Identification and Localization of Endotracheal Tube on Chest Radiographs Using a Cascaded Convolutional Neural Network Approach

Journal of Digital Imaging - Rapid and accurate assessment of endotracheal tube (ETT) location is essential in the intensive care unit (ICU) setting, where timely identification of a mispositioned...     

Using Transfer Learning of Convolutional Neural Network on Neck Radiographs to Identify Acute Epiglottitis

Journal of Digital Imaging - Acute epiglottitis (AE) is a life-threatening condition and needs to be recognized timely. Diagnosis of AE with a lateral neck radiograph yields poor reliability and...     

Detecting Distal Radial Fractures from Wrist Radiographs Using a Deep Convolutional Neural Network with an Accuracy Comparable to Hand Orthopedic Surgeons

In recent years, fracture image diagnosis using a convolutional neural network (CNN) has been reported. The purpose of the present study was to evaluate the ability of CNN to diagnose distal radius fractures (DRFs) using frontal and lateral wrist radiographs. We included 503 cases of DRF diagnosed by plain radiographs and 289 cases without fracture. We implemented the CNN model using Keras and Tensorflow. Frontal and lateral views of wrist radiographs were manually cropped and trained separately. Fine-tuning was performed using EfficientNets. The diagnostic ability of CNN was evaluated using 150 images with and without fractures from anteroposterior and lateral radiographs. The CNN model diagnosed DRF based on three views: frontal view, lateral view, and both frontal and lateral view. We determined the sensitivity, specificity, and accuracy of the CNN model, plotted a receiver operating characteristic (ROC) curve, and calculated the area under the ROC curve (AUC). We further compared performances between the CNN and three hand orthopedic surgeons. EfficientNet-B2 in the frontal view and EfficientNet-B4 in the lateral view showed highest accuracy on the validation dataset, and these models were used for combined views. The accuracy, sensitivity, and specificity of the CNN based on both anteroposterior and lateral radiographs were 99.3, 98.7, and 100, respectively. The accuracy of the CNN was equal to or better than that of three orthopedic surgeons. The AUC of the CNN on the combined views was 0.993. The CNN model exhibited high accuracy in the diagnosis of distal radius fracture with a plain radiograph.

     

Ultra-Low-Dose Spectral CT Based on a Multi-level Wavelet Convolutional Neural Network

Journal of Digital Imaging - Spectral computed tomography (CT) based on a photon-counting detector (PCD) is a promising technique with the potential to improve lesion detection, tissue...     

Convolutional Neural Network to Segment Laminae on 3D Ultrasound Spinal Images to Assist Cobb Angle Measurement

A recent innovation in scoliosis monitoring is the use of ultrasonography, which provides true 3D information in one scan and does not emit ionizing radiation. Measuring the severity of scoliosis on ultrasonographs requires identifying lamina pairs on the most tilted vertebrae, which is difficult and time-consuming. To expedite and automate measurement steps, this paper detailed an automatic convolutional neural network-based algorithm for identifying the laminae on 3D ultrasonographs. The predicted laminae were manually paired to measure the lateral spinal curvature on the coronal view, called the Cobb angle. In total, 130 spinal ultrasonographs of adolescents with idiopathic scoliosis recruited from a scoliosis clinic were selected, with 70 for training and 60 for testing. Data augmentation increased the effective training set size to 140 ultrasonographs. Semi-automatic Cobb measurements were compared to manual measurements on the same ultrasonographs. The semi-automatic measurements demonstrated good inter-method reliability (ICC_3,1 = 0.87) and performed better on thoracic (ICC_3,1 = 0.91) than lumbar curves (ICC_3,1 = 0.81). The mean absolute difference and standard deviation between semi-automatic and manual was 3.6° ± 3.0°. In conclusion, the semi-automatic method to measure the Cobb angle on ultrasonographs is feasible and accurate. This is the first algorithm that automates steps of Cobb angle measurement on ultrasonographs.

     

A Patch-Based Deep Learning Approach for Detecting Rib Fractures on Frontal Radiographs in Young Children

Journal of Digital Imaging - Chest radiography is the modality of choice for the identification of rib fractures in young children and there is value for the development of computer-aided rib...     

