[Research progress on high performance liquid chromatography stationary phases based on mixed mode]. [PDF]
Se PuLi SJ, Li Z, Tang ZX, Lyu M, Wang LT.
europepmc +1 more source
工程科学与技术在日常运维管理中，供水管网（WDNs）有效的长期维护决策对提升城市供水韧性和保障系统的可持续运行至关重要，但现有长期维护方法普遍存在评估指标维度单一、维护决策的全局优化能力不足、模型计算效率低及泛化性差的问题。因此，本研究提出了一种融合图卷积网络（GCN）与深度强化学习（DRL）的WDN多目标韧性维护决策模型（M2DRL），旨在以韧性指数最大化为目标生成最优恢复序列。首先，构建了一种融合水力性能、水质性能与运输时间的多目标韧性评估框架，首次将三类关键指标同时纳入优化目标，实现韧性维护决策的多目标协同优化；
王思琦 +3 more
doaj
[Research progress in artificial intelligence for the diagnosis and management of diseases in preterm infants]. [PDF]
Zhongguo Dang Dai Er Ke Za ZhiYuan Y, Tang LH, Guan LR.
europepmc +1 more source
[Musculoskeletal biomechanical analysis of the three-dimensional gastrocnemius and soleus muscles]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za ZhiTang Z, Xu J, Jin Z, Li J.
europepmc +1 more source
[Hierarchical attention-driven multiple instance learning for clear cell renal cell carcinoma grading and staging in digital pathology]. [PDF]
Nan Fang Yi Ke Da Xue Xue BaoXu J, Mao Y, Zhang Y.
europepmc +1 more source
[A protein-ligand binding affinity prediction model integrating multi-scale interaction features]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za ZhiLiu H, Yan L, Wang S, Peng J.
europepmc +1 more source
Gaoya dianqi油中溶解乙炔作为电力变压器中重要的放电程度表征参量之一，对其进行多步预测可以为变压器故障诊断及预警提供重要依据。现有的状态预测模型主要集中于单步预测，对于未来更长时期变化趋势的预测手段不足。此外，基于深度学习的多步预测模型的超参数选择大多基于经验和朴素的单一控制变量法，超参数之间的耦合关系没有得到充分的研究。文中提出基于多输出策略的门控循环单元（gated recurrent unit，GRU）神经网络多步预测模型，通过改变模型结构超参数和训练超参数研究超参数之间的耦合关系 ...
赵军 +6 more
doaj
[Research on weighted hypergraph attention neural network for the diagnosis of psychiatric disorders using brain functional connectivity networks]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za ZhiLin X +7 more
europepmc +1 more source
[Research progress on intelligent brain age prediction methods in diagnosis of Parkinson's disease]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za ZhiZheng S +6 more
europepmc +1 more source
Jianzhu Gangjiegou Jinzhan模块化钢结构建筑凭借工业化程度高、施工效率快等优势，已成为推动绿色低碳建筑发展的重要方向。模块间连接对模块化建筑的力学性能具有关键影响，然而当前针对模块间连接受剪性能及优化设计方法的研究仍较为缺陷。文中基于课题组前期完成的全装配可吊装节点（fully prefabricated liftable connection， FPLC）受剪性能试验研究，建立了精细化有限元模型并开展了参数分析，从而得到了包含1 000组不同参数的FPLC受剪性能数据库。运用了六种常用机器学习算法对FPLC受剪性能进行评估 ...
邓恩峰 +5 more
doaj