人工智能影像诊断技术在基层医院中的应用与思考

人工智能（artificial intelligence,AI）是研究利用计算机来模拟人的某些思维过程和智能行为的学科。近年来,随着机器学习和深度学习的应用,AI技术在医学影像诊断中得到了飞速发展,且取得了卓越的效果。基层医院医生由于临床诊断能力较薄弱和经验累积较慢等原因,使得基层医院的诊疗受到一定的影响,将AI影像诊断技术应用于基层医院能提高医院的诊断效率及临床诊断准确性。但是,AI与医学影像诊断技术的结合尚处于初步阶段,有一定的不足,需制定相应的解决方案,以推动AI在医学影像诊断领域的不断发展与成熟。     

磁共振成像的安全性与心脏金属移植物

磁共振成像（MRI）是临床上一种重要的影像学诊断方法，由于检查中外加磁场的作用，对体内有金属移植物的待检者（如心脏起搏器和除颤器、冠脉支架、人工心脏瓣膜移植物等）存在一定的风险。MR安全性是指一种装置进行MR检查时，对患者或其他个体不存在危险，但是有可能会影响诊断信息。传统认为体内金属移植物是行MR检查的禁忌证，近年来，有关金属物质MR成像的安全性研究结果表明携带金属移植物不是MR检查的绝对禁忌证。本文就MR与金属移植物的相互效应以及心脏金属移植物（matallicimplants）行MR检查的安全性进行综述，旨在为临床确需行MRI检查的此类患者提供些许参考。     

免疫检查点抑制剂相关心肌炎临床诊疗实施建议

免疫检查点抑制剂（immune-checkpoint inhibitors，ICIs）现已应用于多种恶性肿瘤的治疗，为肿瘤患者带来获益的同时，其严重ICIs相关心肌炎日益为临床带来新挑战。本临床诊疗实施建议聚焦ICIs相关心肌炎的危险因素、诊断与鉴别诊断、临床分型及治疗、监测转归和治疗重启等关键临床问题，参考国内外相关共识或指南和新近发表的循证证据，结合实际临床经验，为ICIs相关心肌炎的诊疗提供具有实践性的指导意见和建议。     

人工智能在神经介入诊疗中的应用前景

人工智能（artificial intelligence，AI）机器人能否逐渐替代人类来诊治疾病是当下全球研究的热点。计算机的海量存储、快速计算、不知疲倦学习、永无止境升级使得AI可能首先在诊断上替代并超越人类。计算机图像分析、工业自动化、智能控制的应用推动了AI机器人治疗系统的研发，并促使其逐步被应用到精准度和稳定性要求极高的显微手术和具有放射损害的介入治疗领域。2019年11月，首例AI机器人实施的颅内动脉瘤介入手术在加拿大多伦多西区医院完成，具有划时代意义。本文介绍了AI在脑血管疾病诊断，尤其介入手术机器人在神经介入诊疗方面的应用现状，并就AI应用前景进行展望。     

腔隙性脑梗塞52例临床分析

随着颅脑计算机断层扫苗（CT）和核磁共振（MR／）的应用，腔隙性脑梗塞病例进行临床分析，从中探讨腔隙性脑梗塞的病因，临床特点、诊断、预后及康复。     

肺结节人工智能评估危险度辅助人工阅片的应用价值

目的 探讨基于CT图像的人工智能（AI）在肺结节筛查和评估危险度的临床应用价值。方法 选取2022年1月～2022年12月在国药葛洲坝中心医院接受手术或穿刺并进行病理检查的肺结节患者100例。AI和医师分别进行结节诊断,比较A组（人工诊断,n=98）、B组（AI诊断,n=99）和C组（人工+AI联合诊断,n=99）的敏感度、特异性、阳性预测值和阴性预测值与准确率。以病理结果作为“金标准”,分析AI辅助医师预测良恶性的价值。结果 A与B组评估肺结节危险度差异无统计学意义（P>0.05）。C组的敏感度（98.75%）、阴性预测值（91.67%）和准确性（90.00%）优于A、B组。C组可明显提高诊断效能,优于A和B组单独阅片,且诊断与病理结果有较好一致性（Kappa值=0.632,P<0.01）。结论 基于AI的肺结节筛查和评级能够明显提高敏感度及准确率,并降低漏诊率,对于辅助影像诊断医生的阅片具有显著的临床价值。AI评估危险度辅助人工阅片肺结节的良恶性预判中同样具备较好的参考及临床应用价值。     

