Lab-in-a-cartridge for real-time detection of tuberculosis via precise measurement of urinary lipoarabinomannan
本研究介绍了一种创新的便携式生物工具——实验室内的卡匣（LIC），用于实时检测尿液中的脂多糖抗原（LAM），以应对结核病（TB）诊断的挑战。LIC系统通过无泵液体处理和磁力富集技术，实现了对低浓度LAM的高灵敏度检测，检测阈值低至0.01 pg/mL，灵敏度达到92%，特异性为88%。该设备的设计旨在支持低资源环境中的分散式诊断，符合世界卫生组织的END TB战略，具有重要的公共卫生意义。
Active-matrix digital microfluidics for high-throughput, precise droplet handling
AM-DMF技术利用半导体电极阵列动态控制微米级液滴，提供高通量和精确的液体样本处理能力。该技术在基因组学、单细胞分析和药物发现等生物医学领域展现出巨大潜力。尽管面临生物污垢和试剂选择性等挑战，AM-DMF的创新性和商业价值使其成为未来生命科学的重要工具。
Thermally programmed one-pot CRISPR assay for on-site pandemic surveillance
TRACE is a novel CRISPR-based diagnostic assay designed for rapid and accurate detection of monkeypox virus (MPXV). It operates in a one-pot format, significantly reducing complexity and time to results. The assay demonstrates a limit of detection of 2.5 copies/test and achieves 99.5% accuracy across various clinical samples. Its design allows for simultaneous detection of pathogen and host genes, addressing critical limitations of existing CRISPR assays. TRACE meets ASSURED criteria, making it suitable for deployment in resource-limited settings, thus enhancing pandemic preparedness.
Comparison between SBMA and MPC coatings on PEEK surface: stability over time and anti-inflammatory effects in vitro
本研究比较了两种不同的涂层策略（SBMA和MPC）在PEEK表面的应用，重点评估了其稳定性和抗炎效果。结果表明，MPC涂层在调节炎症反应方面表现优异，能够显著降低多种炎症标志物的释放，且在生理条件下保持良好的稳定性。这些发现为PEEK植入物的安全性和整合性提供了新的思路，具有重要的临床应用潜力。
Identifying key clinical and biochemical predictors of treatment outcomes in inflammatory bowel disease: a real-world evidence study
该研究通过机器学习模型分析227名炎症性肠病（IBD）患者的临床和生化数据，识别出影响生物治疗反应的关键预测因子。研究结果显示，白细胞计数、C反应蛋白（CRP）和维生素B12水平等变量对治疗效果有显著影响。模型的F1分数分别为0.842和0.869，表明其在临床决策支持中的潜力。研究强调了真实世界数据在个性化治疗中的重要性，并呼吁进行多中心验证以增强模型的适用性。
Reach and implementation of human and AI-assisted diabetic retinopathy screening models in primary healthcare settings in India
该研究评估了在印度初级医疗环境中实施人类和AI辅助的糖尿病视网膜病变筛查模型的可行性，采用RE-AIM框架分析了不同筛查模型的覆盖率和实施效果。结果显示，社区基础筛查模型的参与率显著高于设施基础筛查，主要由于交通和医疗意识等障碍。研究强调了在资源有限的环境中实施有效筛查策略的必要性，为未来的公共卫生干预提供了重要参考。
Microglia across evolution: from conserved origins to functional divergence
该研究探讨了小胶质细胞的起源及其在不同物种中的功能差异，强调了这些细胞在维持中枢神经系统健康和应对病理状态中的重要性。研究表明，小胶质细胞在脊椎动物中具有保守的起源和功能，但在形态、基因表达及对刺激的反应上存在显著差异。这些发现为小胶质细胞靶向治疗神经疾病提供了新的视角。
The relationship between CTCs, TFD and postoperative prognosis of cervical cancer patients and the construction of prediction models
本研究开发了一种基于循环肿瘤细胞（CTCs）和肿瘤纤维化距离（TFD）的术后复发/转移风险预测模型，旨在为宫颈癌患者提供个性化治疗策略。研究结果显示，CTCs和TFD的结合能够有效预测术后复发风险，模型在148名患者中表现出良好的预测能力（AUC为0.91）。此外，动物实验验证了模型的生物学机制，支持其在临床中的应用潜力。这一创新模型为宫颈癌患者的术后监测和管理提供了新的思路。
Development and validation of the risk stratification based on deep learning and radiomics to predict survival of advanced cervical cancer
