifying patients who may be candidates for a nipple-sparing mastectomy. Although multiple risk factors are indicated in the guidelines, it is difficult to predict NAC involvement (NAC-i) preoperatively even if these factors are evaluated individually. This study aimed to develop a more accurate and practical preoperative NAC-i prediction model using magnetic resonance imaging (MRI). Materials and methodsAll tumors in 252 patients were evaluated using postcontrast T1-weighted subtraction on MRI. ResultsThe receiver operating characteristic curves identified cut-off values for tumor size and tumor-to-nipple distance (TND) as 4 cm and 1.2 cm, respectively. Multivariate analysis demonstrated that TND (p < 0.001), ductal enhancement extending to the nipple (DEEN) (p < 0.001), and nipple enhancement (NE) (p = 0.005) were independent clinical risk factors for pathological NAC-i. A formula was constructed using odds ratios for these three independent preoperative risk factors in multivariate analysis: the MRI-based NAC-i predictive index (mNACPI) = TND × 4 + DEEN × 3 + NE × 1. A total score of ≤4 points was defined as low risk and ≥5 points as high risk. NAC-i rates were 2.4% in the low-risk group and 89.4% in the high-risk group; a significant correlation was observed between the risk group and permanent pathological NAC-i (p < 0.001). Assuming that the NAC was preserved in low-risk patients and resected in high-risk patients, NAC-i was verified using the mNACPI. ConclusionmNACPI may contribute greatly to the improvement of selecting suitable patients for NAC preservation in breast reconstructive surgery while maintaining oncological safety.

A preoperative prediction model for predicting coexisting adnexa malignancy of patients with G1/G2 endometrioid endometrial cancer

ObjectiveTo identify the predictors of coexisting adnexa malignancy (CAM) before surgery for patients with G1/G2 endometrioid endometrial cancer (EEC). MethodsPatients with G1/G2 EEC who received surgery in Fudan University Shanghai Cancer Center from 1996 to 2017 were enrolled. Univariate and multivariate logistic regression were performed to identify the predictors for CAM, and the nomogram was constructed and evaluated the discrimination and calibration. ResultsAmong the 1511 patients in the study cohort, 66 (4.4%) coexisted adnexa malignancy (51 metastatic and 15 synchronous primaries). In the univariate logistic regression analysis, CA125 level (>35 U/ml), histologic grades, myometrial invasion depth in magnetic resonance imaging (MRI), adnexal involvement in MRI/surgical exploration (SEP) were found to be significant predictors for CAM (P < .001, 0.047, 0.011, <0.001, respectively). The multivariate analysis demonstrated that high CA125 level (P < .001; OR: 2.945; 95%CI: 1.700–5.101), deep myometrial invasion (P = .011; OR: 2.194; 95%CI: 1.200–4.011), and suspected adnexal involvement in MRI/SEP (P < .001; OR: 11.524; 95%CI: 6.726–19.744) were independent predictors for CAM (AUC = 0.786). In 338 patients with MMR results, eighty-seven (25.7%) were detected MSI-high. There were 5.7% (5/87) patients diagnosed with CAM in the MSI-high group compared with 4.4% (11/251) in the MSS group. ConclusionsA nomogram with pre- and intra-operative factors was constructed to predict CAM in G1/G2 EEC patients, which may help clinicians in decision-making for ovarian preservation for these patients.