2007年慢性稳定性心绞痛防治指南解读

我国第一部慢性稳定性心绞痛诊断治疗指南（以下简称指南）已于2007年发表，为广大临床医师提供了一个适用于大多数患者的诊疗策略。指南充分收集循证医学证据，结合了我国实际情况，能反映慢性稳定性心绞痛防治领域中各个方面的最新进展，能指导临床实践。本文按照指南的写作顺序，就慢性稳定性心绞痛的诊断、治疗（包括药物和血管重建治疗）和特殊诊疗考虑等内容作深入浅出的解读。     

高压注射器在MR增强扫描中的注意事项

随着磁共振（magnetic resonance，MR）在临床的广泛应用，高压注射器在MR增强中的使用也越来越普及，从而使MR增强图像质量发生了明显的变化，加上高压注射器可以在短时间内行快速注射，实施快速扫描，能任意控制在某一造影期完成扫描，同时又进行双期或多期扫描，故而提高了多血供和少血供病灶的检出率，尤其是对难以定性诊断的小病灶，提高了诊断准确率，为患者的治疗提供了准确的依据，对平扫不能显示的病变及鉴别诊断却具有十分重要的应用价值，为临床诊断提供了很大帮助。     

颅脑急症CT、MR检查手段选择

评价CT、MR（核磁共振）在血管性颅脑急症诊断中的应用价值，如何利用CT、MR成像显像特点及优势，选择最佳检查手段，为这类疾病的早期治疗，减少并发症、延长生命创造有利条件。     

基于EasyDL开发糖尿病眼底病变人工智能分级诊断模型及其验证评价

目的 创新性利用人工智能（AI）开放平台EasyDL独立开发糖尿病视网膜病变（DR）的AI辅助诊断模型,并对其诊断准确指标进行评价。方法 采用Kaggle公开的糖尿病眼底疾病数据集的35 126张眼底照片作为训练集,上传至EasyDL开放平台建立AI辅助诊断模型。收集在眼科进行临床DR筛查的150例糖尿病患者共300张双眼的彩色眼底照片作为测试集,以3位副高及以上职称眼科医师的诊断为金标准,分别评价AI诊断模型、初级医师、中级医师及联合诊断对DR分级的诊断准确性。结果 非DR和轻度非增生型DR（NPDR）患者共170例,中度、重度NPDR和增生型DR（PDR）患者共130例。AI诊断模型灵敏度高但特异度低,各项诊断指标和中级医师诊断接近,比初级医师诊断优秀。当AI诊断模型和临床医师诊断相结合时,诊断的准确率和灵敏度均有所提高。在与金标准的一致性评价中,AI诊断模型的Kappa系数为1.00,而中级医师诊断的Kappa系数为0.88（P均< 0.01）。结论 基于开放平台EasyDL建立的AI诊断模型操作简单,能为DR的初筛提供帮助,同时也为不具备深度学习算法知识的临床医师提供有效的科研工具。     

A critical examination of the role of appreciative inquiry within an interprofessional education initiative

Appreciative inquiry (AI) is a relatively new approach to initiating or managing organizational change that is associated with the ‘positiveness’ movement in psychology and its offshoot positive organizational scholarship. Rather than dwelling upon problems related to change, AI encourages individuals to adopt a positive, constructive approach to managing change. In recent years, AI has been used to initiate change across a broad range of public and private sector organizations. In this article, we report findings from a subset of 50 interviews gathered in a wider study of interprofessional education (IPE) in which AI was employed as a change agent for implementing IPE in a number of health care institutions in a North American setting. A multiple case study approach. (Yin, 2002) was employed in the wider study and semi-structured interviews were undertaken with participants both before their IPE programs and directly afterwards to obtain a detailed understanding of their expectations and experiences of IPE. Interviews were analyzed in an inductive thematic manner in order to produce key emergent themes from each of the IPE programs. A process of re-analysis provided a set of themes which offered an understanding of the role of AI within this IPE initiative. Our findings identify a strong resonance and fit for AI both among the health and social care professionals who participated in this initiative. Numerous individuals commented on the enthusiasm and energy AI engendered, while praising its ability to enhance their working lives and interprofessional relationships. Yet a number of difficulties were also reported. These focused on problems with the translation of the AI process into achievable structural level (e.g. professional, cultural) changes. Based on these findings, the article goes on to argue that the use of AI can overlook a number of structural factors, which will ultimately limit its ability to actually secure meaningful and lasting change within health care.     

