Low-Dose CT Image Denoising Based on Improved WGAN-gp

In order to improve the quality of low-dose computational tomography (CT)images, the paper proposes an improved image denoising approach based on WGAN-gpwith Wasserstein distance. For improving the training and the convergence efficiency, thegiven method introduces the gradient penalty term to WGAN network. The novelperceptual loss is introduced to make the texture information of the low-dose imagessensitive to the diagnostician eye. The experimental results show that compared with thestate-of-art methods, the time complexity is reduced, and the visual quality of low-doseCT images is significantly improved.

Robust restoration of low-dose cerebral perfusion CT images using NCS-Unet

Cerebral perfusion computed tomography(PCT)is an important imaging modality for evaluating cerebrovascular diseases and stroke symptoms.With widespread public concern about the potential cancer risks and health hazards associated with cumulative radiation exposure in PCT imaging,considerable research has been conducted to reduce the radiation dose in X-ray-based brain perfusion imaging.Reducing the dose of X-rays causes severe noise and artifacts in PCT images.To solve this problem,we propose a deep learning method called NCS-Unet.The exceptional characteristics of non-subsampled contourlet transform(NSCT)and the Sobel filter are introduced into NCS-Unet.NSCT decomposes the convolved features into high-and low-frequency components.The decomposed high-frequency component retains image edges,contrast imaging traces,and noise,whereas the low-frequency component retains the main image information.The Sobel filter extracts the contours of the original image and the imaging traces caused by the contrast agent decay.The extracted information is added to NCS-Unet to improve its performance in noise reduction and artifact removal.Qualitative and quantitative analyses demonstrated that the proposed NCS-Unet can improve the quality of low-dose cone-beam CT perfusion reconstruction images and the accuracy of perfusion parameter calculations.

Hformer:highly efficient vision transformer for low-dose CT denoising

In this paper,we propose Hformer,a novel supervised learning model for low-dose computer tomography(LDCT)denoising.Hformer combines the strengths of convolutional neural networks for local feature extraction and transformer models for global feature capture.The performance of Hformer was verified and evaluated based on the AAPM-Mayo Clinic LDCT Grand Challenge Dataset.Compared with the former representative state-of-the-art(SOTA)model designs under different architectures,Hformer achieved optimal metrics without requiring a large number of learning parameters,with metrics of33.4405 PSNR,8.6956 RMSE,and 0.9163 SSIM.The experiments demonstrated designed Hformer is a SOTA model for noise suppression,structure preservation,and lesion detection.

Mortality outcomes of low-dose computed tomography screening for lung cancer in urban China:a decision analysis and implications for practice

Background: Mortality outcomes in trials of low-dose computed tomography(CT) screening for lung cancer are inconsistent. This study aimed to evaluate whether CT screening in urban areas of China could reduce lung cancer mortality and to investigate the factors that associate with the screening effect.Methods: A decision tree model with three scenarios(low-dose CT screening, chest X-ray screening, and no screening) was developed to compare screening results in a simulated Chinese urban cohort(100,000 smokers aged45-80 years). Data of participant characteristics were obtained from national registries and epidemiological surveys for estimating lung cancer prevalence. The selection of other tree variables such as sensitivities and specificities of low-dose CT and chest X-ray screening were based on literature research. Differences in lung cancer mortality(primary outcome), false diagnoses, and deaths due to false diagnosis were calculated. Sensitivity analyses were performed to identify the factors that associate with the screening results and to ascertain worst and optimal screening effects considering possible ranges of the variables.Results: Among the 100,000 subjects, there were 448,541, and 591 lung cancer deaths in the low-dose CT, chest X-ray, and no screening scenarios, respectively(17.2% reduction in low-dose CT screening over chest X-ray screening and 24.2% over no screening). The costs of the two screening scenarios were 9387 and 2497 false diagnoses and 7and 2 deaths due to false diagnosis among the 100,000 persons, respectively. The factors that most influenced death reduction with low-dose CT screening over no screening were lung cancer prevalence in the screened cohort, lowdose CT sensitivity, and proportion of early-stage cancers among low-dose CT detected lung cancers. Considering all possibilities, reduction in deaths(relative numbers) with low-dose CT screening in the worst and optimal cases were16(5.4%) and 288(40.2%) over no screening, respectively.Conclusions: In terms of mortality outcomes, our findings favor conducting low-dose CT screening in urban China.However, approaches to reducing false diagnoses and optimizing important screening conditions such as enrollment criteria for screening are highly needed.

