Multi-scale Bottleneck Residual Network for Retinal Vessel Segmentation.

Precise segmentation of retinal vessels is crucial for the prevention and diagnosis of ophthalmic diseases. In recent years, deep learning has shown outstanding performance in retinal vessel segmentation. Many scholars are dedicated to studying retinal vessel segmentation methods based on color fundus images, but the amount of research works on Scanning Laser Ophthalmoscopy (SLO) images is very scarce. In addition, existing SLO image segmentation methods still have difficulty in balancing accuracy and model parameters. This paper proposes a SLO image segmentation model based on lightweight U-Net architecture called MBRNet, which solves the problems in the current research through Multi-scale Bottleneck Residual (MBR) module and attention mechanism. Concretely speaking, the MBR module expands the receptive field of the model at a relatively low computational cost and retains more detailed information. Attention Gate (AG) module alleviates the disturbance of noise so that the network can concentrate on vascular characteristics. Experimental results on two public SLO datasets demonstrate that by comparison to existing methods, the MBRNet has better segmentation performance with relatively few parameters.

Anterior cervical interbody fusion with the Zero-P spacer: mid-term results of two-level fusion.

PURPOSE: Our aim was to compare the safety and efficacy of anterior cervical discectomy and fusion (ACDF) using the Zero-P spacer versus the plate method in patients with cervical spine spondylosis. METHODS: Clinical and radiologic data from 69 patients undergoing two-level ACDF from January 2009 to May 2011 were collected prospectively. The Zero-P spacer was implanted in 37 patients (group A) and the anterior cervical plate and interbody cage in 32 (group B). Patients were followed for at least 3 years after surgery. Clinical outcomes were analyzed using the Neck Disability Index and Japanese Orthopaedic Association (JOA) scoring. The thickness of the prevertebral soft tissue at the fused levels was measured on the lateral cervical spine radiographs and dysphagia was assessed using the Bazaz score. Fusion rate, change in cervical lordosis, and adjacent segment degeneration were analyzed. RESULTS: Neurologic outcomes were statistically equivalent between the two groups. The incidence of postoperative dysphagia was significantly lower in group A than in group B at 2 and 6 months (p < 0.05). At the final follow-up, there were no significant differences in the C2-C7 Cobb angles between the two groups (p > 0.05). Also, degenerative changes in adjacent segments occurred in five group A patients and seven group B patients (p = 0.361). There were no differences in fusion rate during the radiologic follow-up. CONCLUSIONS: Clinical results with the Zero-P spacer used for two-level ACDF were satisfactory. The device is superior to the traditional plate for preventing postoperative dysphagia and avoiding possible complications associated with a plate. Prospective trials with more patients and longer follow-ups are required to confirm these observations.

Ameliorating and refining islet organoids to illuminate treatment and pathogenesis of diabetes mellitus.

Diabetes mellitus, a significant global public health challenge, severely impacts human health worldwide. The organoid, an innovative in vitro three-dimensional (3D) culture model, closely mimics tissues or organs in vivo. Insulin-secreting islet organoid, derived from stem cells induced in vitro with 3D structures, has emerged as a potential alternative for islet transplantation and as a possible disease model that mirrors the human body's in vivo environment, eliminating species difference. This technology has gained considerable attention for its potential in diabetes treatment. Despite advances, the process of stem cell differentiation into islet organoid and its cultivation demonstrates deficiencies, prompting ongoing efforts to develop more efficient differentiation protocols and 3D biomimetic materials. At present, the constructed islet organoid exhibit limitations in their composition, structure, and functionality when compared to natural islets. Consequently, further research is imperative to achieve a multi-tissue system composition and improved insulin secretion functionality in islet organoid, while addressing transplantation-related safety concerns, such as tumorigenicity, immune rejection, infection, and thrombosis. This review delves into the methodologies and strategies for constructing the islet organoid, its application in diabetes treatment, and the pivotal scientific challenges within organoid research, offering fresh perspectives for a deeper understanding of diabetes pathogenesis and the development of therapeutic interventions.

MAS-UNet: a U-shaped network for prostate segmentation.

Prostate cancer is a common disease that seriously endangers the health of middle-aged and elderly men. MRI images are the gold standard for assessing the health status of the prostate region. Segmentation of the prostate region is of great significance for the diagnosis of prostate cancer. In the past, some methods have been used to segment the prostate region, but segmentation accuracy still has room for improvement. This study has proposed a new image segmentation model based on Attention UNet. The model improves Attention UNet by using GN instead of BN, adding dropout to prevent overfitting, introducing the ASPP module, adding channel attention to the attention gate module, and using different channels to output segmentation results of different prostate regions. Finally, we conducted comparative experiments using five existing UNet-based models, and used the dice coefficient as the metric to evaluate the segmentation result. The proposed model achieves dice scores of 0.807 and 0.907 in the transition region and the peripheral region, respectively. The experimental results show that the proposed model is better than other UNet-based models.