Evaluating mammographic density′s contribution to improve a breast cancer risk model with questionnaire-based and polygenic factors
本研究评估了乳腺密度对乳腺癌风险模型的贡献，结合问卷基础风险因素和多基因风险评分，验证了其在风险分层中的有效性。研究结果表明，加入乳腺密度后，模型的风险预测能力有所提高，尤其是在50岁以上的女性中，能够更好地识别高风险患者。这一发现为个性化乳腺癌筛查和预防策略提供了重要的临床依据，具有广泛的应用潜力。
Artificial Intelligence (AI)-assisted readout method for the evaluation of skin prick automated test results
AI-assisted readout method for skin prick tests demonstrates high accuracy and significantly reduces variability in results. The study involved 897 patients across multiple hospitals, validating the AI algorithm against physician measurements. Results show a specificity of 98.4% and sensitivity of 85.0%, with a notable reduction in readout time by 3.7 times. This innovation enhances standardization in allergy diagnostics, improving both efficiency and patient outcomes.
Graph neural network model using radiomics for lung CT image segmentation
GEANet is a novel framework for lung CT image segmentation that integrates radiomics and Graph Neural Networks (GNN). It addresses challenges in segmentation accuracy and model interpretability, achieving superior performance compared to existing methods. The framework employs a hybrid loss function to enhance robustness against class imbalance and improve boundary delineation. Experimental results indicate that GEANet significantly outperforms traditional models, making it a promising tool for lung cancer diagnosis and management.
Bilateral collaborative streams with multi-modal attention network for accurate polyp segmentation
BiCoMA网络通过结合卷积神经网络和变换器的优势，显著提高了结肠镜图像中息肉的分割精度。该研究展示了BiCoMA在多个基准数据集上的优越性能，尤其是在复杂的临床场景中，强调了其在早期癌症检测和预防中的重要性。
Long non-coding RNAs as a biomarker for homologous recombination deficiency and parp inhibitor sensitivity in high-grade serous ovarian cancers
本研究探讨了长非编码RNA（lncRNA）ENSG00000272172.1在高等级浆液性卵巢癌（HGSC）中的潜在生物标志物价值，特别是在同源重组缺陷（HRD）和PARP抑制剂敏感性方面。研究表明，ENSG00000272172.1在HRD阳性肿瘤中显著上调，并且可通过血浆检测，显示出其作为非侵入性生物标志物的潜力。该lncRNA的高表达与患者的整体生存率显著相关，提示其在临床决策中的应用前景。
Decoding the molecular and genomic landscape of hepatocellular carcinoma: biomarker discovery, classification frameworks, and therapeutic targeting
该文综述了肝细胞癌（HCC）的分子和基因组特征，强调了新兴生物标志物如DCLK1、GPC3等在早期诊断中的潜力。传统的生物标志物如AFP的敏感性不足，导致HCC的早期诊断困难。通过整合新技术如液体活检和单细胞测序，研究者们希望能够改善HCC的早期检测和个性化治疗策略。
Novel peptides from the edible bivalve Ruditapes decussatus target apoptosis, autophagy, and FGF19-FGFR4 signaling in human cancer cell lines
本研究从可食用双壳类Ruditapes decussatus中提取了新型生物活性肽，显示出显著的抗癌潜力，尤其针对肝癌和结肠癌细胞。通过诱导细胞凋亡和自噬，这些肽表现出选择性细胞毒性，且对正常细胞影响较小。研究结果为开发新一代抗癌疗法奠定了基础，具有良好的商业投资前景。
Improved vascular response to a novel polymeric blend PLLA/PLGA coronary scaffold material: preclinical study
研究评估了一种新型聚合物混合物PLLA/PLGA生物可吸收支架在猪冠状动脉再狭窄模型中的表现。结果显示，该支架在减少炎症和促进血管愈合方面优于传统的PLA支架，具有重要的临床应用潜力。研究结果表明，PLLA/PLGA支架在90天后保持通畅且无再狭窄现象，支持其作为下一代生物可吸收支架的潜力。
Rapid thrombogenesis prediction in COVID-19 patients using DNN with data labeling
研究提出了一种基于深度神经网络(DNN)的框架，用于快速预测COVID-19患者的血栓形成风险，显著缩短了传统CFD模拟所需的时间。通过对多种机器学习算法的比较，发现XGBoost模型表现最佳，准确率超过95%。该框架的创新性和临床应用潜力为医疗决策提供了新的工具，具有较高的商业价值。
Intermittent fasting attenuates glial hyperactivation and photoreceptor degeneration in a NaIO3-induced mouse model of age-related macular degeneration
Intermittent fasting (IF) has been shown to attenuate glial hyperactivation and photoreceptor degeneration in a sodium iodate (NaIO3)-induced mouse model of age-related macular degeneration (AMD). This study highlights IF's potential as a therapeutic strategy by demonstrating its ability to preserve visual function, reduce oxidative stress, and inhibit inflammatory pathways. The findings suggest that dietary interventions like IF could play a significant role in managing AMD and other retinal degenerative diseases.
Design of optimized epigenetic regulators for durable gene silencing with application to PCSK9 in nonhuman primates
研究开发了一种优化的表观遗传调控因子EpiReg-T，能够在非人灵长类动物中有效抑制PCSK9基因表达，从而降低LDL胆固醇水平。该技术展示了高效的基因沉默能力和较低的脱靶效应，具有重要的临床应用潜力，适合早期投资。
An adaptive autoregressive diffusion approach to design active humanized antibodies and nanobodies
HuDiff is an innovative adaptive diffusion approach designed for humanizing antibodies and nanobodies, showing significant advancements in humanness and binding affinity. This method eliminates the need for humanized templates and has been validated through experiments targeting SARS-CoV-2, demonstrating its potential for clinical applications. The research is backed by teams from Lanzhou University and Tencent AI Lab, highlighting its relevance in the biotechnology sector.
Cassini: streamlined and scalable method for in situ profiling of RNA and protein
Cassini是一种新颖的空间基因组学方法，能够在短时间内高效、低成本地同时分析RNA和蛋白质特征。该方法结合了强大的信号放大技术和优化的免疫染色缓冲液，能够在同一实验中实现多重检测。研究表明，Cassini在灵敏度和准确性上与现有技术相当，且其开放源代码和在线探针设计平台使其更易于使用，具有广泛的应用潜力，尤其在脑科学和生物技术领域。
Targeted clearance of extracellular Tau using aptamer-armed monocytes alleviates neuroinflammation in mice with Alzheimer’s disease
研究开发了一种基于aptamer的单核细胞治疗方法，能够有效清除阿尔茨海默病小鼠模型中的细胞外Tau，减轻神经炎症并改善认知功能。该方法通过生物正交化学将高亲和力的Tau特异性aptamer共价结合到单核细胞表面，确保其有效穿越血脑屏障并在Tau富集的脑区积累。研究结果显示，该治疗方法具有良好的安全性和长期疗效，提供了阿尔茨海默病治疗的新思路。
Elevated maternal testosterone induces sex-specific neurodevelopmental changes and ASD-related behavioral phenotypes in rat offspring
研究发现，孕期母体睾酮水平升高会导致后代出现性别特异性的神经发育变化和自闭症相关行为特征。实验结果表明，母体睾酮影响后代的神经结构，尤其是大脑DHA水平的降低，可能是导致认知和社交功能缺陷的机制。该研究为孕期高睾酮水平的潜在风险提供了新的视角，并强调了DHA补充作为治疗策略的可能性，具有重要的临床和商业价值。