Pediatric CT of the Lung: Influences on Image Quality

Objective: To assess influential factors of CT on image quality of the lung in children. Materials and methods: Retrospective evaluation of 82 consecutive chest-CT-scans in 50 children (1-16 years, 17 females and 33 males). Two pediatric radiologists evaluated in consensus the subjective image quality on lung windows using a 4-point scale (1 = very good, 2 = good, 3 = moderate, 4 = poor). Ventilation, motion artifacts and beam hardening artifact were included in this score. The effects of the following factors were evaluated: 1) CT-settings (tube energy, tube current);2) Patient’s age, weight, chest size, ventilation;3) Artifacts of devices, tubes and lines;4) Combination MRI of the abdomen prior to CT of the chest with the same sedation/anesthesia in oncological patients. Results: The odds of having a better image quality increase with patient’s age, weight and chest diameter in a multiple-factor model. There was no difference between tube current protocols. In infants (15 kg) subjective image quality was good in only 1 (8%), moderate in 8 (67%) and poor in 3 (25%) scans. In childhood and adolescence (15 - 90 kg) 25 (36%) scans were very good, 28 (40%) good, 15 (21%) moderate and 2 (3%) poor. Artifacts of tubes and lines have no statistical significant influence on image quality. Lower lung densities were related to better ventilation and older children. Conclusion: Increasing dose parameters may not increase necessarily subjective image quality in infants (15 kg), rather than good ventilation, optimal preparation and avoiding artifacts. A possible explanation of the rather moderate image quality in infants may be the alveolar stage of the lung. Up to two years of age the lung has a high specific lung volume per kg and a low total lung volume with a low alveolar surface.

Optimization of Image Quality in Retrospective Respiratory-Gated Micro-CT for Quantitative Measurements of Lung Function in Free-Breathing Rats

Objective: To optimize scan time and X-ray dose with no loss of image quality for retrospectively gated micro-CT scans of free-breathing rats. Methods: Five free-breathing rats were scanned using a dynamic micro-CT scanner over 10 continuous gantry rotations (50 seconds and entrance dose of 0.28 Gy). The in-phase projection views were selected and reconstructed, representing peak inspiration and end expiration from all 10 rotations and progressively fewer rotations. A least error method was also used to ensure that all angular positions were filled. Image quality and reproducibility for physiological measurements were compared for the two techniques. Results: The least error approach underestimated the lung volume, air content in the lung at peak inspiration, and tidal volume. Other measurements showed no differences between the projection-sorting techniques. Conclusions: Seven gantry rotations (35 seconds and 0.2 Gy dose) proved to be the optimal protocol for both the in-phase images and the least error images.

术前能谱CT多参数成像对晚期肺癌患者淋巴结转移的预测价值

目的探讨晚期肺癌患者术前采用能谱CT多参数成像检查对淋巴结转移的预测价值。方法回顾性收集2023年1—6月在医院经手术病理检查诊断为晚期肺癌的60例患者的临床资料,所有患者术前均接受能谱CT多参数成像检查,以手术病理学检查结果为金标准,绘制受试者工作特征(ROC)曲线分析能谱CT多参数成像参数预测晚期肺癌患者淋巴结转移的价值。结果经手术病理检查确诊,60例晚期肺癌患者中淋巴结转移患者19例,非淋巴结转移患者41例;淋巴结转移患者淋巴结能谱曲线斜率(λ_(HU))、淋巴结λ_(HU)/原发病灶λ_(HU)、淋巴结标准化碘密度(NIC)、淋巴结NIC/原发病灶NIC、淋巴结Neff-Z/原发病灶Neff-Z参数低于非淋巴结转移患者,原发病灶λ_(HU)、原发病灶NIC参数高于非淋巴结转移患者(P<0.05);绘制ROC曲线结果显示,淋巴结λ_(HU)/原发病灶λ_(HU)、淋巴结NIC/原发病灶NIC、淋巴结Neff-Z/原发病灶Neff-Z参数评估晚期肺癌患者淋巴结转移的曲线下面积均>0.7,具有一定预测价值。结论术前应用能谱CT多参数成像有助于诊断晚期肺癌患者是否存在淋巴结转移,且晚期肺癌患者术前采用能谱CT多参数成像检查对淋巴结转移具有一定的预测价值。