Artificial neural network model for preoperative prediction of severe liver failure after hemihepatectomy in patients with hepatocellular carcinoma

BackgroundPosthepatectomy liver failure is a worrisome complication after major hepatectomy for hepatocellular carcinoma and is the leading cause of postoperative mortality. Recommendations for hepatectomy for hepatocellular carcinoma are based on the risk of severe posthepatectomy liver failure, and accurately predicting posthepatectomy liver failure risk before undertaking major hepatectomy is of great significance. Thus, herein, we aimed to establish and validate an artificial neural network model to predict severe posthepatectomy liver failure in patients with hepatocellular carcinoma who underwent hemihepatectomy. MethodsThree hundred and fifty-three patients who underwent hemihepatectomy for hepatocellular carcinoma were included. We randomly divided the patients into a development set (n = 265, 75%) and a validation set (n = 88, 25%). Multivariate logistic analysis facilitated identification of independent variables that we incorporated into the artificial neural network model to predict severe posthepatectomy liver failure in the development set and then verified in the validation set. ResultsThe morbidity of patients with severe posthepatectomy liver failure in the development and validation sets was 24.9% and 23.9%, respectively. Multivariate analysis revealed that platelet count, prothrombin time, total bilirubin, aspartate aminotransferase, and standardized future liver remnant were all significant predictors of severe posthepatectomy liver failure. Incorporating these factors, the artificial neural network model showed satisfactory area under the receiver operating characteristic curve for the development set of 0.880 (95% confidence interval, 0.836–0.925) and for the validation set of 0.876 (95% confidence interval, 0.801–0.950) in predicting severe posthepatectomy liver failure and achieved well-fitted calibration ability. The predictive performance of the artificial neural network model for severe posthepatectomy liver failure outperformed the traditional logistic regression model and commonly used scoring systems. Moreover, stratification into 3 risk groups highlighted significant differences between the incidences and grades of posthepatectomy liver failure. ConclusionThe artificial neural network model accurately predicted the risk of severe posthepatectomy liver failure in patients with hepatocellular carcinoma who underwent hemihepatectomy. Our artificial neural network model might help surgeons identify intermediate and high-risk patients to facilitate earlier interventions.

Preoperative computed tomography-guided disease-free survival prediction in gastric cancer: a multicenter radiomics study.

Preoperative and noninvasive prognosis evaluation remains challenging for gastric cancer. Novel preoperative prognostic biomarkers should be investigated. This study aimed to develop multidetector-row computed tomography (MDCT)-guided prognostic models to direct follow-up strategy and improve prognosis. A retrospective dataset of 353 gastric cancer patients were enrolled from two centers and allocated to three cohorts: training cohort (n=166), internal validation cohort (n=83), and external validation cohort (n=104). Quantitative radiomic features were extracted from MDCT images. The least absolute shrinkage and selection operator penalized Cox regression was adopted to construct a radiomic signature. A radiomic nomogram was established by integrating the radiomic signature and significant clinical risk factors. We also built a preoperative tumor-node-metastasis staging model for comparison. All models were evaluated considering the abilities of risk stratification, discrimination, calibration, and clinical use. In the two validation cohorts, the established four-feature radiomic signature showed robust risk stratification power (P=0.0260 and 0.0003, log-rank test). The radiomic nomogram incorporated radiomic signature, extramural vessel invasion, clinical T stage, and clinical N stage, outperforming all the other models (concordance index=0.720 and 0.727) with good calibration and decision benefits. Also, the 2-yr disease-free survival (DFS) prediction was most effective (time-dependent area under curve=0.771 and 0.765). Moreover, subgroup analysis indicated that the radiomic signature was more sensitive in risk stratifying patients with advanced clinical T/N stage. The proposed MDCT-guided radiomic signature was verified as a prognostic factor for gastric cancer. The radiomic nomogram was a noninvasive auxiliary model for preoperative individualized DFS prediction, holding potential in promoting treatment strategy and clinical prognosis.

