Preoperative prediction of perineural invasion of rectal cancer based on a magnetic resonance imaging radiomics model: A dual-center study.

BACKGROUND: Perineural invasion (PNI) has been used as an important pathological indicator and independent prognostic factor for patients with rectal cancer (RC). Preoperative prediction of PNI status is helpful for individualized treatment of RC. Recently, several radiomics studies have been used to predict the PNI status in RC, demonstrating a good predictive effect, but the results lacked generalizability. The preoperative prediction of PNI status is still challenging and needs further study. AIM: To establish and validate an optimal radiomics model for predicting PNI status preoperatively in RC patients. METHODS: This retrospective study enrolled 244 postoperative patients with pathologically confirmed RC from two independent centers. The patients underwent pre-operative high-resolution magnetic resonance imaging (MRI) between May 2019 and August 2022. Quantitative radiomics features were extracted and selected from oblique axial T2-weighted imaging (T2WI) and contrast-enhanced T1WI (T1CE) sequences. The radiomics signatures were constructed using logistic regression analysis and the predictive potential of various sequences was compared (T2WI, T1CE and T2WI + T1CE fusion sequences). A clinical-radiomics (CR) model was established by combining the radiomics features and clinical risk factors. The internal and external validation groups were used to validate the proposed models. The area under the receiver operating characteristic curve (AUC), DeLong test, net reclassification improvement (NRI), integrated discrimination improvement (IDI), calibration curve, and decision curve analysis (DCA) were used to evaluate the model performance. RESULTS: Among the radiomics models, the T2WI + T1CE fusion sequences model showed the best predictive performance, in the training and internal validation groups, the AUCs of the fusion sequence model were 0.839 [95% confidence interval (CI): 0.757-0.921] and 0.787 (95%CI: 0.650-0.923), which were higher than those of the T2WI and T1CE sequence models. The CR model constructed by combining clinical risk factors had the best predictive performance. In the training and internal and external validation groups, the AUCs of the CR model were 0.889 (95%CI: 0.824-0.954), 0.889 (95%CI: 0.803-0.976) and 0.894 (95%CI: 0.814-0.974). Delong test, NRI, and IDI showed that the CR model had significant differences from other models (P < 0.05). Calibration curves demonstrated good agreement, and DCA revealed significant benefits of the CR model. CONCLUSION: The CR model based on preoperative MRI radiomics features and clinical risk factors can preoperatively predict the PNI status of RC noninvasively, which facilitates individualized treatment of RC patients.

The white matter characteristic of the genu of corpus callosum coupled with pain intensity and negative emotion scores in patients with trigeminal neuralgia: a multivariate analysis.

Background: Trigeminal neuralgia (TN) is a chronic neuropathic pain disorder that not only causes intense pain but also affects the psychological health of patients. Since TN pain intensity and negative emotion may be grounded in our own pain experiences, they exhibit huge inter-individual differences. This study investigates the effect of inter-individual differences in pain intensity and negative emotion on brain structure in patients with TN and the possible pathophysiology mechanism underlying this disease. Methods: T1 weighted magnetic resonance imaging and diffusion tensor imaging scans were obtained in 46 patients with TN and 35 healthy controls. All patients with TN underwent pain-related and emotion-related questionnaires. Voxel-based morphometry and regional white matter diffusion property analysis were used to investigate whole brain grey and white matter quantitatively. Innovatively employing partial least squares correlation analysis to explore the relationship among pain intensity, negative emotion and brain microstructure in patients with TN. Results: Significant difference in white matter integrity were identified in patients with TN compared to the healthy controls group; The most correlation brain region in the partial least squares correlation analysis was the genus of the corpus callosum, which was negatively associated with both pain intensity and negative emotion. Conclusion: The genu of corpus callosum plays an important role in the cognition of pain perception, the generation and conduction of negative emotions in patients with TN. These findings may deepen our understanding of the pathophysiology of TN.

Radiomics nomogram based on MRI water imaging identifying symptomatic nerves of patients with primary trigeminal neuralgia: A preliminary study.