采用MSCT灌注成像检查评估周围型非小细胞肺癌分化程度的可行性分析

目的分析多层螺旋CT(MSCT)灌注成像检查评估周围型非小细胞肺癌(NSCLC)分化程度的可行性。方法选取本院2017年7月至2018年10月本院收治确诊的52例周围型NSCLC患者作为研究对象,比较不同分化级别患者的MSCT灌注成像参数;分析灌注参数与分化程度的相关性。结果高分化、中分化周围型NSCLC患者BF、BV、PS、MTT及PH数值均高于低分化周围型NSCLC,以高分化周围型NSCLC的BF、BV、PS、MTT及PH数值最高。各个灌注参数值,其中高分化、中分化周围型NSCLC的BF、PH与低分化周围型NSCLC比较差异显著(P<0.05),三者BV、PS及MTT数值比较,均为明显差异(P>0.05)。周围型NSCLC患者灌注参数BF、PH与其分化程度成负相关,且相关性显著(P<0.05)。结论MSCT灌注成像检查可有效反映周围型NSCLC的分化程度,其灌注参数中BF、PH对评估其分化程度有一定帮助,与周围型NSCLC分化程度具有一定相关性。

CT检查影像特征对小细胞肺癌的诊断价值

目的 探讨CT检查影像特征对小细胞肺癌的诊断价值。方法 选取26例小细胞肺癌(SCLC)患者,作为SCLC组,采用倾向匹配法按1∶2的比例对性别、年龄进行匹配,共选取52例非小细胞肺癌(NSCLC)患者,作为NSCLC组。比较两组患者的CT影像特征,将差异有统计学意义的CT影像特征纳入多因素Logistic回归分析,分析SCLC的影响因素;绘制受试者工作特征(ROC)曲线,计算曲线下面积(AUC),评估CT影像特征对SCLC的诊断价值。结果 两组患者分型、病灶形态、支气管闭塞和/或肺不张、肺门纵隔淋巴结情况、血管受累情况比较,差异均有统计学意义(P﹤0.05)。多因素Logistic回归分析结果显示,周围型肺癌、有支气管闭塞和/或肺不张均为SCLC的独立保护因素(P﹤0.05),肿瘤形态不规则、肺门纵隔淋巴结肿大、有血管受累均为SCLC的独立危险因素(P﹤0.05)。ROC曲线显示,病灶形态、血管受累情况对SCLC具有中等诊断价值,肺门纵隔淋巴结肿大情况、分型、支气管闭塞和/或肺不张情况对SCLC具有较低诊断价值,联合模型诊断SCLC的AUC为0.989(95%CI:0.893~0.997),具有较高诊断价值,均高于各指标单独检测(P﹤0.05)。结论 CT检查见中央型、病灶形态不规则、有肺门纵隔淋巴结肿大、血管受累任意一项可提示为SCLC,上述指标联合检查对SCLC的诊断价值较高。