A radiomics model for preoperative prediction of brain invasion in meningioma non-invasively based on MRI: A multicentre study

BackgroundPrediction of brain invasion pre-operatively rather than postoperatively would contribute to the selection of surgical techniques, predicting meningioma grading and prognosis. Here, we aimed to predict the risk of brain invasion in meningioma pre-operatively using a nomogram by incorporating radiomic and clinical features.MethodsIn this case-control study, 1728 patients from Beijing Tiantan Hospital (training cohort: n = 1070) and Lanzhou University Second Hospital (external validation cohort: n = 658) were diagnosed with meningiomas by histopathology. Radiomic features were extracted from the T1-weighted post-contrast and T2-weighted magnetic resonance imaging. The least absolute shrinkage and selection operator was used to select the most informative features of different modalities. The support vector machine algorithm was used to predict the risk of brain invasion. Furthermore, a nomogram was constructed by incorporating radiomics signature and clinical risk factors, and decision curve analysis was used to validate the clinical usefulness of the nomogram.FindingsSixteen features were significantly correlated with brain invasion. The clinicoradiomic model derived from the fusing MRI sequences and sex resulted in the best discrimination ability for risk prediction of brain invasion, with areas under the curves (AUCs) of 0•857 (95% CI, 0•831–0•887) and 0•819 (95% CI, 0•775–0•863) and sensitivities of 72•8% and 90•1% in the training and validation cohorts, respectively.InterpretationOur clinicoradiomic model showed good performance and high sensitivity for risk prediction of brain invasion in meningioma, and can be applied in patients with meningiomas.FundingThis work was supported by the 10.13039/501100001809National Natural Science Foundation of China (81772006, 81922040); the 10.13039/501100004739Youth Innovation Promotion Association CAS (grant numbers 2019136); special fund project for doctoral training program of 10.13039/100012899Lanzhou University Second Hospital (grant numbers YJS-BD-33).

Can Machine Learning Methods Produce Accurate and Easy-to-Use Preoperative Prediction Models of One-Year Improvements in Pain and Functioning After Knee Arthroplasty?

BackgroundApproximately 15%-20% of total knee arthroplasty (TKA) patients do not experience clinically meaningful improvements. We sought to compare the accuracy and parsimony of several machine learning strategies for developing predictive models of failing to experience minimal clinically important differences in patient-reported outcome measures (PROMs) 1 year after TKA. MethodsPatients (N = 587) in 3 large Veteran Health Administration facilities completed PROMs before and 1 year after TKA (92% follow-up). Preoperative PROMs and electronic health record data were used to develop and validate models to predict failing to experience at least a minimal clinically important difference in Knee Injury and Osteoarthritis Outcome Score (KOOS) Total, KOOS JR, and KOOS subscales (Pain, Symptoms, Activities of Daily Living, Quality of Life, and recreation). Several machine learning strategies were used for model development. Ten-fold cross-validation and bootstrapping were used to produce measures of overall accuracy (C-statistic, Brier Score). The sensitivity and specificity of various predicted probability cut-points were examined. ResultsThe most accurate models produced were for the Activities of Daily Living, Pain, Symptoms, and Quality of Life subscales of the KOOS (C-statistics 0.76, 0.72, 0.72, and 0.71, respectively). Strategies varied substantially in terms of the numbers of inputs required to achieve similar accuracy, with none being superior for all outcomes. ConclusionModels produced in this project provide estimates of patient-specific improvements in major outcomes 1 year after TKA. Integrating these models into clinical decision support, informed consent and shared decision making could improve patient selection, education, and satisfaction. Level of EvidenceLevel III, diagnostic study.

External validation and modification of nomogram for predicting positive resection margins before breast conserving surgery.