弥散张量成像在脑白质病变中的应用价值初探

ThisstudyperformedacombinedconventionalanddiffusiontensorMRimagingfrom10multiplesclerosis,10multiplelacunarinfarction,3cysticercosis,1angiitis,1morphinistand10healthycontrolvolunteerstoinvestigatedmorphologicandquantitativeindex.1Subjectsandmethods1.1SubjectsTwenty-fivepatientsandtenhealthycontrolvol-unteersunderwentacombinedconventionalanddiffusiontensorMRimagingduringoneyear.Thepatientsincludetenmultiplesclerosis(7womenand3men)whoseagewasfrom17to48years,themediandurationofthediseasewas1yea…     

应用FAIR技术评价难治性和非难治性颞叶癫痫灌注异常

目的用FAIR技术评价难治性及非难治性颞叶癫痫患者发作间期rCBF异常的模式是否不同.方法 9例难治性颞叶癫痫、21例非难治性颞叶癫痫患者及13名正常志愿者,测量其双侧大脑半球及双侧内侧颞叶的rCBF,并计算不对称指数(AI).结果难治性及非难治性癫痫患者大脑半球AI值与正常对照组之间有统计学差异(P值分别为0.012和0.029);难治性颞叶癫痫患者内侧颞叶AI值与正常对照组之间无统计学差异(P＝0.102),而非难治性癫痫患者与正常对照组之间有统计学差异(P＝0.049).结论难治性和非难治性颞叶癫痫患者发作间期rCBF异常的模式有所不同,虽然均可出现较广泛的rCBF异常,但难治性癫痫更倾向于出现双侧内侧颞叶低灌注,这可为判断预后提供帮助.     

A critical examination of the role of appreciative inquiry within an interprofessional education initiative

Appreciative inquiry (AI) is a relatively new approach to initiating or managing organizational change that is associated with the 'positiveness' movement in psychology and its offshoot positive organizational scholarship. Rather than dwelling upon problems related to change, AI encourages individuals to adopt a positive, constructive approach to managing change. In recent years, AI has been used to initiate change across a broad range of public and private sector organizations. In this article, we report findings from a subset of 50 interviews gathered in a wider study of interprofessional education (IPE) in which AI was employed as a change agent for implementing IPE in a number of health care institutions in a North American setting. A multiple case study approach. (Yin, 2002) was employed in the wider study and semi-structured interviews were undertaken with participants both before their IPE programs and directly afterwards to obtain a detailed understanding of their expectations and experiences of IPE. Interviews were analyzed in an inductive thematic manner in order to produce key emergent themes from each of the IPE programs. A process of re-analysis provided a set of themes which offered an understanding of the role of AI within this IPE initiative. Our findings identify a strong resonance and fit for AI both among the health and social care professionals who participated in this initiative. Numerous individuals commented on the enthusiasm and energy AI engendered, while praising its ability to enhance their working lives and interprofessional relationships. Yet a number of difficulties were also reported. These focused on problems with the translation of the AI process into achievable structural level (e.g. professional, cultural) changes. Based on these findings, the article goes on to argue that the use of AI can overlook a number of structural factors, which will ultimately limit its ability to actually secure meaningful and lasting change within health care.     

Risk factors and predictive model of adrenocortical insufficiency in patients with traumatic brain injury

BACKGROUND: Neuroendocrine dysfunction after traumatic brain injury (TBI) has received increased attention due to its impact on the recovery of neural function. The purpose of this study is to investigate the incidence and risk factors of adrenocortical insufficiency (AI) after TBI to reveal independent predictors and build a prediction model of AI after TBI.     

Mannose receptor and its putative ligands in normal murine lymphoid and nonlymphoid organs: In situ expression of mannose receptor by selected macrophages, endothelial cells, perivascular microglia, and mesangial cells, but not dendritic cells.

The mannose receptor (MR) has established roles in macrophage (Mphi) phagocytosis of microorganisms and endocytic clearance of host-derived glycoproteins, and has recently been implicated in antigen capture by dendritic cells (DCs) in vitro. MR is the founder member of a family of homologous proteins, and its recognition properties differ according to its tissue of origin. Given this heterogeneity and our recent discovery of a soluble form of MR in mouse serum, we studied the sites of synthesis of MR mRNA and expression of MR protein in normal mouse tissues. We demonstrate that synthesis and expression occur at identical sites, and that mature Mphi and endothelium are heterogeneous with respect to MR expression, additionally describing MR on perivascular microglia and glomerular mesangial cells. However, MR was not detected on DCs in situ, or on marginal zone or subcapsular sinus Mphi, both of which have MR-like binding activities. We also compared expression of MR to the binding of a recombinant probe containing the cysteine-rich domain of MR. We show that MR and its putative ligand(s) are expressed at nonoverlapping sites within lymphoid organs, consistent with a transfer function for soluble MR. Therefore, in addition to endocytic and phagocytic roles, MR may play an important role in antigen recognition and transport within lymphoid organs.     

