Multiparametric MRI-Based Interpretable Radiomics Machine Learning Model Differentiates Medulloblastoma and Ependymoma in Children: A Two-Center Study.

RATIONALE AND OBJECTIVES: Medulloblastoma (MB) and Ependymoma (EM) in children, share similarities in age group, tumor location, and clinical presentation. Distinguishing between them through clinical diagnosis is challenging. This study aims to explore the effectiveness of using radiomics and machine learning on multiparametric magnetic resonance imaging (MRI) to differentiate between MB and EM and validate its diagnostic ability with an external set. MATERIALS AND METHODS: Axial T2 weighted image (T2WI) and contrast-enhanced T1weighted image (CE-T1WI) MRI sequences of 135 patients from two centers were collected as train/test sets. Volume of interest (VOI) was manually delineated by an experienced neuroradiologist, supervised by a senior. Feature selection analysis and the least absolute shrinkage and selection operator (LASSO) algorithm identified valuable features, and Shapley additive explanations (SHAP) evaluated their significance. Five machine-learning classifiers-extreme gradient boosting (XGBoost), Bernoulli naive Bayes (Bernoulli NB), Logistic Regression (LR), support vector machine (SVM), linear support vector machine (Linear SVC) classifiers were built based on T2WI (T2 model), CE-T1WI (T1 model), and T1 + T2WI (T1 + T2 model). A human expert diagnosis was developed and corrected by senior radiologists. External validation was performed at Sun Yat-Sen University Cancer Center. RESULTS: 31 valuable features were extracted from T2WI and CE-T1WI. XGBoost demonstrated the highest performance with an area under the curve (AUC) of 0.92 on the test set and maintained an AUC of 0.80 during external validation. For the T1 model, XGBoost achieved the highest AUC of 0.85 on the test set and the highest accuracy of 0.71 on the external validation set. In the T2 model, XGBoost achieved the highest AUC of 0.86 on the test set and the highest accuracy of 0.82 on the external validation set. The human expert diagnosis had an AUC of 0.66 on the test set and 0.69 on the external validation set. The integrated T1 + T2 model achieved an AUC of 0.92 on the test set, 0.80 on the external validation set, achieved the best performance. Overall, XGBoost consistently outperformed in different classification models. CONCLUSION: The combination of radiomics and machine learning on multiparametric MRI effectively distinguishes between MB and EM in childhood, surpassing human expert diagnosis in training and testing sets.

Differentiation between cerebral alveolar echinococcosis and brain metastases with radiomics combined machine learning approach.

BACKGROUND: Cerebral alveolar echinococcosis (CAE) and brain metastases (BM) share similar in locations and imaging appearance. However, they require distinct treatment approaches, with CAE typically treated with chemotherapy and surgery, while BM is managed with radiotherapy and targeted therapy for the primary malignancy. Accurate diagnosis is crucial due to the divergent treatment strategies. PURPOSE: This study aims to evaluate the effectiveness of radiomics and machine learning techniques based on magnetic resonance imaging (MRI) to differentiate between CAE and BM. METHODS: We retrospectively analyzed MRI images of 130 patients (30 CAE and 100 BM) from Xinjiang Medical University First Affiliated Hospital and The First People's Hospital of Kashi Prefecture, between January 2014 and December 2022. The dataset was divided into training (91 cases) and testing (39 cases) sets. Three dimensional tumors were segmented by radiologists from contrast-enhanced T1WI images on open resources software 3D Slicer. Features were extracted on Pyradiomics, further feature reduction was carried out using univariate analysis, correlation analysis, and least absolute shrinkage and selection operator (LASSO). Finally, we built five machine learning models, support vector machine, logistic regression, linear discrimination analysis, k-nearest neighbors classifier, and Gaussian naïve bias and evaluated their performance via several metrics including sensitivity (recall), specificity, positive predictive value (precision), negative predictive value, accuracy and the area under the curve (AUC). RESULTS: The area under curve (AUC) of support vector classifier (SVC), linear discrimination analysis (LDA), k-nearest neighbors (KNN), and gaussian naïve bias (NB) algorithms in training (testing) sets are 0.99 (0.94), 1.00 (0.87), 0.98 (0.92), 0.97 (0.97), and 0.98 (0.93), respectively. Nested cross-validation demonstrated the robustness and generalizability of the models. Additionally, the calibration plot and decision curve analysis demonstrated the practical usefulness of these models in clinical practice, with lower bias toward different subgroups during decision-making. CONCLUSION: The combination of radiomics and machine learning approach based on contrast enhanced T1WI images could well distinguish CAE and BM. This approach holds promise in assisting doctors with accurate diagnosis and clinical decision-making.