A positive resection margin after breast conserving surgery (BCS) is the most important risk factor for tumor recurrence. In 2012, Seoul National University Hospital (SNUH) breast surgery team developed a nomogram for predicting positive resection margins before BCS to provide individual surgical plans that could reduce local recurrence without increasing re-excision rates. The purpose of this study was to validate this nomogram using an external cohort and to test if addition of surgeon-related factor could improve its use as a predictive model. A total of 419 patients with breast cancer who underwent BCS from January to December 2018 were retrospectively reviewed. Using the SNUH BCS nomogram, risk score for positive resection margins was calculated for 343 patients. The predictive accuracy of the nomogram was assessed, and multivariable logistic regression analyses were performed to evaluate the nomogram's predictive variables. The positive resection margin rate of the current external validation cohort was 13.5% (46 out of 343), compared to 14.6% (151 out of 1034) of the original study. The discrimination power of the SNUH BCS nomogram as measure by area under the receiver operating characteristics curve (AUC) was 0.656 [95% confidence interval (CI) 0.576-0.735]. This result is lower than expected value of 0.823 [95% CI 0.785-0.862], the AUC of the original study. Multivariable logistic regression analysis showed that, among the five nomogram variables, presence of tumor size discrepancy greater than 0.5cm between MRI and ultrasonography (OR 2.445, p = 0.019) and presence of ductal carcinoma in situ on needle biopsy (OR 2.066, p = 0.048) were significantly associated with positive resection margins. Finally, the nomogram score was re-calculated by adding each surgeon's resection margin positive rate as odds ratio and the AUC was increased to 0.733. Validation of the SNUH BCS nomogram was not successful in the current study as much as its original publication. However, we could improve its predictive power by including surgeon-related factor. Before applying a published nomogram as a preoperative predictive model, we suggest each institution to validate the model and adjust it with surgeon-related factor. Addition of new factors to currently available nomograms holds promise for improving its applicability for breast cancer patients at the actual clinical level.

Prediction Models for Recurrence of Chronic Subdural Hematoma in Patients Underwent Twist-Drill Craniostomy Combined With Urokinase Instillation.

The recurrence of chronic subdural hematoma (CSDH) is high post-treatment. In this study, we aimed to construct individualized models for prediction of the postoperative recurrence of CSDH in patients underwent twist-drill craniostomy combined with urokinase (UK) instillation. In total, 183 patients with CSDH were retrospectively enrolled. In summary, 21 candidate factors were retrieved from past medical records. The least absolute shrinkage and selection operator regression was adopted to reduce the high dimensionality of data. Four predictors: preoperative hematoma volume, encephalatrophy, brain re-expansion, and UK instillation frequency were filtered from the 21 candidate factors using the least absolute shrinkage and selection operator method. Binary logistic regression model was employed to establish preoperative and postoperative prediction models. The preoperative model included preoperative hematoma volume and encephalatrophy whereas the postoperative model included brain re-expansion and UK instillation frequency. The predictive performance of the nomograms was evaluated by the receiver operating characteristic curve and calibration chart. Area under curve of the preoperative and postoperative models were 0.755 (95% confidence interval: 0.690-0.889) and 0.782 (95% confidence interval: 0.720-0.936), respectively, indicating good discrimination ability. The calibration results showed good fitting between the predicted probability and the actual probability. Finally, a decision curve analysis revealed excellent clinical performance of the proposed nomograms. Functionally, the preoperative model was used to identify high-risk patients with CSDH and application of UK, while the postoperative model was applied to guide physician-patients communication during follow-up. These 2 prediction models provide a basis for further clinical and experimental studies.

A Practical Pre-Operative Predictive Model for Determining Hamstring Autograft Size for Anterior Cruciate Ligament Reconstruction in Children and Adolescents

