BMC Medical Imaging最新文献

Purpose: This study aimed to develop and evaluate an adaptive weighted ensemble learning model using multiple CNN feature extractors for multi-modal MRI classification of PI-RADS 3-5 prostate lesions. The primary goal was to reduce unnecessary invasive biopsies while maintaining high diagnostic accuracy for prostate cancer detection.

Methods: A retrospective diagnostic accuracy study analyzed 196 patients (mean age 64 ± 8.5 years) with PI-RADS 3-5 lesions who underwent multiparametric MRI and biopsy between January 2023-November 2024. Five CNN feature extractors (MobileNet_v2, VGG16, DenseNet121, EfficientNet_b0, ResNet50) were compared within an adaptive weighted ensemble model integrating DCE, DWI, and T2-weighted sequences. The model incorporated expert architectures (CNN, Transformer, Attention LSTM) for each modality with dynamic weighting mechanisms. Performance was evaluated using 5-fold cross-validation with data augmentation and ADASYN balancing, comparing against histopathological reference standards and radiologist interpretations.

Results: VGG16 achieved the highest diagnostic accuracy (99.0 ± 0.7%, AUC 99.9 ± 0.1%), followed by MobileNet_v2 (97.5 ± 0.7%, AUC 99.7 ± 0.2%). The ensemble model demonstrated superior specificity compared to radiologists' biopsy recommendations for PI-RADS 3-5 lesions (98.9% vs. 0.0%) while maintaining high sensitivity (99.1% vs. 100%). Learned modality weights showed DCE contributed most significantly (41.6 ± 2.0%), followed by T2-weighted (33.9 ± 2.1%) and DWI (24.6 ± 1.6%) sequences.

Conclusion: The proposed adaptive weighted ensemble model achieved superior diagnostic performance for prostate cancer classification compared to radiologist interpretation, demonstrating significant potential to reduce unnecessary biopsies while maintaining high sensitivity for cancer detection. These findings highlight the potential of the approach to improve the efficiency of prostate cancer diagnosis and support better clinical decision-making in prostate cancer management.

Objectives: Metabolic syndrome (MetS) presents significant health risks, but studies on individual component of MetS or its combined impact on bone mass have shown conflicting results. Therefore, this study aimed to analyze the relationship between abdominal fat and bone mineral density (BMD) in women with MetS using gemstone spectral imaging (GSI).

Methods: A retrospective study was performed on 76 women with MetS scheduled for sleeve gastrectomy between June and November 2021. Based on cluster analysis of BMD parameters, the patients were categorized into the high (54) and low (22) BMD groups. Clinical, BMD, and body composition metrics were analyzed separately. Univariate and multivariate logistic regression analyses were used to evaluate patients' clinical and body composition parameters. Receiver operating characteristic (ROC) curves were generated to determine the optimal diagnostic thresholds of various parameters for diagnosing the high and low BMD groups. Furthermore, taking lumbar vertebral BMD as the dependent variable, multiple linear regression analysis was performed.

Results: Significant differences in body composition were observed between the high and low BMD groups, with lower abdominal fat in patients in the high BMD group. The ROC curves showed a total abdominal fat volume threshold of 4733.2mL for predicting BMD (P = 0.008). Furthermore, using multiple linear regression adjusted for age, a statistically significant negative correlation was observed between the lumbar vertebral BMD and abdominal fat volume.

Conclusion: Abdominal fat volume influenced the GSI-BMD in women with MetS. As the abdominal fat increased, the patients' GSI-BMD in the lumbar spine also decreased.

Purpose: Cognitive impairment is a common but poorly understood comorbidity in chronic obstructive pulmonary disease (COPD). Although gray matter abnormalities have been observed in this population, the contribution of cortical gyrification-a structural feature linked to cognitive development and brain plasticity-remains unclear. This study aimed to characterize region-specific cortical gyrification alterations and examine their associations with domain-specific cognitive function and disease severity.

Methods: We enrolled 59 patients with stable COPD and 49 healthy controls who underwent pulmonary function testing, Montreal Cognitive Assessment, and high-resolution T1WI. The Toro's Gyrification Index quantified cortical gyrification. Group comparisons, partial correlations, and multiple linear regression analyses were conducted with adjustments for age, sex, and total intracranial volume.

Results: Compared with healthy controls, the patient group showed increased Toro's Gyrification Index in the bilateral superior temporal gyrus and left insula, and decreased values in the bilateral lingual gyri (P < .05). In the patient group, Toro's Gyrification Index in the left superior temporal gyrus was negatively correlated with abstract thinking (r = - .46, P = .003) and attention scores (r = - .39, P = .01). A regression model incorporating Toro's Gyrification Index in the left superior temporal and right lingual gyri explained 31% of the variance in abstract thinking score (F = 3.68, P = .004). The Global Initiative for Chronic Obstructive Lung Disease stage significantly predicted the right superior temporal gyrus Toro's Gyrification Index (F = 3.98, P = .002), with higher values observed in patients with disease stages 3 and 4 than stages 1 and 2 (F = 4.74, P = .005).

Conclusions: COPD is associated with region-specific, bidirectional cortical gyrification changes that are closely linked to cognitive impairment and disease severity. These findings suggest that gyrification-based metrics may offer a novel neuroimaging perspective for understanding brain reorganization in COPD.

Clinical trial number: Not applicable.