Machine learning model to preoperatively predict T2/T3 staging of laryngeal and hypopharyngeal cancer based on the CT radiomic signature.

OBJECTIVES: To develop and assess a radiomics-based prediction model for distinguishing T2/T3 staging of laryngeal and hypopharyngeal squamous cell carcinoma (LHSCC) METHODS: A total of 118 patients with pathologically proven LHSCC were enrolled in this retrospective study. We performed feature processing based on 851 radiomic features derived from contrast-enhanced CT images and established multiple radiomic models by combining three feature selection methods and seven machine learning classifiers. The area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity were used to assess the performance of the models. The radiomic signature obtained from the optimal model and statistically significant morphological image characteristics were incorporated into the predictive nomogram. The performance of the nomogram was assessed by calibration curve and decision curve analysis. RESULTS: Using analysis of variance (ANOVA) feature selection and logistic regression (LR) classifier produced the best model. The AUCs of the training, validation, and test sets were 0.919, 0.857, and 0.817, respectively. A nomogram based on the model integrating the radiomic signature and a morphological imaging characteristic (suspicious thyroid cartilage invasion) exhibited C-indexes of 0.899 (95% confidence interval (CI) 0.843-0.955), fitting well in calibration curves (p > 0.05). Decision curve analysis further confirmed the clinical usefulness of the nomogram. CONCLUSIONS: The nomogram based on the radiomics model derived from contrast-enhanced CT images had good diagnostic performance for distinguishing T2/T3 staging of LHSCC. CLINICAL RELEVANCE STATEMENT: Accurate T2/T3 staging assessment of LHSCC aids in determining whether laryngectomy or laryngeal preservation therapy should be performed. The nomogram based on the radiomics model derived from contrast-enhanced CT images has the potential to predict the T2/T3 staging of LHSCC, which can provide a non-invasive and robust approach for guiding the optimization of clinical decision-making. KEY POINTS: • Combining analysis of variance with logistic regression yielded the optimal radiomic model. • A nomogram based on the CT-radiomic signature has good performance for differentiating T2 from T3 staging of laryngeal and hypopharyngeal squamous cell carcinoma. • It provides a non-invasive and robust approach for guiding the optimization of clinical decision-making.

Abnormal neonatal brain microstructure in gestational diabetes mellitus revealed by MRI texture analysis.

To investigate the value of MRI texture analysis in evaluating the effect of gestational diabetes mellitus (GDM) on neonatal brain microstructure development, we retrospectively collected images of neonates undergoing head MRI scans, including a GDM group (N1 = 37) and a healthy control group (N2 = 34). MaZda texture analysis software was used to extract the texture features from different sequence images and perform dimensionality reduction, and then the texture features selected by the lowest misjudgement rate method were imported into SPSS software for statistical analysis. In our study, we found that GDM affects the development of the microstructure of the neonatal brain, and different combinations of texture features have different recognition performances, such as different sequences and different brain regions. As a consequence, texture analysis combining multiple conventional MRI sequences has a high recognition performance in revealing the abnormal development of the brain microstructure of neonates born of mothers with GDM.