Objectives:The most commonly used graft for pediatric and adolescent anterior cruciate ligament (ACL) reconstruction is the hamstring autograft. Recent evidence has suggested that graft sizes less than 8 mm and age less than 20 years old are the most significant factors influencing graft failure. Additionally, studies examining the role of augmenting smaller grafts with allograft have demonstrated mixed results and potentially increased failure rates. Efforts to predict hamstring size preoperatively using MRI and ultrasound in the adult population have shown promise. However, there have been no studies in an entirely pediatric population. The purpose of this study was to evaluate the predictive value of MRI in determining hamstring autograft size preoperatively for pediatric ACL reconstructions.Methods:Children and adolescents from 10-20 years old that had undergone ACL reconstruction using hamstring autograft were identified from 2017-2018. Semitendinosus and gracilis cross-sectional area and diameter were measured on preoperative knee MRIs using the field of interest (FOI) and distance measurement tools. The slice used for measurement was identified as the largest section by the measuring surgeon. Two surgeons independently measured all MRIs. Preoperative demographic data (height, weight, body mass index (BMI), gender) and operative data (graft size) was collected. Data was analyzed by an independent statistician. Ordinal regression analysis was performed to determine the relationship between demographic data and graft size. A Classification and Regression Tree (CART) was constructed to identify predictors for the pre-determined graft sizes of 8 mm. Receiver operating characteristics (ROC) analysis were then created to assess the performance of the predictive model and determine sensitivity and specificity. All analyses were performed using SPSS v. 24 with alpha set at p<0.05 to declare significance.Results:One hundred and ten children (54 boys, 56 girls) were included in this study with an average age of 15.93 years (range 11.81-20.22) and average BMI of 25.98 (range 16.02-40.08). CART analysis determined that if the sum of the cross-sectional areas of the semitendinosus and gracilis tendons were greater than 31.17 mm then 87.5% of children had graft sizes of 8 mm or greater. The ROC analysis determined the model to have an 80% sensitivity and 74% specificity for predicting a graft size of 8 mm or larger.Conclusion:A preoperative summation of the cross-sectional areas of the semitendinosus and gracilis tendons greater than 31.17 mm predicts an intraoperative graft size of 8 mm or larger in 87.5% of children.Figure 1.Table 1:Magnetic resonance imaging (MRI) measurements comparing children with graft sizes less than 8 mm with those 8 mm or greater. 95% Confidence Interval Mean Lower Upper P-value Height (cm)<8162.28159.66164.908+169.02166.56171.47<0.001Weight (kg)<867.3562.2772.438+76.8871.1682.600.014BMI<825.3823.7726.998+26.5725.1228.020.274Age (years)<815.8915.3916.388+15.9715.5716.370.793Semitendinosus<815.8314.8616.81 Cross-sectional Area8+19.1518.1420.16<0.001Semitendinosus Diameter<84.824.664.99 (Anteroposterior)8+5.174.975.370.009Semitendinosus Diameter<84.023.844.21 (Medial/lateral)8+4.494.334.65<0.001Gracillis<89.849.2210.46 Cross-sectional Area8+11.3310.6811.980.001Gracillis Diameter<84.224.094.36 (Anteroposterior)8+4.584.364.800.01Gracillis Diameter<82.872.752.99 (Medial/lateral)8+3.102.973.220.013Semitendinosus + Gracillis<825.6724.3227.02 Cross-sectional Area8+30.4829.0331.93<0.001Diameter Sum<815.9415.5516.338+17.3316.9017.77<0.001

Preoperative Nomogram Predicts Non-home Discharge in Patients Undergoing Pancreatoduodenectomy

BackgroundIn patients undergoing pancreatoduodenectomy, non-home discharge is common and often results in an unnecessary delay in hospital discharge. This study aimed to develop and validate a preoperative prediction model to identify patients with a high likelihood of non-home discharge following pancreatoduodenectomy. MethodsPatients undergoing pancreatoduodenectomy from 2013 to 2018 were identified using an institutional database. Patients were categorized according to discharge location (home vs. non-home). Preoperative risk factors, including social determinants of health associated with non-home discharge, were identified using Pearson’s chi-squared test and then included in a multiple logistic regression model. A training cohort composed of 80% of the sampled patients was used to create the prediction model, and validation carried out using the remaining 20%. Statistical significance was defined as P < 0.05. ResultsSeven hundred sixty-six pancreatoduodenectomy patients met the study criteria for inclusion in the analysis (non-home, 126; home, 640). Independent predictors of non-home discharge on multivariable analysis were age, marital status, mental health diagnosis, functional health status, dyspnea, and chronic obstructive pulmonary disease. The prediction model was then used to generate a nomogram to predict likelihood of non-home discharge. The training and validation cohorts demonstrated comparable performances with an identical area under the curve (0.81) and an accuracy of 84%. ConclusionA prediction model to reliably assess the likelihood of non-home discharge after pancreatoduodenectomy was developed and validated in the present study.

