Balancing High-performance and Lightweight: HL-UNet for 3D Cardiac Medical Image Segmentation.

RATIONALE AND OBJECTIVES: Cardiac magnetic resonance imaging is a crucial tool for analyzing, diagnosing, and formulating treatment plans for cardiovascular diseases. Currently, there is very little research focused on balancing cardiac segmentation performance with lightweight methods. Despite the existence of numerous efficient image segmentation algorithms, they primarily rely on complex and computationally intensive network models, making it challenging to implement them on resource-constrained medical devices. Furthermore, simplified models designed to meet the requirements of device lightweighting may have limitations in comprehending and utilizing both global and local information for cardiac segmentation. MATERIALS AND METHODS: We propose a novel 3D high-performance lightweight medical image segmentation network, HL-UNet, for application in cardiac image segmentation. Specifically, in HL-UNet, we propose a novel residual-enhanced Adaptive attention (REAA) module that combines residual-enhanced connectivity with an adaptive attention mechanism to efficiently capture key features of input images and optimize their representation capabilities, and integrates the Visual Mamba (VSS) module to enhance the performance of HL-UNet. RESULTS: Compared to large-scale models such as TransUNet, HL-UNet increased the Dice of the right ventricular cavity (RV), left ventricular myocardia (MYO), and left ventricular cavity (LV), the key indicators of cardiac image segmentation, by 1.61%, 5.03% and 0.19%, respectively. At the same time, the Params and FLOPs of the model decreased by 41.3 M and 31.05 G, respectively. Furthermore, compared to lightweight models such as the MISSFormer, the HL-UNet improves the Dice of RV, MYO, and LV by 4.11%, 3.82%, and 4.33%, respectively, when the number of parameters and computational complexity are close to or even lower. CONCLUSION: The proposed HL-UNet model captures local details and edge information in images while being lightweight. Experimental results show that compared with large-scale models, HL-UNet significantly reduces the number of parameters and computational complexity while maintaining performance, thereby increasing frames per second (FPS). Compared to lightweight models, HL-UNet shows substantial improvements across various key metrics, with parameter count and computational complexity approaching or even lower.

Blocking Nonspecific Interactions Using Y-Shape Poly(ethylene glycol).

Nonspecific interactions play a significant role in physiological activities, surface chemical modification, and artificial adhesives. However, nonspecificity sometimes causes sticky problems, including surface fouling, decreased target specificity, and artifacts in single-molecule measurements. Adjusting the liquid pH, using protein-blocking additives, adding nonionic surfactants, or increasing the salt concentration are common methods to minimize nonspecific binding to achieve high-quality data. Here, we report that grafting heteromorphic polyethylene glycol (Y-shape PEG) with two inert terminates could noticeably decrease nonspecific binding. As a proof-of-concept, we performed single-molecule force spectroscopy and fluorescence staining imaging experiments to verify the feasibility of Y-shape PEG in blocking nonspecific interactions. Our results indicate that Y-shape PEG could serve as a prominent and efficient candidate to minimize nonspecificity for scientific and biomedical applications.

Sequence Tendency for the Interaction between Low-Complexity Intrinsically Disordered Proteins.

Reversible interaction between intrinsically disordered proteins (IDPs) is considered as the driving force for liquid-liquid phase separation (LLPS), while the detailed description of such a transient interaction process still remains a challenge. And the mechanisms underlying the behavior of IDP interaction, for example, the possible relationship with its inherent conformational fluctuations and sequence features, remain elusive. Here, we use atomistic molecular dynamics (MD) simulation to investigate the reversible association of the LAF-1 RGG domain, the IDP with ultra-low LLPS concentration (0.06 mM). We find that the duration of the association between two RGG domains is highly heterogeneous, and the sustained associations essentially dominate the IDP interaction. More interestingly, such sustained associations are mediated by a finite region, that is, the C-terminal region 138-168 (denoted as a contact-prone region). We noticed that such sequence tendency is attributed to the extended conformation of the RGG domain during its inherent conformational fluctuations. Hence, our results suggest that there is a certain region in this low-complexity IDP which can essentially dominate their interaction and should be also important to the LLPS. And the inherent conformational fluctuations are actually essential for the emergence of such a hot region of IDP interaction. The importance of this hot region to LLPS is verified by experiment.

Computed Tomography Image Analysis and Clinical Diagnosis Value of Nasal Olfactory Neuroblastoma.

BACKGROUND: This article retrospectively analyzed the data of all patients diagnosed with malignant tumors of the nasal cavity and sinuses by pathologic section in all patients (including outpatients and inpatients) in our hospital from January 2008 to December 2017 (10 years) to discuss the nasal cavity The clinical characteristics of sinus malignant tumors and the clinical characteristics (symptoms, signs, imaging examinations) of 35 cases of olfactory neuroblastoma with high incidence were analyzed, and their treatment methods and prognosis were summarized for future nasal tumors. CLINICAL BASIS FOR PREVENTION AND TREATMENT: In our department, 300 cases of malignant tumors of nasal cavity and sinuses were treated within 10 years. AGE: Eighty-six cases (28.6%) were in the age group 50-59 years old, occupying first place, and 55 cases (18.3%) were in the age group 60-69 years old, occupying second place. By comparing groups A and B with malignant tumors of the nasal cavity and sinuses, the difference in age distribution of olfactory neuroblastoma is statistically significant. Of the 35 cases of olfactory neuroblastoma, 23 cases were diagnosed with nasal tumors on admission, 12 cases were initially diagnosed with nasal polyps (including 7 cases with hemorrhagic and necrotic nasal polyps), and 20 cases underwent rapid pathologic definite diagnosis during the operation. DIAGNOSIS FOR POSTOPERATIVE PATHOLOGIC RETURNS: On the basis of our analysis, the following conclusions can be drawn: Of the 300 cases of malignant tumors of the nasal cavity and paranasal sinuses, age distribution tends to be younger, the overall incidence of males is higher than that of females, but the proportion of male and female olfactory neuroblastoma is equal, and the incidence of adenoid cystic carcinoma and sarcoma. The morphology of cancer can be divided as follows: the tissue source is epithelial tissue, the cancer is harder, paler, drier, and often forms cancer nests, the parenchymal and interstitial boundaries are clear, and the fibrous tissue is hyperplastic. The incidence rate of females is higher than that of males, with the primary origin in the sinuses. Among them, the maxillary sinus is more common. Olfactory neuroblastoma, lymphoma, and sarcoma are gradually increasing in malignant tumors of the nasal cavity and sinuses.