本研究开发了一种基于深度学习的风险分层模型，旨在预测晚期宫颈癌患者的生存率。通过对396名患者的CT图像进行分析，构建了一个整合临床变量的预后模型。该模型在训练和验证队列中表现出良好的预测能力，C-index分别为0.784和0.726，显示出其在临床决策中的潜在应用价值。研究结果表明，深度学习和放射组学的结合能够有效提高生存预测的准确性，为个性化治疗提供支持。
Sequential lymphotoxin-β receptor and retinoic acid receptor signals regulate cDC2 fate
该研究揭示了小鼠类型2常规树突细胞（cDC2）的发育机制，特别是其亚群的分化路径和微环境信号的作用。研究表明，cDC2亚群在脾脏中占主导地位，并依赖于细胞内的视黄酸信号和Notch信号维持其分化。这些发现为理解免疫细胞的发育提供了新的视角，并可能对未来的免疫治疗策略产生影响。
Impact of statins on short and long term mortality in severe community acquired pneumonia in the intensive care unit
This study investigates the impact of statins on mortality in patients with severe community-acquired pneumonia (sCAP) in the ICU. Using real-world data from the MIMIC-IV database, it simulated a randomized controlled trial and found that statin therapy significantly reduced both acute and long-term mortality rates. The average treatment effect was −4.03%, with Kaplan-Meier analysis confirming improved survival for statin users. These findings suggest that statins could serve as a valuable co-adjuvant therapy in treating sCAP, warranting further research to optimize treatment protocols.
A multi-head YOLOv12 with self-supervised pretraining for urinary sediment particle detection
本研究提出了一种基于多头YOLOv12架构的深度学习方法，用于自动化尿沉渣分析，结合自监督预训练和Slicing Aided Hyper Inference (SAHI)技术。通过创建OpenUrine数据集，包含790个标记和5640个未标记图像，研究显示该模型在39类尿沉渣颗粒的检测中达到了76.59%的精度，显著提高了尿沉渣分析的准确性和效率，具有重要的临床应用潜力。
Analysis of the clinical features of neurocristopathy-related hearing loss and how these relate to outcomes after cochlear implantation
该研究分析了神经嵴病相关的听力损失及其与人工耳蜗植入后结果的关系，涉及16名儿童的临床特征和基因评估。研究发现，Waardenburg综合征和Noonan综合征患者在植入后取得了良好的听力恢复，而CHARGE综合征患者的恢复效果较差。通过整合基因、耳部结构和神经发育评估，研究为临床决策提供了依据，强调了个体化治疗的重要性。
Multimodal deep learning framework integrating multiphase CT and histopathological whole slide imaging for predicting recurrence in ccRCC
本研究提出了一种多模态深度学习框架CPNet，结合多相CT和病理全切片成像，旨在预测清细胞肾细胞癌（ccRCC）患者的术后复发风险。通过对274名患者的回顾性分析，CPNet模型在术后无病生存率的预测中表现出色，AUC达到0.8363，显示出其在临床应用中的潜力。该模型的创新之处在于其双向注意机制，有效整合了CT和病理图像的信息，为ccRCC患者的个性化管理提供了新的思路。
Synergistic anticancer effects of Chrysin and trastuzumab in HER2-Positive breast cancer cells
Chrysin, a natural flavonoid, significantly enhances the efficacy of trastuzumab in HER2-positive breast cancer cells, as demonstrated by improved apoptosis rates and reduced HER2 expression. This study highlights the potential of combining natural compounds with targeted therapies to overcome resistance and improve treatment outcomes in oncology. The findings suggest that further investigation into the mechanisms and clinical applicability of this combination therapy is warranted.