BMC Medical Imaging最新文献

Background: Lymphovascular invasion (LVI) is a critical prognostic factor in bladder cancer, affecting recurrence, survival, and overall prognosis. Traditional methods for diagnosing LVI, such as immunohistochemical staining, are costly and time-consuming, making non-invasive alternatives like radiomics-based models valuable. This study aimed to construct an interpretable machine learning model to predict LVI status and survival outcomes in patients with bladder urothelial carcinoma using preoperative CT images.

Methods: This study retrospectively enrolled patients with urothelial carcinoma who underwent radical cystectomy and preoperative contrast-enhanced CT from three medicine centers. Tumor regions were manually segmented, and radiomics features were extracted and selected through reproducibility testing, correlation analysis, and LASSO. Based on the selected radiomics features, machine learning classifiers, including SVM, were trained using five-fold cross-validation. A combined model was then constructed by integrating the radiomics signature with clinical risk factors. Model performance was evaluated by AUC, ACC, sensitivity, specificity, and survival analysis.

Results: The SVM model showed high performance, with an AUC of 0.944 in the training set and 0.872 in the testing set. The combined model integrating clinical factor performed better, achieving an AUC of 0.952 in the training set and 0.901 in the testing set. The model's interpretability was enhanced using SHAP analysis, identifying key radiomics features associated with LVI, such as tumor shape and texture. Survival analysis indicated that patients predicted to be LVI-negative had significantly better disease-free survival compared to patients predicted to be LVI-positive.

Conclusions: This multicenter study demonstrates that the interpretable machine learning model based on preoperative CT images can effectively predict LVI status and survival outcomes in bladder urothelial carcinoma.

Trial registration: This study was retrospectively registered by Ethics Committee of the First Affiliated Hospital of Chongqing Medical University (Approval No. K2024-187-01) on April 12, 2024, and informed consent was waived.

Background: Assessing radiographic parameters in pediatric pelvic X-rays is crucial for evaluating hip development, yet existing deep learning (DL)-based methods lack both age-specific reliability analysis and a comprehensive solution for measuring multiple key parameters.

Methods: This retrospective study developed and validated a DL-based system using separate, nonoverlapping datasets of 1495 and 1300 anteroposterior (AP) pelvic radiographs of normal Korean children for model training and evaluation, respectively. The system measured the acetabular index (AcI), Shenton line (ShL), pelvic rotation index (PRI), and pelvic tilt index (PTI). Subgroup analyses were conducted to evaluate the effects of age-related pelvic bone development. Evaluation metrics included the intraclass correlation coefficient (ICC), Pearson correlation coefficient (r), mean absolute error (MAE), root mean square error (RMSE), Hausdorff distance (HD), and Frechet distance (FD). Agreement between the system's and clinician's measurements was assessed using Bland-Altman analysis.

Results: For all evaluation data, automatically measured AcI, PRI, PTI, and ShL values strongly matched and correlated with radiologist-assessed values (AcI: ICC = 0.89, r = 0.91, MAE = 2.07°, RMSE = 2.99°; PRI: ICC = 0.94, r = 0.94, MAE = 0.03, RMSE = 0.04; PTI: ICC = 0.97, r = 0.97, MAE = 0.04, RMSE = 0.09; ShL: HD = 3.62 mm, FD = 2.27 mm). The subgroup analysis revealed that the system's performance varied with age-related differences in pelvic bone development.

Conclusion: The DL-based system exhibited high reliability and accuracy in measuring radiographic parameters for differentiating normal from dislocated hips and assessing pelvic radiograph quality.

Aim: To investigate the dynamic functional connectivity (DFC) pattern in diverse severity of white matter hyperintensity (WMH) and explore the relationship between DFC temporal properties and cognitive impairment in WMH severity.

Methods: Our study enrolled 85 CSVD patients diagnosed by the Neurology Department of the Fifth People's Hospital of Shanghai, Fudan University. All participants underwent demographic surveys, assessments of vascular risk factors, neuropsychological testing, both structural and resting-state functional MRI scans. To explore the influence of WMH severity on cognition, subjects were categorized into two groups based on their WMH Fazekas rating scale scores: (1)mild WMH group scored 1-2 (n = 55); (2)severe WMH group scored 3-6 (n = 30). We calculated DFC by using sliding window correlation approach. Subsequently, we employed k-means clustering to identify distinct DFC states and calculated DFC temporal properties (including mean dwell time, fractional windows and transition numbers).

Results: The intrinsic brain functional connectivity of both mild and severe WMH subjects was clustered into four distinct connectivity states (state 1: intermediate pattern, state 2: a more frequent, sparsely connected pattern, state 3: intermediate pattern, state 4: a less frequent, strongly connected pattern). Compared to mild WMH group, severe WMH subjects dwelled relatively longer in hypoconnected state 2, and shorter in hyperconnected state 4. Across the whole WMH subjects, mean dwell time and fractional windows of hypoconnected state 2 showed negative exploratory correlations with executive function, whereas the corresponding metrics of hyperconnected state 4 showed positive exploratory correlations before multiple comparison correction. Additionally, transition numbers demonstrated a negative correlation with the WMH Fazekas rating scale scores.

Conclusion: The severity of WMH affects DFC temporal properties to a certain extent, and exploratory correlations suggest a possible link to cognition. These exploratory findings need replication in larger, multicenter cohorts before firm conclusions can be drawn.

Background: Preoperative differentiation between rectal mucinous adenocarcinoma (MAC) and non-mucinous adenocarcinoma (NMAC) remains a clinical challenge. This study aimed to develop and validate a nomogram incorporating baseline clinical characteristics and magnetic resonance imaging (MRI) features to distinguish MAC from NMAC.

Methods: This retrospective study included clinical baseline characteristics, laboratory parameters, and MRI features of patients with MAC and NMAC from two medical centers. Relevant variables were identified using univariate logistic regression analysis. Separate models based on clinical and imaging features were developed and subsequently integrated into a nomogram. Model performance was evaluated using receiver operating characteristic (ROC) curves and the area under the curve (AUC), and decision curve analysis (DCA) was conducted to assess clinical utility.

Results: Data from 221 patients (NMAC = 160, MAC = 61) from Center 1 were collected for this study. Data from 76 patients (NMAC = 54, MAC = 22) from Center 2 were used as an external validation cohort to verify the robustness of the models. We developed three models: a clinical model, an imaging feature model, and a nomogram. The nomogram integrating both clinical and imaging features demonstrated the best performance, with an AUC of 0.937 (95% CI, 0.894-0.979) in the training cohort and 0.882 (95% CI, 0.793-0.971) in the validation cohort. In the validation cohort, the nomogram achieved a sensitivity of 0.869, specificity of 0.925, and accuracy of 0.909. Furthermore, calibration curves confirmed good agreement between the predicted and observed outcomes.

Conclusions: The nomogram integrating clinical characteristics with MRI features enables efficient and practical differentiation between rectal MAC and NMAC, providing a valuable reference for individualized treatment decisions.