运用色度学分析技术对离体乳牙脱矿与再矿化时间的实验研究

目的运用数码照相和色度分析技术观测不同酸蚀时间对离体乳牙釉质脱矿以及再矿化时间的影响,以寻找乳牙合适的脱矿及再矿化时间。方法将60颗离体上颌乳中切牙牙冠唇面开窗包埋。标本随机分为4个处理组,32%的磷酸凝胶分别酸蚀开窗面15s(组1)、30s(组2)、45s(组3)和60s(组4),使开窗面脱矿,再将其置于人工唾液中分别再矿化8h、24h、48h、72h及1周。脱矿后采用2%亚甲基蓝对开窗面进行染色,数码照相,运用图像分析软件对图像进行色度分析,记录色度参数,运用SPSS19.0对实验数据进行统计学分析。结果酸蚀15s后染色,其色度值变化即可被计算机所识别;酸蚀15s再矿化72h、酸蚀30s再矿化48h、酸蚀45s和60s再矿化24h时的色度值与矿化前的色度值相比,差异具有统计学意义,其P值分别为0.01、0.037、0.047和0.011。结论运用色度学技术分析乳牙脱矿与再矿化的时间,提示乳牙釉质最合适的酸蚀时间为15s,再矿化时间为72h。     

优化扫描范围对CT冠状动脉成像辐射剂量和图像质量的影响

目的 初步探讨优化扫描范围(FOV)对冠状动脉CT成像(CTA)辐射剂量和图像质量的影响.方法 选取100例临床疑似冠心病的患者进行冠状动脉CTA检查,随机分成A、B两组,每组各50例.A组FOV为气管分叉处至心脏下缘1 cm;B组FOV根据平扫图像确定,即冠状动脉树上缘1 cm至心尖部心包即将消失的层面.记录患者增强扫描的剂量长度乘积(DLP),选择冠状动脉不同节段测量其对比剂增强水平.分析A、B两组的FOV、有效辐射剂量(ED)及图像质量有无统计学差异.结果 两组患者平均年龄、性别、体重指数(BMI)比较均无统计学意义;A组平均FOV、ED分别为(15.31±1.51)cm、(15.10±2.00) mSv,均高于B组的(13.51±1.68)cm、(12.02±1.40)mSv,且差异均有统计学意义(P＜0.05).两组共显示冠状动脉1303个节段,A组669个节段,B组634个节段,两组间比较差异无统计学意义(P＞0.05).右冠状动脉(RCA)近段、左冠状动脉前降支(LAD)近段及左回旋支(LCX)近段CT值水平差异无统计学意义.结论 根据平扫图像优化冠状动脉CTA FOV,既能有效降低辐射剂量,同时又能保证图像质量,为临床诊断冠心病提供参考信息.     

孕中期超声胎儿结构筛查在超声培训中的质量控制

目的:对孕中期胎儿超声结构筛查进行了图像评分和cumulative sum(CUSUM)曲线联合质量控制法,以提高超声培训中胎儿结构筛查的准确性。方法:选取胎儿中枢神经系统、心脏、消化系统等标准切面,共28幅图象,要求每位参加超声培训超声医生扫查100例孕中期胎儿,采集各个切面并存储图像;由两位上级医师评估每幅图像的标准程度并予以评分,将评分值与总分值进行CUSUM分析,将每位超声医生评分的失败率设定为10%。用组内差异性分析(ICC)分别对两位评估医生和第一位评估医生进行自身差异性分析。结果:超声医生1、2和3图象234分(90%)以上均在80幅以上;评分医生A1、A2与评分医生B评分之间的ICC分别是0.985和0.977;评分医生A1与评分医生A2之间的ICC是0.992。评分医生A对三位超声医生评分的CUSUM曲线显示了三位超声医生分别在筛查74、89和61次后筛查能力合格。结论:胎儿结构筛查图象评分及CUSUM曲线联合质量控制法,有效地提高了超声培训中超声医生产前结构筛查的质量,认证了其独立筛查能力。     

Kinematic patellar tracking from MR images for knee pain analysis

Kinematic approaches using MR images have been regarded of more accuracy in knee pain (AKP) detection than stationary approaches. However, the challenge in segmenting femur, patellar and tibia due to the intensity non-uniformity caused by magnetic propagation degradation in MR images, and the strong adhesion of the soft tissue around the knee organs, has restricted the use of kinematic approaches. This paper proposes a combinatorial based kinematic patellar tracking (CKPT) for AKP detection. The CKPT uses a hybrid approach for extracting knee organs, where an edge-constrained wavelet enhancement followed by moment preserving segmentation is employed for conquering the soft tissue adhesion for extracting the femur and tibia from axial MR images, and a sliding window based moment preserving for resolving the segmentation difficulty associated with intensity non-uniformity in sagittal MR images. The location constraints are then applied for extracting landmark points from femur and patellar, and three inclination angles reflecting patellar position and orientation, during leg movement, are calculated as the measurement of patellar dislocation. The experiment shows that the hybrid approach can accurately extract femur, patellar and tibia. It also demonstrates the prominent of the calculated inclination angles in detecting AKP.     

