基于V-Net卷积神经网络深度学习模型自动分割腰椎CT图像中的椎旁肌

目的 观察基于V-Net卷积神经网络(CNN)的深度学习(DL)模型自动分割腰椎CT图像中的椎旁肌的价值。方法 收集471例接受腰椎CT检查患者,按7∶3比例将其分为训练集(n=330)和测试集(n=141);采用2D V-Net进行训练,建立DL模型;观察其分割腰大肌、腰方肌、椎后肌群及椎旁肌的价值。结果 基于V-Net CNN的DL模型分割椎旁肌精度良好,戴斯相似系数(DSC)均较高、肌肉横截面积误差率(CSA error)均较低;其分割训练集图像中的腰大肌、腰方肌及椎旁肌的DSC均高于测试集(P均<0.05),而分割训练集中4组肌肉的CSA error均低于测试集(P均<0.05)。测试集内两两比较结果显示,该模型分割椎后肌群的DSC最高、腰方肌的DSC最低;分割腰方肌的CSA error最高、椎旁肌的CSA error最低(P均<0.05)。结论 以基于V-Net的DL模型自动分割椎旁肌的效能较佳。     

基于深度学习构建模型识别下肢全长正位X线片中的下肢力线关键点及自动测量关键角度

目的 基于深度学习（DL）方法构建自动测量下肢全长正位X线片关键角度模型,评估其临床应用价值。方法 回顾性选取634幅下肢全长正位X线片,由5名骨科医师分别标注下肢力线关键点,包括髋关节中心、股骨髁间窝顶点、胫骨髁间嵴中点、股骨内侧和外侧髁最低点、胫骨内侧和外侧平台最低点、距骨宽度中点,并建立数据集。采用高分辨率网络（HRNet）进行迁移学习,构建自动检测关键点模型,以5折交叉验证筛选最优模型,确定关键点坐标后,通过余弦定律计算关键角度机械股骨远端外侧角（mLDFA）、胫骨近端内侧角（MPTA）、股骨胫骨关节线夹角（JLCA）及髋-膝-踝角（HKA）,实现自动测量关键角度,并以关键点自动检测模型和通过余弦定律计算所得关键角度共同构建自动测量关键角度模型。随机选取50幅图像,由另3名骨科医师手动测量上述关键角度,评估自动测量关键角度模型与医师测量结果的一致性。结果 3名骨科医师所测mLDFA、MPTA、JLCA及HKA的均值分别为（88.50±2.59）°、（86.41±2.25）°、（2.90±2.27）°及（174.62±3.97）°;自动测量关键角度模型所获结果分别为（88.48±2.60）°、（86.52±2.57）°、（3.11±2.41）°及（174.53±3.99）°,与医师测量结果的一致性较好（ICC=0.897、0.888、0.826、0.996）。结论 所构建的自动测量下肢全长正位X线片关键角度模型有助于识别骨科关键点和测量关键角度。     

基于胸部CT身体成分自动分析系统评估肌肉及脂肪含量

目的 构建基于胸部CT的身体成分自动分析系统,观察其评估胸部肌肉及脂肪含量的价值。方法 收集108例肺炎患者T7~T8层面轴位胸部CT图像(分割数据集),于COVID 19-CT数据集筛选984例胸部CT数据(随机抽取10例为整体测试数据集,余974例为选层数据集);基于卷积神经网络(CNN)衍生网络,包括ResNet、ResNeXt、MobileNet、ShuffleNet、DenseNet、EfficientNet及ConvNeXt,于选层数据集中分类T7~T8层面,以准确率、精确率、召回率及特异度进行评价;基于经典全CNN(FCN)衍生网络,包括FCN、SegNet、UNet、Attention UNet、UNet++、nnUNet、UNeXt及CMUNeXt于分割数据集中分割骨骼肌(SM)、皮下脂肪组织(SAT)、肌间脂肪组织(IMAT)及内脏脂肪组织(VAT),以戴斯相似系数(DSC)、交并比(IoU)及95豪斯多夫距离(HD)进行评价;基于表现最优的选层网络及分层网络构建身体成分自动分析系统,对整体测试数据集进行测试,以平均绝对误差(MAE)、均方根误差(RMSE)及MAE的标准差(SD)进行评价。结果 DenseNet网络自动于完整胸部CT图中分类T7~T8层面的准确率、精确率、召回率及特异度分别为95.06%、84.83%、92.27%及95.78%,均高于其余选层网络。在分割SM、SAT、IMAT及整体分割方面,UNet++网络DSC及IoU均高于、而95HD均低于其余分割网络。以DenseNet为选层网络、UNet++为分割网络测试整体测试数据集,其预测SM、SAT、IMAT及VAT的MAE分别为27.09、6.95、6.65及3.35 cm²。结论 基于胸部CT身体成分自动分析系统可用于评估胸部肌肉及脂肪含量;其中最佳分割网络UNet++分割脂肪组织精准度优于SM。     

肌骨系统X线平片基本征象

随着科学技术的进步,CT、MRI等大型医疗设备已广泛应用到临床各系统的检查中,并且发挥越来越重要的作用.但是,X线平片仍然是骨骼肌肉系统最基本和最常用的影像学检查,全面了解和掌握骨骼肌肉系统病变的基本X线平片征象是很有必要的.     

