Alzheimer's disease,the primary cause of dementia,is characterized by neuropathologies,such as amyloid plaques,synaptic and neuronal degeneration,and neurofibrillary tangles.Although amyloid plaques are the primar...Alzheimer's disease,the primary cause of dementia,is characterized by neuropathologies,such as amyloid plaques,synaptic and neuronal degeneration,and neurofibrillary tangles.Although amyloid plaques are the primary characteristic of Alzheimer's disease in the central nervous system and peripheral organs,targeting amyloid-beta clearance in the central nervous system has shown limited clinical efficacy in Alzheimer's disease treatment.Metabolic abnormalities are commonly observed in patients with Alzheimer's disease.The liver is the primary peripheral organ involved in amyloid-beta metabolism,playing a crucial role in the pathophysiology of Alzheimer's disease.Notably,impaired cholesterol metabolism in the liver may exacerbate the development of Alzheimer's disease.In this review,we explore the underlying causes of Alzheimer's disease and elucidate the role of the liver in amyloid-beta clearance and cholesterol metabolism.Furthermore,we propose that restoring normal cholesterol metabolism in the liver could represent a promising therapeutic strategy for addressing Alzheimer's disease.展开更多
目的:评价基于深度学习的继发性肺结核CT辅助诊断模型在临床应用中的价值。方法:回顾性收集2018年12月至2023年4月在重庆市公共卫生医疗救治中心接受胸部CT平扫的2004例患者的病例资料,分为肺部正常组(544例)、普通肺部感染组(526组)和...目的:评价基于深度学习的继发性肺结核CT辅助诊断模型在临床应用中的价值。方法:回顾性收集2018年12月至2023年4月在重庆市公共卫生医疗救治中心接受胸部CT平扫的2004例患者的病例资料,分为肺部正常组(544例)、普通肺部感染组(526组)和继发性肺结核组(934例)。按照随机分组(通过R语言的sample函数实现训练集和测试集的完全随机分组)的方式,将数据集划分为训练集(1402例,70.0%)和测试集(602例,30.0%)。所有图像采用肺野自动分割算法,获得肺野区域。进一步采用BasicNet和DenseNet算法进行三组间的分类研究。采用曲线下面积(area under curve,AUC)、敏感度、特异度和准确率评价模型的分类性能。最后,在测试数据中,将最优模型与3位不同年资的放射科医生的诊断结果进行比较。结果:602例独立测试集中,DenseNet模型的性能优于BasicNet模型,两种模型的平均AUC、敏感度、特异度和准确率分别为92.1%和89.4%、79.7%和74.0%、89.4%和86.6%、86.2%和83.3%。其中,DenseNet模型的诊断性能优于低年资医生(准确率分别为90.7%和89.1%,Kappa=0.677),与中年资和高年资医生的诊断水平(准确率分别为90.7%、92.2%和95.3%,Kappa值分别为0.746和0.819)保持高度一致性。结论:DenseNet模型能较准确地识别继发性肺结核,与放射科中年资医师的诊断水准相当,可以作为继发性肺结核的辅助诊断工具。展开更多
基金financially supported by the Science and Technology Innovation Program of Hunan Province,No.2022RC1220(to WP)China Postdoctoral Science Foundation,No.2022M711733(to ZZ)+2 种基金the National Natural Science Foundation of China,No.82160920(to ZZ)Hebei Postdoctoral Scientific Research Project,No.B2022003040(to ZZ)Hunan Flagship Department of Integrated Traditional Chinese and Western Medicine(to WP)。
文摘Alzheimer's disease,the primary cause of dementia,is characterized by neuropathologies,such as amyloid plaques,synaptic and neuronal degeneration,and neurofibrillary tangles.Although amyloid plaques are the primary characteristic of Alzheimer's disease in the central nervous system and peripheral organs,targeting amyloid-beta clearance in the central nervous system has shown limited clinical efficacy in Alzheimer's disease treatment.Metabolic abnormalities are commonly observed in patients with Alzheimer's disease.The liver is the primary peripheral organ involved in amyloid-beta metabolism,playing a crucial role in the pathophysiology of Alzheimer's disease.Notably,impaired cholesterol metabolism in the liver may exacerbate the development of Alzheimer's disease.In this review,we explore the underlying causes of Alzheimer's disease and elucidate the role of the liver in amyloid-beta clearance and cholesterol metabolism.Furthermore,we propose that restoring normal cholesterol metabolism in the liver could represent a promising therapeutic strategy for addressing Alzheimer's disease.
文摘目的:评价基于深度学习的继发性肺结核CT辅助诊断模型在临床应用中的价值。方法:回顾性收集2018年12月至2023年4月在重庆市公共卫生医疗救治中心接受胸部CT平扫的2004例患者的病例资料,分为肺部正常组(544例)、普通肺部感染组(526组)和继发性肺结核组(934例)。按照随机分组(通过R语言的sample函数实现训练集和测试集的完全随机分组)的方式,将数据集划分为训练集(1402例,70.0%)和测试集(602例,30.0%)。所有图像采用肺野自动分割算法,获得肺野区域。进一步采用BasicNet和DenseNet算法进行三组间的分类研究。采用曲线下面积(area under curve,AUC)、敏感度、特异度和准确率评价模型的分类性能。最后,在测试数据中,将最优模型与3位不同年资的放射科医生的诊断结果进行比较。结果:602例独立测试集中,DenseNet模型的性能优于BasicNet模型,两种模型的平均AUC、敏感度、特异度和准确率分别为92.1%和89.4%、79.7%和74.0%、89.4%和86.6%、86.2%和83.3%。其中,DenseNet模型的诊断性能优于低年资医生(准确率分别为90.7%和89.1%,Kappa=0.677),与中年资和高年资医生的诊断水平(准确率分别为90.7%、92.2%和95.3%,Kappa值分别为0.746和0.819)保持高度一致性。结论:DenseNet模型能较准确地识别继发性肺结核,与放射科中年资医师的诊断水准相当,可以作为继发性肺结核的辅助诊断工具。