一体化18F-FDG PET/MR血流代谢显像评估慢性单侧头颈动脉狭窄

目的 观察一体化¹⁸F-PET/MR血流代谢显像评估单侧头颈动脉狭窄的价值。方法 纳入68例慢性单侧头颈动脉狭窄患者,记录其美国国立卫生研究院卒中量表(NIHSS)及改良Rankin量表(mRS)评分;采用一体化PET/MR同步采集MR结构像、动脉自旋标记图像和¹⁸F-FDG PET图像,基于血流、代谢图像获取并比较幕上病变侧血流减低区、代谢减低区、血流代谢同步减低区及其各项指标,分析其与临床神经功能评分的相关性。结果 68例患者中,与血流减低区比较血流代谢同步减低区脑血流量(CBF)更低(P<0.001)、不对称指数(AI)更高(P<0.001);血流代谢同步减低区与代谢减低区标准摄取值比值(SUVR)差异无统计学意义(P=0.548),但前者AI更高(P<0.001)。同步减低区体积比与NIHSS、mRS均呈正相关(r_s=0.507、0.467,P均<0.001),SUVR AI与NIHSS、mRS均呈正相关(r_s=0.337、0.317,P均<0.05)。结论 一...     

Substitution of the carboxyl-terminal domain of apo AI with apo AII sequences restores the potential of HDL to reduce the progression of atherosclerosis in apo E knockout mice.

HDL metabolism and atherosclerosis were studied in apo E knockout (KO) mice overexpressing human apo AI, a des- (190-243)-apo AI carboxyl-terminal deletion mutant of human apo AI or an apo AI-(1-189)-apo AII-(12-77) chimera in which the carboxyl-terminal domain of apo AI was substituted with the pair of helices of apo AII. HDL cholesterol levels ranked: apo AI/apo E KO approximately apo AI-(1-189)-apo AII- (12-77)/apo E KO > > des-(190-243)-apo AI/apo E KO > apo E KO mice. Progression of atherosclerosis ranked: apo E KO > des-(190-243)-apo AI/apo E KO > > apo AI-(1-189)- apo AII-(12-77)/apo E KO approximately apo AI/apo E KO mice. Whereas the total capacity to induce cholesterol efflux from lipid-loaded THP-1 macrophages was higher for HDL of mice overexpressing human apo AI or the apo AI/apo AII chimera, the fractional cholesterol efflux rate, expressed in percent cholesterol efflux/microg apolipoprotein/h, for HDL of these mice was similar to that for HDL of mice overexpressing the deletion mutant and for HDL of apo E KO mice. This study demonstrates that the tertiary structure of apo AI, e.g., the number and organization of its helices, and not its amino sequence is essential for protection against atherosclerosis because it determines HDL cholesterol levels and not cholesterol efflux. Amino acid sequences of apo AII, which is considered to be less antiatherogenic, can be used to restore the structure of apo AI and thereby its antiatherogenicity.     

Clinical oxygen-enhanced magnetic resonance imaging of the lung

SUMMARY: Oxygen-enhanced magnetic resonance (MR) ventilation imaging is a new technique, and the full extent of its physiologic significance has not been elucidated. This review article includes (1) theory of oxygen enhancement; (2) respiratory physiology; (3) oxygen-enhanced MR imaging (MRI) sequence design; (4) basic study of oxygen-enhanced MRI in animal models and humans; (5) clinical study of oxygen-enhanced MRI; and (6) merits and demerits of the technique in comparison with hyperpolarized noble gas MR ventilation imaging. Oxygen-enhanced MRI provides not only ventilation-related information but also respiration-related information. Although application of oxygen-enhanced MR ventilation imaging to patients with pulmonary diseases has been limited, oxygen-enhanced MRI offers the possibility of demonstrating regional pulmonary function and substituting for nuclear medicine ventilation-perfusion study, when combined with MR perfusion imaging. We believe that further basic studies and clinical applications of this new technique will define the real significance of oxygen-enhanced MR ventilation imaging in the future of pulmonary functional imaging and its usefulness for diagnostic radiology.     

人工智能在角膜相关疾病领域的应用研究

人工智能(artificial intelligence，AI)是计算机领域的前沿科学，近年来在众多领域发展迅猛，其在眼科的研究和应用也日益增多。AI在角膜相关疾病领域的研究主要包括圆锥角膜的早期诊断及分级、角膜屈光手术相关评估、感染性角膜炎的分类及程度判断、角膜移植术后再干预的评估等，主要采用的算法包括神经网络、支持向量机及决策树，模型的灵敏度和特异度均达90%以上。AI可为医生提供客观的临床决策、为患者提供精准的治疗奠定基础。本文将对近年来AI在角膜相关疾病领域的应用研究进行综述。