Development and validation of a novel nomogram to predict the risk of the prolonged postoperative length of stay for lumbar spinal stenosis patients.

BACKGROUND: Lumber spinal stenosis (LSS) is the increasingly reason for spine surgery for elder patients since China is facing the fastest-growing aging population. The aim of this research was to create a model to predict the probabilities of requiring a prolonged postoperative length of stay (PLOS) for lumbar spinal stenosis patients, minimizing the healthcare burden. METHODS: A total of 540 LSS patients were enrolled in this project. The outcome was a prolonged PLOS after spine surgery, defined as hospitalizations ≥ 75th percentile for PLOS, including the day of discharge. The least absolute shrinkage and selection operator (LASSO) was used to identify independent risk variables related to prolonged PLOS. Multivariable logistic regression analysis was utilized to generate a prediction model utilizing the variables employed in the LASSO approach. The receiver operating characteristic (ROC) curve's area under the curve (AUC) and the calibration curve's respective curves were used to further validate the model's calibration with predictability and discriminative capabilities. By using decision curve analysis, the resulting model's clinical effectiveness was assessed. RESULTS: Among 540 individuals, 344 had PLOS that was within the usual range of P75 (8 days), according to the interquartile range of PLOS, and 196 had PLOS that was above the normal range of P75 (prolonged PLOS). Four variables were incorporated into the predictive model, named: transfusion, operation duration, blood loss and involved spine segments. A great difference in clinical scores can be found between the two groups (P < 0.001). In the development set, the model's AUC for predicting prolonged PLOS was 0.812 (95% CI: 0.768-0.859), while in the validation set, it was 0.830 (95% CI: 0.753-0.881). The calibration plots for the probability showed coherence between the expected probability and the actual probability both in the development set and validation set respectively. When intervention was chosen at the potential threshold of 2%, analysis of the decision curve revealed that the model was more clinically effective. CONCLUSIONS: The individualized prediction nomogram incorporating five common clinical features for LSS patients undergoing surgery can be suitably used to smooth early identification and improve screening of patients at higher risk of prolonged PLOS and minimize health care.

Development and validation of a diagnostic model for differentiating tuberculous spondylitis from brucellar spondylitis using machine learning: A retrospective cohort study.

Background: Tuberculous spondylitis (TS) and brucellar spondylitis (BS) are commonly observed in spinal infectious diseases, which are initially caused by bacteremia. BS is easily misdiagnosed as TS, especially in underdeveloped regions of northwestern China with less sensitive medical equipment. Nevertheless, a rapid and reliable diagnostic tool remains to be developed and a clinical diagnostic model to differentiate TS and BS using machine learning algorithms is of great significance. Methods: A total of 410 patients were included in this study. Independent factors to predict TS were selected by using the least absolute shrinkage and selection operator (LASSO) regression model, permutation feature importance, and multivariate logistic regression analysis. A TS risk prediction model was developed with six different machine learning algorithms. We used several metrics to evaluate the accuracy, calibration capability, and predictability of these models. The performance of the model with the best predictability was further verified with the area under the curve (AUC) of the receiver operating characteristic (ROC) curve and the calibration curve. The clinical performance of the final model was evaluated by decision curve analysis. Results: Six variables were incorporated in the final model, namely, pain severity, CRP, x-ray intervertebral disc height loss, x-ray endplate sclerosis, CT vertebral destruction, and MRI paravertebral abscess. The analysis of appraising six models revealed that the logistic regression model developed in the current study outperformed other methods in terms of sensitivity (0.88 ± 0.07) and accuracy (0.79 ± 0.07). The AUC of the logistic regression model predicting TS was 0.86 (95% CI, 0.81-0.90) in the training set and 0.86 (95% CI, 0.78-0.92) in the validation set. The decision curve analysis indicated that the logistic regression model displayed a higher clinical efficiency in the differential diagnosis. Conclusions: The logistic regression model developed in this study outperformed other methods. The logistic regression model demonstrated by a calculator exerts good discrimination and calibration capability and could be applicable in differentiating TS from BS in primary health care diagnosis.