A mixed Mamba U-net for prostate segmentation in MR images.

The diagnosis of early prostate cancer depends on the accurate segmentation of prostate regions in magnetic resonance imaging (MRI). However, this segmentation task is challenging due to the particularities of prostate MR images themselves and the limitations of existing methods. To address these issues, we propose a U-shaped encoder-decoder network MM-UNet based on Mamba and CNN for prostate segmentation in MR images. Specifically, we first proposed an adaptive feature fusion module based on channel attention guidance to achieve effective fusion between adjacent hierarchical features and suppress the interference of background noise. Secondly, we propose a global context-aware module based on Mamba, which has strong long-range modeling capabilities and linear complexity, to capture global context information in images. Finally, we propose a multi-scale anisotropic convolution module based on the principle of parallel multi-scale anisotropic convolution blocks and 3D convolution decomposition. Experimental results on two public prostate MR image segmentation datasets demonstrate that the proposed method outperforms competing models in terms of prostate segmentation performance and achieves state-of-the-art performance. In future work, we intend to enhance the model's robustness and extend its applicability to additional medical image segmentation tasks.

Screening of Associated Factors for Erectile Dysfunction after Radical Prostatectomy and Construction of a Clinical Risk Assessment Model: A Retrospective Study.

OBJECTIVE: In this article, the associated factors for erectile dysfunction (ED) after radical prostatectomy (RP) were explored, and a clinical risk assessment model was constructed. METHODS: A total of 155 patients who underwent RP in People's Hospital of Hunan Province from November 2020, to November 2021, were selected as the study group. In accordance with the results of International Index of Erectile Function (IIEF-5) at 6 months after surgery, 88 patients were included in the ED group (IIEF-5 <22), and 67 patients were included in the non-ED group (IIEF-5 ≥22). Univariate and multivariate logistic regression analyses were conducted to screen the risk factors for ED after RP, and a risk model was constructed on this basis. In addition, 43 patients with ED after RP and 41 patients with non-ED after RP from January 2022, to January 2023, were included in the test group to evaluate the predictive efficacy of the clinical risk assessment model on the basis of the receiver operating characteristic curve. RESULTS: The study group had a lower postoperative IIEF-5 score than before surgery (p < 0.001). The incidence of ED after RP in the study group was 56.77% (88/155). Multivariate analysis showed that advanced age (odds ratio (OR) = 1.155), large prostate volume (OR = 1.077), smoking (OR = 5.676), drinking (OR = 3.495), hypertension (OR = 8.079), diabetes (OR = 6.082), low preoperative serum testosterone (T) level (OR = 0.684) and high preoperative serum endothelin-1 (ET-1) level (OR = 1.192) were risk factors for ED after RP (p < 0.05). A risk model was constructed as follows: Z = 0.144 × (age) + 0.074 × (prostate volume) + 1.736 × (smoking) + 1.251 × (drinking) + 2.089 × (hypertension) + 1.805 × (diabetes) - 0.380 × (preoperative serum T) + 0.175 × (preoperative serum ET-1). The area under curve (AUC), sensitivity, specificity and 95% CI of this model were 0.906, 97.70%, 73.20%, and 0.848-0.964, respectively (p < 0.001). CONCLUSIONS: The clinical risk assessment model constructed on the basis of the above factors provides some references for the scientific prevention and treatment of ED after RP.