SOSNet:一种非对称编码器-解码器结构的非小细胞肺癌CT图像分割模型

非小细胞肺癌严重损害人类健康,早期非小细胞肺癌CT(Computed Tomography)图像中的肿瘤结节体积小,不易发现,极易造成漏诊和误诊.为了精确分割非小细胞肺癌CT图像中的小体积肿瘤结节,本文提出SOSNet(Small Object Segmentation Networks)自动分割模型,利用ResNet(Residual Network)基础层和空洞卷积构造非对称编码器-解码器结构作为分割主网络,利用轴向取反注意力模块ARA(Axial Reverse Attention)逐步擦除背景中对分割有影响的结构,再使用结构细化模块SR(Structure Refinement)对主网络输出的粗略特征图进行结构细化,从而实现非小细胞肺癌肿瘤结节分割.在非小细胞肺癌公开数据集的实验测试表明,本文提出的小目标自动分割模型SOSNet可以有效分割出非小细胞肺癌CT图像中的小体积肿瘤结节,其mDice(mean-Dice)、mIoU(mean Intersection over Union)、Sensitivity、F1、Specificity、平均绝对误差MAE(Mean Absolute Error)均优于当前最先进的小目标分割模型CaraNet(Context Axial Reverse Attention Network).

CT影像征象对非小细胞肺癌胸膜浸润的影响分析

目的:基于CT影像征象探究非小细胞肺癌胸膜浸润的影响因素。方法:回顾性分析2021年1月至2023年5月于我院就诊的78例非小细胞肺癌患者临床资料,其中发生脏层胸膜浸润者(n=40)为浸润组,未发生脏层胸膜浸润者(n=38)为无浸润组。采用多因素Logistic回归分析非小细胞肺癌胸膜浸润的影响因素。结果:两组患者的肿瘤部位、病理类型、分化程度情况均无明显差异(P>0.05),肿瘤直径、N分期、M分期差异具有统计学意义(P<0.05)。两组患者CT影像空泡/空洞征、支气管充气征、分叶征、毛刺征、血管集束征、及胸腔积液等征象均无明显差异(P>0.05),胸膜凹陷征、病灶紧贴胸壁情况差异具有统计学意义(P<0.05)。回归分析显示,肿瘤直径、N分期、M分期、胸膜凹陷征及病灶紧贴胸壁是NSCLC胸膜浸润的特征指标(P<0.05)。结论:部分CT征象与NSCLC胸膜转移具有相关性,肿瘤直径、N分期、M分期、胸膜凹陷征、病灶紧贴胸壁均为NSCLC患者胸膜浸润的特征指标。

A New Medical Image Enhancement Algorithm Based on Fractional Calculus

The enhancement of medical images is a challenging research task due to the unforeseeable variation in the quality of the captured images.The captured images may present with low contrast and low visibility,which might inuence the accuracy of the diagnosis process.To overcome this problem,this paper presents a new fractional integral entropy(FITE)that estimates the unforeseeable probabilities of image pixels,posing as the main contribution of the paper.The proposed model dynamically enhances the image based on the image contents.The main advantage of FITE lies in its capability to enhance the low contrast intensities through pixels’probability.Initially,the pixel probability of the fractional power is utilized to extract the illumination value from the pixels of the image.Next,the contrast of the image is then adjusted to enhance the regions with low visibility.Finally,the fractional integral entropy approach is implemented to enhance the low visibility contents from the input image.Tests were conducted on brain MRI,lungs CT,and kidney MRI scans datasets of different image qualities to show that the proposed model is robust and can withstand dramatic variations in quality.The obtained comparative results show that the proposed image enhancement model achieves the best BRISQUE and NIQE scores.Overall,this model improves the details of brain MRI,lungs CT,and kidney MRI scans,and could therefore potentially help the medical staff during the diagnosis process.

Automated Classification of Lung Diseases in Computed Tomography Images Using a Wavelet Based Convolutional Neural Network

Recently, convolutional neural networks (CNNs) have been utilized in medical imaging research field and have successfully shown their ability in image classification and detection. In this paper we used a CNN combined with a wavelet transform approach for classifying a dataset of 448 lung CT images into 4 categories, e.g. lung adenocarcinoma, lung squamous cell carcinoma, metastatic lung cancer, and normal. The key difference between the commonly-used CNNs and the presented method is that in this method, we adopt the use of redundant wavelet coefficients at level 1 as inputs to the CNN, instead of using original images. One of the main advantages of the proposed method is that it is not necessary to extract regions of interest from original images. The wavelet coefficients of the entire image are used as inputs to the CNN. We compare the classification performance of the proposed method to that of an existing CNN classifier and a CNN-based support vector machine classifier. The experimental results show that the proposed method outperforms the other two methods and achieve the highest overall accuracy of 91.9%. It demonstrates the potential for use in classification of lung diseases in CT images.