256层CT肝脏体积自动测量研究

目的评估采用256层CT进行肝脏体积测量(liver volume,LV)不同方法的精确性和一致性。方法收集20例肝脏无明显异常因其他原因行256层螺旋CT上腹部检查患者的平扫及增强扫描图像资料。利用手工勾勒法及3D自动法测量LV,与标准化肝体积(standardize liver volume,SLV)进行相关性、双因素方差及Bland-Altman分析,并记录测量LV两种方法平均用时。结果手工法所测LV为(1 277.97±99.83)cm3,自动法所测LV为(1 247.77±116.27)cm3,SLV为(1 252.34±121.85)cm3;手工法、自动法所测LV与SLV分别呈显著正相关(r=0.8754、0.9066,P均<0.01),两种方法所测LV与SLV之间差异均无明显统计学意义(P均>0.05);手工法所测LV数值更接近SLV,但平均测量时间3D自动法为(7.6±1.8)min,较手工法(22.5±5.7)min短。结论肝脏体积自动测量法是一种较好的活体测量肝体积的方法。     

基于深度学习计算机辅助诊断系统测量脑出血量

目的探讨基于深度学习计算机辅助诊断系统(DL-CAD)测量脑出血量的应用价值。方法回顾性分析超急性期、急性期、亚急性期各50例脑出血患者脑部平扫CT资料,分别采用DL-CAD、多田公式法和CT定量法测量脑出血量,记录测量时间,并计算绝对百分误差率(APE)。比较3种方法测量结果、测量时间及APE差异。结果应用DL-CAD测量超急性期及亚急性期脑出血量结果与CT定量法差异无统计学意义(P均>0.05),而对急性期出血差异有统计学意义(P<0.05)。应用多田公式法测量脑出血结果与CT定量法差异均有统计学意义(P均<0.01)。采用DL-CAD测量不同时期脑出血量,测量时间均显著低于CT定量法及多田公式法(P均<0.01),且APE均低于多田公式法(P均<0.01)。结论与传统多田公式法相比,DL-CAD测量各期脑出血量准确性更高,且速度更快。     

基于特征金字塔网络自动分割平扫CT所示自发性脑出血血肿并判断其语义特征

目的 观察基于特征金字塔网络(FPN)自动分割平扫CT所示自发性脑出血(sICH)血肿并判断其语义特征的价值。方法 回顾性收集A医院408例(训练集)及B医院103例(验证集)sICH平扫CT图像;基于FPN构建深度学习(DL)分割模型分割血肿区域,并以交并比(IoU)、戴斯相似系数(DSC)及准确率评价其效能;以DL分类模型判断血肿语义特征,绘制受试者工作特征曲线,计算曲线下面积(AUC),评估其识别sICH血肿语义特征的效能。结果 DL分割模型分割训练集95% sICH血肿的IoU、DSC及准确率分别为0.84±0.07、0.91±0.04及(88.78±8.04)%,在验证集分别为0.83±0.07、0.91±0.05及(88.59±7.76)%。DL分类模型识别训练集及验证集sICH血肿不规则形态、不均匀密度、卫星征、混杂征及漩涡征的AUC分别为0.946~0.993及0.714~0.833。结论 基于FPN可准确、高效地自动分割sICH血肿,对于判断血肿语义特征亦具有较高效能。     

二尖瓣脱垂合并胸廓骨骼畸形

本文对83例二尖瓣脱垂进行了X线分析,发现有40例至少合并有一种胸廓骨骼畸形,占48％;其中扁胸直背27例,占27％,胸椎侧弯9例,占11％,漏斗胸4例,占5％。指出,胸廓骨骼畸形是二尖瓣脱垂的重要征象之一。胸廓骨骼畸形也可能影响二尖瓣脱垂的X线诊断。     

肌肉骨骼超声与X线在痛风性关节炎诊断中的价值分析

目的:探讨肌肉骨骼超声与X线在痛风性关节炎诊断中的价值.方法:回顾性选取2018年1月—2019年12月我院收治的100例痛风性关节炎患者作为研究对象,其中男性患者58例,女性42例,均采用肌肉骨骼超声与X线进行关节检查,评估肌肉骨骼超声与X线在痛风性关节炎中的早期诊断价值.结果:肌肉骨骼超声检出痛风性关节炎55例,与...     

罕见先天性右侧足缺如1例报道

患儿,女,53天,维吾尔族,出生后发现右足缺如,来我院就诊。查体：营养发育中等,心肺听诊正常,腹软,肝、脾无肿大,腹部无压痛,全身皮肤无黄染。专科检查：双下肢基本等长．右足缺如,残端皮肤完整,未见破溃及红肿,健侧肢体发育正常。X线表现见图1。     

Deep learning-based automatic blood pressure measurement: evaluation of the effect of deep breathing,talking and arm movement

Abstract

Objectives: It is clinically important to evaluate the performance of a newly developed blood pressure (BP) measurement method under different measurement conditions. This study aims to evaluate the performance of using deep learning-based method to measure BPs and BP change under non-resting conditions.