人工智能对胃肠疾病诊疗的推动作用

高速发展的计算机技术给日常生活及工作带来巨大变化。人工智能是计算机科学的一个分支,是让计算机去行使通常情况下具备智能生命才可能行使的活动,广义的人工智能涵盖机器学习和机器人等等,本文主要聚焦于机器学习与相关的医学领域,深度学习是机器学习中的人工神经网络,卷积神经网络(CNN)是深度神经网络的一种,是在深度神经网络基础上,进一步模仿大脑的视觉皮层构造和视觉活动原理而开发;目前在医疗大数据分析中应用的机器学习方式主要为CNN。在未来数年内,人工智能作为常规工具进入医学图像解读相关的科室是发展趋势。本文主要分享人工智能与生物医学的融合进展,并结合实际案例,重点介绍CNN在胃肠道疾病的病理诊断、影像学诊断及内镜诊断等方面的应用研究现状。     

BPBSAM: Body part-specific burn severity assessment model

Background and objectiveBurns are a serious health problem leading to several thousand deaths annually, and despite the growth of science and technology, automated burns diagnosis still remains a major challenge. Researchers have been exploring visual images-based automated approaches for burn diagnosis. Noting that the impact of a burn on a particular body part can be related to the skin thickness factor, we propose a deep convolutional neural network based body part-specific burns severity assessment model (BPBSAM).MethodConsidering skin anatomy, BPBSAM estimates burn severity using body part-specific support vector machines trained with CNN features extracted from burnt body part images. Thus BPBSAM first identifies the body part of the burn images using a convolutional neural network in training of which the challenge of limited availability of burnt body part images is successfully addressed by using available larger-size datasets of non-burn images of different body parts considered (face, hand, back, and inner forearm). We prepared a rich labelled burn images datasets: BI & UBI and trained several deep learning models with existing models as pipeline for body part classification and feature extraction for severity estimation.ResultsThe proposed novel BPBSAM method classified the severity of burn from color images of burn injury with an overall average F1 score of 77.8% and accuracy of 84.85% for the test BI dataset and 87.2% and 91.53% for the UBI dataset, respectively. For burn images body part classification, the average accuracy of around 93% is achieved, and for burn severity assessment, the proposed BPBSAM outperformed the generic method in terms of overall average accuracy by 10.61%, 4.55%, and 3.03% with pipelines ResNet50, VGG16, and VGG19, respectively.ConclusionsThe main contributions of this work along with burn images labelled datasets creation is that the proposed customized body part-specific burn severity assessment model can significantly improve the performance in spite of having small burn images dataset. This highly innovative customized body part-specific approach could also be used to deal with the burn region segmentation problem. Moreover, fine tuning on pre-trained non-burn body part images network has proven to be robust and reliable.     

自MR-T2图像勾画肝细胞癌患者放疗靶区精度研究

目的 研究MRI-T2加权像扫描(MR-T2)图像对肝细胞癌(HCC)患者放疗靶区勾画精度的影响。方法 2019年5月～2020年5月我院诊治的36例HCC患者,在接受放疗前行CT和MRI检查,勾画大体肿瘤体积(GTV),经形变配准后获得大体内靶区(IGTV)和计划靶区(PTV)。应用Reg Refine图像形变配准。记录2%、50%、95%和98%PTV体积的受照剂量(分别记为D2、D50、D95和D98)。结果 经Reg Refine配准后,门静脉x、y和z轴向位移分别为(0.5±0.2)mm、(0.8±0.3)mm和(0.7±0.3)mm,显著小于自动形变配准【分别为(2.9±0.4)mm、(4.8±0.5)mm和(3.9±0.6)mm,P<0.05】,腹腔干x、y和z轴向位移分别为(0.8±0.4)mm、(0.7±0.3)mm和(0.6±0.3)mm,显著小于自动形变配准【分别为(3.2±0.5)mm、(3.0±0.7)mm和(2.4±0.4)mm,P<0.05】;形变配准后GTV和IGTV分别为(379.8±103.9)cm³和(430.1±120.6)cm³,显著大于形变配准前【分别为(330.4±92.8)cm³和(375.2±95.5)cm³,P<0.05】, PTV为(542.7±130.2)cm³,显著小于形变配准前【(690.2±150.3)cm³,P<0.05】;形变配准后D2、D50、D95和D98靶区剂量分别为(56.0±12.9)Gy、(53.0±6.8)Gy、(49.8±5.3)Gy和(52.9±8.1)Gy,与形变配准前【(55.7±13.2)Gy、(52.8±7.3)Gy、(50.1±4.6)Gy和(53.3±7.7)Gy】比,差异均无统计学意义(P>0.05);形变配准后肝脏V5、V10、V20、V30和V40受照剂量分别为(52.7±9.4)%、(32.0±5.6)%、(20.1±4.6)%、(13.5±5.2)%和(7.4±3.8)%,显著低于配准前【分别为(58.2±10.1)%、(39.3±7.5)%、(24.8±5.8)%、(17.2±4.4)%和(10.1±3.9)%,P<0.05】。结论 应用MR-T2图像勾画HCC放疗靶区有助于扩大靶区范围和运动轨迹,提高肝癌靶区勾画精度。