18F-FDG PET/CT显像在肺癌术前分期诊断及复发转移预测中的应用价值

目的分析放射性荧光脱氧葡萄糖(18F-FDG)正电子发射断层扫描(PET)/计算机断层扫描(CT)显像在肺癌术前分期诊断及复发转移预测中的应用价值。方法 回顾性分析2022年9月至2023年6月期间80例初诊肺癌患者的临床和影像学数据,所有患者均在术前1周内进行了18F-F DG P ET/CT显像检查,并在术后3~6个月内进行了复查,监测复发或转移情况。术前的TNM分期和术后的复发转移情况均以手术病理结果或临床随访结果为金标准进行评估。结果术前分期诊断中,18F-FD(G P ET/CT显像的T分期、N分期和M分期的符合率分别为86.59%、81.93%和100%,一致性检验Kappa值分别为0.834、0.793和1.000。术后的复发转移检测中,18F-FDG PET/CT显像在术后6个月内成功检出了22例(88.00%)的复发转移病例,其诊断灵敏度为88.00%,特异度为100.00%。结论18F-FDG PET/CT显像在肺癌术前分期以及术后复发转移的预测中具有较高的准确性和可靠性,该方法可以为临床提供有效的参考信息,有助于医生制定更准确的治疗方案和更有效的随访策略。

基于双重注意力机制的间质性肺病高分辨率CT图像分类方法

为了更精确地分类间质性疾病,提出了一种基于深度学习的分类网络,首先将多头自注意力机制模DenseNet-121结合,使得模型能够同时关注多个重点区域。然后采用卷积注意力模块实现更高效的特征提取,提升网络的空间感知能力,从而增强分类性能。最后,添加改进的空间金字塔池化层将不同尺度的特征图拼接起来以捕获更丰富的空间信息。此外针对高分辨率C图像数据集类别不均衡问题,引入FocalLoss损失函数,使得模型在训练时更专注于难分类的样本,从而进一步增强模型的分类能力。所提方法在未经训练的数据集上进行测试,达到了88.28%的准确率。相较于原始DenseNet-121在准确率、召回率、精确率、F1分数和Kappa系数提高了4.65%、5.08%、5.82%、5.45%和6.38%。实验结果表明,该方法具有特征提取能力强和分类准确率高的特点。

基于CT数据的肺部影像可视化系统设计

设计一种基于CT数据的肺部影像二维可视化与三维重建系统,首先对DICOM图像进行解析,分割和标记出肺结节的位置;然后利用CT序列的重采样、面绘制的三维重建、形态学处理等技术,实现肺实质和结节的多视角、多分辨率三维显示;最后设计交互界面,包括图像增强、肺部二维可视化、结节勾勒、肺实质和结节三维重建、旋转、缩放切换视角等功能。实验表明,本系统对于二维图像的可视化和病灶区域勾勒位置清晰、准确,并使三维图像呈现的结节完整且光滑。本系统相较于已有的类似医学处理软件,大幅度提高重建和可视化效率,使医生能够更加快速、精确地观察三维图像,辅助疾病诊断和手术方案制定。