Materials and methods: Forty healthy subjects were studied. Systolic and diastolic BPs (SBPs and DBPs) were measured under four conditions using deep learning and manual auscultatory method. The agreement between BPs determined by the two methods were analysed under different conditions. The performance of using deep learning-based method to measure BP changes was finally evaluated.

Results: There were no significant BPs differences between two methods under all measurement conditions (all p?>?.1). SBP and DBP measured by deep learning method changed significantly in comparison with the resting condition: decreased by 2.3 and 4.2?mmHg with deeper breathing (both p?<?.05), increased by 3.6 and 6.4?mmHg with talking, and increased by 5.9 and 5.8?mmHg with arm movement (all p?<?.05). There were no significant differences in BP changes measured by two methods (all p?>?.4, except for SBP change with deeper breathing).

Conclusion: This study demonstrated that the deep learning method could achieve accurate BP measurement under both resting and non-resting conditions.

• Key messages

• Accurate and reliable blood pressure measurement is clinically important. We evaluated the performance of our developed deep learning-based blood pressure measurement method under resting and non-resting measurement conditions.

• The deep learning-based method could achieve accurate BP measurement under both resting and non-resting measurement conditions.

     

Miniature Silicon diode matrix-detector for in vivo measurement of 133Xenon disappearance in the canine myocardium following local tissue injection

Summary. After local tissue depositioning of ¹³³Xenon (¹³³Xe) the regional washout is usually registered by a Nal(Tl) detector. The residual radioactivity of ¹³³Xe is usually measured at its 81 keV photopeak. However, using small Silicon (Si) photodiodes it is feasible to measure only the low-energy activity in the X-ray energy range. In the myocardium of open chest dogs ¹³³Xe washout measurements by a matrix of Si diodes composed in a 4×4 array and a conventional NaI(TI) detector were carried out simultaneously. Fourteen separate pairs of measurements were performed in 3 dogs. When the Si-diodes in the matrix were selected individually in accordance to the position with reference to the diode with maximum count rate or pooled, comparisons could be made between the corresponding washout rate constants measured by the reference detector. In the correlation between the rate constants the intercepts with the y axis were not significantly different from zero allowing the correlation lines to be fitted through (0.0). The slope of the correlation line was close to unity. The registration of the low-X-ray energy of ¹³³Xe by the Si-detectors is an alternative to the conventional high energy activity recording appearing from the gamma-energy of the photopeak. The detector matrix concept allows elimination of motion artefacts and indicator distribution in the myocardial tissue. Due to the uniformity and low cost of Si-diodes the perspective may be the introduction as a disposable transducer useful during cardiac surgery for example.     

Deep learning-based image analysis for automated measurement of eyelid morphology before and after blepharoptosis surgery

Background and aimEyelid position and contour abnormality could lead to various diseases, such as blepharoptosis, which is a common eyelid disease. Accurate assessment of eyelid morphology is important in the management of blepharoptosis. We aimed to proposed a novel deep learning-based image analysis to automatically measure eyelid morphological properties before and after blepharoptosis surgery.MethodsThis study included 135 ptotic eyes of 103 patients who underwent blepharoptosis surgery. Facial photographs were taken preoperatively and postoperatively. Margin reflex distance (MRD) 1 and 2 of the operated eyes were manually measured by a senior surgeon. Multiple eyelid morphological parameters, such as MRD1, MRD2, upper eyelid length and corneal area, were automatically measured by our deep learning-based image analysis. Agreement between manual and automated measurements, as well as two repeated automated measurements of MRDs were analysed. Preoperative and postoperative eyelid morphological parameters were compared. Postoperative eyelid contour symmetry was evaluated using multiple mid-pupil lid distances (MPLDs).ResultsThe intraclass correlation coefficients (ICCs) between manual and automated measurements of MRDs ranged from 0.934 to 0.971 (p < .001), and the bias ranged from 0.09 mm to 0.15 mm. The ICCs between two repeated automated measurements were up to 0.999 (p < .001), and the bias was no more than 0.002 mm. After surgery, MRD1 increased significantly from 0.31 ± 1.17 mm to 2.89 ± 1.06 mm, upper eyelid length from 19.94 ± 3.61 mm to 21.40 ± 2.40 mm, and corneal area from 52.72 ± 15.97 mm² to 76.31 ± 11.31mm² (all p < .001). Postoperative binocular MPLDs at different angles (from 0° to 180°) showed no significant differences in the patients.ConclusionThis technique had high accuracy and repeatability for automatically measuring eyelid morphology, which allows objective assessment of blepharoptosis surgical outcomes. Using only patients’ photographs, this technique has great potential in diagnosis and management of other eyelid-related diseases.