Noninvasive prediction of perineural invasion in intrahepatic cholangiocarcinoma by clinicoradiological features and computed tomography radiomics based on interpretable machine learning: a multicenter cohort study.

BACKGROUND: Perineural invasion (PNI) of intrahepatic cholangiocarcinoma (ICC) is a strong independent risk factor for tumour recurrence and long-term patient survival. However, there is a lack of noninvasive tools for accurately predicting the PNI status. The authors develop and validate a combined model incorporating radiomics signature and clinicoradiological features based on machine learning for predicting PNI in ICC, and used the Shapley Additive explanation (SHAP) to visualize the prediction process for clinical application. METHODS: This retrospective and prospective study included 243 patients with pathologically diagnosed ICC (training, n =136; external validation, n =81; prospective, n =26, respectively) who underwent preoperative contrast-enhanced computed tomography between January 2012 and May 2023 at three institutions (three tertiary referral centres in Guangdong Province, China). The ElasticNet was applied to select radiomics features and construct signature derived from computed tomography images, and univariate and multivariate analyses by logistic regression were used to identify the significant clinical and radiological variables with PNI. A robust combined model incorporating radiomics signature and clinicoradiological features based on machine learning was developed and the SHAP was used to visualize the prediction process. A Kaplan-Meier survival analysis was performed to compare prognostic differences between PNI-positive and PNI-negative groups and was conducted to explore the prognostic information of the combined model. RESULTS: Among 243 patients (mean age, 61.2 years ± 11.0 (SD); 152 men and 91 women), 108 (44.4%) were diagnosed as PNI-positive. The radiomics signature was constructed by seven radiomics features, with areas under the curves of 0.792, 0.748, and 0.729 in the training, external validation, and prospective cohorts, respectively. Three significant clinicoradiological features were selected and combined with radiomics signature to construct a combined model using machine learning. The eXtreme Gradient Boosting exhibited improved accuracy and robustness (areas under the curves of 0.884, 0.831, and 0.831, respectively). Survival analysis showed the construction combined model could be used to stratify relapse-free survival (hazard ratio, 1.933; 95% CI: 1.093-3.418; P =0.021). CONCLUSIONS: We developed and validated a robust combined model incorporating radiomics signature and clinicoradiological features based on machine learning to accurately identify the PNI statuses of ICC, and visualize the prediction process through SHAP for clinical application.

A Preliminary Study of Brain Functional Magnetic Resonance Imaging in Text Reading and Comprehension.

BACKGROUND: Few studies have focused on the changes in human brain function activities caused by reading Chinese characters with different intelligibility and whether it can reflect the understanding and cognitive ability of the human brain. OBJECTIVE: Task-fMRI based on Chinese character reading tasks with different intelligibility was used to explore activated brain regions and their cognitive changes. METHODS: Volunteers were randomly recruited using advertisements. Forty volunteers were recruited based on strict inclusion and exclusion criteria, and 40 volunteers were recruited. Brain function data of 40 healthy right-handed volunteers in fuzzy/clear Chinese reading tasks were collected using a Siemens Skyra 3.0T magnetic resonance scanner. Data were preprocessed and statistically analyzed using the statistical software SPM12.0 to observe the activation of the cortex and analyze its characteristics and possible changes in cognitive function. RESULTS: Task-fMRI analysis: (1) The main brain regions activated in fuzzy/clear reading tasks were located in the occipital visual cortex (P < 0.001); (2) a paired sample t-test suggested that there was a significant difference in BOLD signals in the brain regions activated by fuzzy/clear reading tasks (P < 0.001, equiv Z = 4.25). Compared with the fuzzy reading task, the brain regions more strongly activated in the clear reading task were mainly located in the right superior frontal gyrus and the bilateral temporal lobe. Compared with the clear reading task, the brain region that was more strongly activated in the fuzzy reading task was mainly located in the right fusiform gyrus. CONCLUSION: Clear Chinese character information mainly activates the dorsal stream of the visual-spatial network. This reflects the information transmission of the brain after understanding the text content and is responsible for guiding and controlling attention. Fuzzy words that cannot provide clear text content activate the fusiform gyrus of the ventral stream of the visual-spatial network, strengthening the function of orthographic processing.

Exploration of the correlation between superficial cerebral veins identified using susceptibility-weighted imaging findings and cognitive differences between sexes based on deep learning: a preliminary study.

Background: This study aimed to investigate the association of superficial cerebral veins (SCVs) with sex-related cognitive differences and the possible hemodynamic mechanisms underlying these associations. Methods: This investigation was a prospective case-control study. A total of 344 healthy volunteers were recruited. In all, 200 volunteers were included to establish the deep learning model, and 144 volunteers were used for the research, including 72 males (50%) and 72 females (50%). No significant differences in age (P=0.358) or education (P=0.779) were observed between the sexes. Cognitive functioning was evaluated using neuropsychological tests, including the Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment-Basic (MOCA-B). Susceptibility-weighted imaging scans were acquired with a 3.0 T magnetic resonance imaging system using a 32-channel high-resolution phased array coil. Minimum intensity projection images were obtained by reconstructing susceptibility-weighted imaging data. A deep learning model was trained on the minimum intensity projection images to quantify the diameter, tortuosity index, length, and the number of SCVs in the bilateral cerebral hemispheres. Finally, the association between cognitive differences between males and females and the properties of the SCVs was analyzed. Results: The MMSE and MOCA-B scores of males were significantly higher than those of females (P<0.05). Males had more SCVs in the bilateral cerebral hemispheres than did females (right hemisphere: P<0.01; left hemisphere: P<0.05). The number of SCVs in the right cerebral hemisphere was significantly and positively correlated with the MMSE and MOCA-B scores (correlation coefficients: 0.246 and 0.201, respectively; P<0.05). The number of SCVs in the left cerebral hemisphere was positively correlated with the MMSE scores (correlation coefficient: 0.196; P<0.05) and the MOCA-B scores. In this study, no significant correlations were observed between cognition and the diameter, length, or tortuosity index of the SCVs in the bilateral cerebral hemispheres. Conclusions: The cognitive function of males was better than that of females, and the different numbers of SCVs may be one of the explanations for this phenomenon of sex-based differences in cognition.

Diethyl 4-(4,5-dihydro-furan-2-yl)-3,5-di-methyl-1-phenyl-1,4-dihydro-pyrazine-2,6-dicarboxyl-ate.

In the title compound, C(22)H(26)N(2)O(5), the central 1,4-dihydro-pyrazine ring adopts a boat conformation, while the benzene ring and the two disordered components of the furan ring are inclined at angles of 77.9 (5) and 61.9 (7)°. Three of the C atoms of the furan ring are disordered over two positions with occupancies of 0.655 (18) and 0.345 (18). In the crystal structure, weak inter-molecular C-H⋯O hydrogen bonds link the mol-ecules into chains propagating in [010].