规范化流程管理在非小细胞肺癌患者PET-CT检查中的应用效果

目的:探讨规范化流程管理在非小细胞肺癌患者PET-CT检查中的应用效果,为确保临床非小细胞肺癌的诊疗质量提供参考依据。方法:选取2019年3月—2021年10月河南省肿瘤医院收治的112例非小细胞肺癌患者作为研究对象,按照随机数表法分组,患者均接受PET-CT检查,且检查期间,予以对照组(56例)患者原工作流程操作,予以观察组(56例)患者规范化流程管理,比较两组患者图像合格率、检查前准备完好率、检查时准备完好率、检测的等待时间和检查时间,干预前后焦虑、抑郁情绪评分,以及患者对护理工作的满意度。结果:观察组患者检查前准备完好率、检查时准备完好率高于对照组,差异有统计学意义(χ^(2)=5.617、1.734,P<0.05),同时观察组患者检测的等待时间和检查时间均较对照组缩短,差异有统计学意义(t=5.469、4.941,P<0.05)。与干预前比,干预后两组患者情绪焦虑自评量表(SAS)、抑郁自评量表(SDS)评分降低,且试验组患者低于对照组,差异有统计学意义(t=11.298、8.414,P<0.05)。观察组患者对护理工作的满意度高于对照组,差异有统计学意义(χ^(2)=4.350,P<0.05)。结论:非小细胞肺癌患者PET-CT检查中应用规范化流程管理可缩短患者的检查等待和检查时间,提升诊疗效率,并取得良好的图像质量,患者对护理工作的满意度高。

利用跨模态轻量级YOLOv5模型的PET/CT肺部肿瘤检测

多模态医学图像可在同一病灶处提供更多语义信息,针对跨模态语义相关性未充分考虑和模型复杂度过高的问题,该文提出基于跨模态轻量级YOLOv5(CL-YOLOv5)的肺部肿瘤检测模型。首先,提出学习正电子发射型断层显像(PET)、计算机断层扫描(CT)和PET/CT不同模态语义信息的3分支网络;然后,设计跨模态交互式增强块充分学习多模态语义相关性,余弦重加权计算Transformer高效学习全局特征关系,交互式增强网络提取病灶的能力;最后,提出双分支轻量块,激活函数簇(ACON)瓶颈结构降低参数同时增加网络深度和鲁棒性,另一分支为密集连接的递进重参卷积,特征传递达到最大化,递进空间交互高效地学习多模态特征。在肺部肿瘤PET/CT多模态数据集中,该文模型获得94.76%mAP最优性能和3238 s最高效率,以及0.81 M参数量,较YOLOv5s和EfficientDet-d0降低7.7倍和5.3倍,多模态对比实验中总体上优于现有的先进方法,消融实验和热力图可视化进一步验证。

双源CT灌注成像参数对非小细胞肺癌患者化疗效果的预测价值

目的:探讨双源CT灌注成像参数对非小细胞肺癌患者化疗效果的预测价值。方法:回顾性分析2020年6月至2022年6月该院收治的124例非小细胞肺癌患者的临床资料,入院后均给予化疗治疗,治疗前后分别进行双源CT灌注成像检查,根据化疗效果分为敏感组(n=81)和不敏感组(n=43),比较两组治疗前后双源CT灌注成像参数[血流量(BF)、血容量(BV)、表面通透性(PS)],并绘制受试者工作特征(ROC)曲线,分析双源CT灌注成像参数对非小细胞肺癌患者化疗效果的预测价值。结果:治疗前,敏感组BF、PS、BV均高于不敏感组,差异有统计学意义(P<0.05);治疗后,两组BF、PS、BV均低于治疗前,且敏感组低于不敏感组,差异有统计学意义(P<0.05);ROC曲线结果显示,BF、PS、BV单项及联合检测预测非小细胞肺癌患者化疗效果的曲线下面积分别为0.767、0.824、0.803、0.932,联合检测的预测价值最高。结论:双源CT灌注成像参数联合检测预测非小细胞肺癌患者化疗效果的价值高于三者单项检测。

基于Transformer的肺肿瘤三维CT图像分割

基于信息学技术自动分割病人的肺部CT图像,有助于医生对于肺癌患者的早期诊断,提取和整合图像区域间的空间关联,对于提升肺肿瘤分割性能是十分重要的。本文提出了一个新的基于Transformer的分割模型,用于肺肿瘤三维CT图像分割、学习和整合此类关联。本文分别设计了带有混合多头图像区域节点注意力的Transformer模块和类别注意力模块,学习并融合了肺部CT图像的空间层面和通道层面的信息。将新的基于Transformer的分割模型同其他较为先进的模型进行了对比实验,实验结果表明新的模型在骰子系数、交并比和豪斯多夫距离等方面优于其他模型。