The study proposes a combined nomogram based on radiomics features from magnetic resonance neurohydrography and clinical features to identify symptomatic nerves in patients with primary trigeminal neuralgia. We retrospectively analyzed 140 patients with clinically confirmed trigeminal neuralgia. Out of these, 24 patients constituted the external validation set, while the remaining 116 patients contributed a total of 231 nerves, comprising 118 symptomatic nerves, and 113 normal nerves. Radiomics features were extracted from the MRI water imaging (t2-mix3d-tra-spair). Radiomics feature selection was performed using L1 regularization-based regression, while clinical feature selection utilized univariate analysis and multivariate logistic regression. Subsequently, radiomics, clinical, and combined models were developed by using multivariate logistic regression, and a nomogram of the combined model was drawn. The performance of nomogram in discriminating symptomatic nerves was assessed through the area under the curve (AUC) of receiver operating characteristics, accuracy, and calibration curves. Clinical applications of the nomogram were further evaluated using decision curve analysis. Five clinical factors and 13 radiomics signatures were ultimately selected to establish predictive models. The AUCs in the training and validation cohorts were 0.77 (0.70-0.84) and 0.82 (0.72-0.92) with the radiomics model, 0.69 (0.61-0.77) and 0.66 (0.53-0.79) with the clinical model, 0.80 (0.74-0.87), and 0.85 (0.76-0.94) with the combined model, respectively. In the external validation set, the AUCs for the clinical, radiomics, and combined models were 0.70 (0.60-0.79), 0.78 (0.65-0.91), and 0.81 (0.70-0.93), respectively. The calibration curve demonstrated that the nomogram exhibited good predictive ability. Moreover, The decision curve analysis curve indicated shows that the combined model holds high clinical application value. The integrated model, combines radiomics features from magnetic resonance neurohydrography with clinical factors, proves to be effective in identify symptomatic nerves in trigeminal neuralgia. The diagnostic efficacy of the combined model was notably superior to that of the model constructed solely from conventional clinical features.

Deep learning-driven MRI trigeminal nerve segmentation with SEVB-net.

Purpose: Trigeminal neuralgia (TN) poses significant challenges in its diagnosis and treatment due to its extreme pain. Magnetic resonance imaging (MRI) plays a crucial role in diagnosing TN and understanding its pathogenesis. Manual delineation of the trigeminal nerve in volumetric images is time-consuming and subjective. This study introduces a Squeeze and Excitation with BottleNeck V-Net (SEVB-Net), a novel approach for the automatic segmentation of the trigeminal nerve in three-dimensional T2 MRI volumes. Methods: We enrolled 88 patients with trigeminal neuralgia and 99 healthy volunteers, dividing them into training and testing groups. The SEVB-Net was designed for end-to-end training, taking three-dimensional T2 images as input and producing a segmentation volume of the same size. We assessed the performance of the basic V-Net, nnUNet, and SEVB-Net models by calculating the Dice similarity coefficient (DSC), sensitivity, precision, and network complexity. Additionally, we used the Mann-Whitney U test to compare the time required for manual segmentation and automatic segmentation with manual modification. Results: In the testing group, the experimental results demonstrated that the proposed method achieved state-of-the-art performance. SEVB-Net combined with the ωDoubleLoss loss function achieved a DSC ranging from 0.6070 to 0.7923. SEVB-Net combined with the ωDoubleLoss method and nnUNet combined with the DoubleLoss method, achieved DSC, sensitivity, and precision values exceeding 0.7. However, SEVB-Net significantly reduced the number of parameters (2.20 M), memory consumption (11.41 MB), and model size (17.02 MB), resulting in improved computation and forward time compared with nnUNet. The difference in average time between manual segmentation and automatic segmentation with manual modification for both radiologists was statistically significant (p < 0.001). Conclusion: The experimental results demonstrate that the proposed method can automatically segment the root and three main branches of the trigeminal nerve in three-dimensional T2 images. SEVB-Net, compared with the basic V-Net model, showed improved segmentation performance and achieved a level similar to nnUNet. The segmentation volumes of both SEVB-Net and nnUNet aligned with expert annotations but SEVB-Net displayed a more lightweight feature.