A CT-based Radiomics Model for Prediction of Lymph Node Metastasis in Early Stage Gastric Cancer

To develop and validate a CT-based radiomics model for preoperative prediction of lymph node metastasis (LNM) in early stage gastric cancer (EGC). Four hundred and sixty-three consecutive EGC patients were enrolled in this retrospective study. Radiomics features were extracted from portal venous phase CT scans. A radiomics signature was built based on highly reproducible features using the least absolute shrinkage and selection operator method. The predictive performance of radiomics signature was tested in the training and testing cohorts. Multivariate logistic regression analysis was conducted to build a radiomics-based model combined radiomics signature and lymph node status according to CT. Performance of the model was determined by its discrimination, calibration, and clinical usefulness. The radiomics signature comprised six robust features showed significant association with LNM in both cohorts. A radiomics model that incorporated radiomics signature and CT-reported lymph node status showed good calibration and discrimination in the training cohort (AUC = 0.91) and testing cohort (AUC = 0.89). Decision curve analysis confirmed the clinical utility of this model. The CT-based radiomics model showed favorable accuracy for prediction of LNM in EGC and may help to improve clinical decision-making.

Development and validation of a preoperative noninvasive predictive model based on circular tumor DNA for lymph node metastasis in resectable non-small cell lung cancer.

BackgroundClinical lymph node staging in resectable non-small cell lung cancer (NSCLC) patients not only indicates prognosis, but also determines primary treatment strategy. The demand of noninvasive tool for preoperative lymph node metastasis prediction remains significant. This study aimed to develop and externally validate a preoperative noninvasive predictive model based on circular tumor DNA (ctDNA) for the lymph node metastasis in resectable NSCLC patients.MethodsResectable NSCLC patients in TRACERx cohort were included as training group. Potential preoperative noninvasively accessible predictors were incorporated into the development of a nomogram via multivariate logistic regression. The predictive model was externally validated by a similar cohort from our hospital.ResultsOverall, 58 patients from TRACERx cohort were included as training group and 37 patients from our hospital were included as external validation group. Variant allele frequency (VAF) level of ctDNA was significantly associated with lymph node metastasis (OR: 4.89, 95% CI: 1.22–19.54, P=0.03). The predictive model incorporating age, tumor size and VAF demonstrated satisfactory discrimination and calibration in both training group (AUC =0.77, 95% CI: 0.65–0.90, P=0.001) and external validation group (AUC =0.84, 95% CI: 0.70–0.99, P=0.005).ConclusionsHigh VAF level in preoperative ctDNA may indicate lymph node metastasis of resectable NSCLC. And a preoperative noninvasive predictive model based on ctDNA for the lymph node metastasis in resectable NSCLC patients was developed and externally validated with satisfactory discrimination and calibration.

Preoperative Predictive Model of Narrow Pelvis in Laparoscopic Radical Prostatectomy Through Computed Tomography.

Objective: At present, prostate cancer radical surgery still lacks an objective standard for predicting the anatomic difficulty of surgery through preoperative examination. Due to the poor prognosis and surgical complications caused by "Narrow Pelvis," a pelvic model for prediction is urgently needed. The purpose of this study is to present objective preoperative assessing indicators to predict the surgical difficulty caused by pelvic anatomic abnormalities during laparoscopic radical prostatectomy (LRP). Methods: Patients undergoing LRP were retrospectively analyzed and separated into "Common Pelvis" or "Difficult Pelvis" groups according to the preset criteria. The clinical data and pelvic imaging data of the two groups were compared to obtain statistically significant differences. The results were compared and validated in 500 ordinary males. Logistic regression analysis was performed to optimize these indicators into a scoring model, "Laparoscopic Radical Prostatectomy Narrow Pelvic Predictive Index (LRP-NPPI)" for predicting "Difficult Pelvis." Results: Of the 145 patients undergoing LRP, 22 (15.2%) were included in the "Difficult Pelvis" group. Patients in the "Difficult Pelvis" group were more likely to have a narrower, shorter, deeper, more flattened pelvis, greater body mass index, more history of previous pelvic surgery, and greater prostate volume. Moreover, the larger amount of intraoperative blood loss and longer operation time were related. Compared with the "Common Pelvis" group, patients in the "Difficult Pelvis" have higher score of LRP-NPPI (p < 0.05). Conclusion: With the model we proposed, it is possible to predict patients with pelvic anatomical difficulties during LRP.

A radiomics-based model for prediction of lymph node metastasis in gastric cancer

To develop and validate a radiomics-based model for preoperative prediction of lymph node metastasis (LNM) in gastric cancer (GC). A total of 768 GC patients were enrolled in this retrospective study. Radiomics features were extracted from portal venous phase computed tomography (CT) scans. A radiomics signature was built with highly reproducible features using the least absolute shrinkage and selection operator (LASSO) method in the training cohort (n = 486). The signature was further validated in internal validation (n = 240) and external testing cohorts (n = 42). Multivariate logistic regression analysis was conducted to build a model that combined radiomics signature, serum biomarkers, and lymph node status according to CT. Performance of the model was determined by its discrimination, calibration, and clinical usefulness. The predictive value of the model was also evaluated in early stage GC (EGC) subgroup. The radiomics signature comprised 7 robust features showed favorable prediction efficacy in all cohorts. A radiomics-based model that incorporated radiomics signature, serum CA72-4, and CT-reported lymph node status had good calibration and discrimination in training cohort [AUC, 0.92; 95% confidence interval (CI), 0.89-0.95] and validation cohort (AUC 0.86; 95% CI, 0.81-0.91). The model also showed a favorable predictive performance for EGC patients with an AUC of 0.85 (95% CI, 0.76-0.94). Decision curve analysis confirmed the clinical utility of this model. The radiomics-based model showed favorable accuracy for prediction of LNM in GC. The model may also serve as a noninvasive tool for preoperative evaluation of LNM in EGC.

Preoperative Prediction of Cervical Lymph Node Metastasis in Papillary Thyroid Carcinoma via Conventional and Contrast-Enhanced Ultrasound.

To investigate the independent risk factors for cervical lymph node metastasis (CLNM) in patients with papillary thyroid carcinoma (PTC) and establish a prediction model via conventional ultrasound (US) and contrast-enhanced ultrasound (CEUS). A total of 275 patients with single PTC, who were admitted to our hospital from January 2016 to December 2018, were enrolled in this study. The clinical characteristics and features of the tumor on conventional US and CEUS were retrospectively analyzed. A binary logistic regression model was established, and the diagnostic efficacy of conventional US and CEUS was compared. The binary logistic regression analysis showed that age younger than 38 years, size of 10.0 mm or greater, solid composition, peak of the nodule interior of 28.3750 or greater, and area under the curve (AUC) of the peripheral ring of less than 3.2500 were independent risk factors of CLNM (P < .05, for all). Prediction model: Logit (p) = -4.135 + 1.040 (age) + 1.920 (size) + 1.769 (composition) + 1.230 (peak of nodule interior) + 0.812 (AUC of peripheral ring). The model positively predicted CLNM at a value of L > -0.199. The AUC, sensitivity, and negative predictive value of the receiver operating characteristic curve for the model were 0.727, 71.7%, and 75.2%, respectively, which were significantly higher than those of conventional US (Z = -2.403; P = .016; Z = -5.330; P < .001; and Z = -2.390; P = .017). The specificity of the model was 73.7%, which was lower than that of conventional US (Z = 3.508; P < .001). The preoperative prediction model established via conventional US and CEUS may be helpful to evaluate CLNM in patients with PTC and determine the appropriate treatment options.