Effect of Dataset Size and Medical Image Modality on Convolutional Neural Network Model Performance for Automated Segmentation: A CT and MR Renal Tumor Imaging Study

The aim of this study is to investigate the use of an exponential-plateau model to determine the required training dataset size that yields the maximum medical image segmentation performance. CT and MR images of patients with renal tumors acquired between 1997 and 2017 were retrospectively collected from our nephrectomy registry. Modality-based datasets of 50, 100, 150, 200, 250, and 300 images were assembled to train models with an 80–20 training-validation split evaluated against 50 randomly held out test set images. A third experiment using the KiTS21 dataset was also used to explore the effects of different model architectures. Exponential-plateau models were used to establish the relationship of dataset size to model generalizability performance. For segmenting non-neoplastic kidney regions on CT and MR imaging, our model yielded test Dice score plateaus of \(0.93\pm 0.02\) and \(0.92\pm 0.04\) with the number of training-validation images needed to reach the plateaus of 54 and 122, respectively. For segmenting CT and MR tumor regions, we modeled a test Dice score plateau of \(0.85\pm 0.20\) and \(0.76\pm 0.27\), with 125 and 389 training-validation images needed to reach the plateaus. For the KiTS21 dataset, the best Dice score plateaus for nn-UNet 2D and 3D architectures were \(0.67\pm 0.29\) and \(0.84\pm 0.18\) with number to reach performance plateau of 177 and 440. Our research validates that differing imaging modalities, target structures, and model architectures all affect the amount of training images required to reach a performance plateau. The modeling approach we developed will help future researchers determine for their experiments when additional training-validation images will likely not further improve model performance.

     

Recognition of Digital Dental X-ray Images Using a Convolutional Neural Network

Journal of Digital Imaging - Digital dental X-ray images are an important basis for diagnosing dental diseases, especially endodontic and periodontal diseases. Conventional diagnostic methods...     

基于卷积神经网络的复合菌落智能分类识别

为满足复合菌落智能形态分类的需求,构建菌落分类卷积神经网络。通过水平集演化分割,获取培养皿内部所有的连通域;通过极限腐蚀,判别种子点数目大于1的连通域,即为粘连连通域;获取粘连连通域的凸闭包,检测凹点并连接对应凹点,对该连通域进行分割。归一化获取的600张单个菌落样本,通过旋转翻转并叠加信噪比不超过5%的随机噪声,将数据扩增至30 000例。以其中70%样本数据作为菌落分类卷积神经网络的训练集,对网络模型进行10折交叉验证,再以30%样本数据进行测试,4种菌落的加权平均准确率达到87.50%;其中斑点状光滑菌落分类准确率为86.40%,类圆波状菌落分类准确率为87.21%,椭圆形菌落分类准确率为88.11%,不规则其他菌落分类准确率为87.25%。最后采用通用计算设备架构(CUDA),对各个算法模块进行并行优化加速,算法运行时间最优提升至原耗时的1/10,在运行速度和便利性方面远远超过传统菌落分类方法。所设计的方法可以有效完成复合菌落智能分类识别任务,并具有良好的扩展性和自学习功能,对基于图像的生化样本智能分析具有一定的借鉴价值。     

Intelligent Evaluation of Global Spinal Alignment by a Decentralized Convolutional Neural Network

Degenerative changes of the spine can cause spinal misalignment, with part of the spine arching beyond normal limits or moving in an incorrect direction, potentially resulting in back pain and significantly limiting a person’s mobility. The most important parameters related to spinal misalignment include pelvic incidence, pelvic tilt, lumbar lordosis, thoracic kyphosis, and cervical lordosis. As a general rule, alignment of the spine for diagnosis and surgical treatment is estimated based on geometrical parameters measured manually by experienced doctors. However, these measurements consume the time and effort of experts to perform repetitive tasks that could be automated, especially with the powerful support of current artificial intelligence techniques. This paper focuses on creation of a decentralized convolutional neural network to precisely measure 12 spinal alignment parameters. Specifically, this method is based on detecting regions of interest with its dimensions that decrease by three orders of magnitude to focus on the necessary region to provide the output as key points. Using these key points, parameters representing spinal alignment are calculated. The quality of the method’s performance, which is the consistency of the measurement results with manual measurement, is validated by 30 test cases and shows 10 of 12 parameters with a correlation coefficient?>?0.8, with pelvic tilt having the smallest absolute deviation of 1.156°.

     

基于CNN算法的稳态体感诱发电位的特征识别

脑-机接口研究可为瘫痪病人的康复带来一种新的治疗方法。已有研究表明对手指或者正中神经施加一定频率的体感刺激,会引发相同频率且具有空间特异性的稳态体感诱发电位。为优化基于稳态体感诱发电位的脑-机接口的性能,通过快速傅里叶变换寻找12个健康被试的个人左手特定共振频率,采用事件相关谱扰动进行时频分析,检测其稳态体感诱发电位信号。基于共振频率对实验诱发的脑电信号进行1 Hz带通滤波,获得特定频带的数据,采用卷积神经网络(CNN)学习算法对其进行分类,并与采用共空间模式和支持向量机的特征提取及特征分类的方法(CSP+SVM)进行比较。所有被试的结果显示:基于共振频率滤波方法,采用CNN学习算法获得的离线分类准确率均高于85%,并且CNN学习算法的分类准确率显著性优于CSP+SVM的分类准确率(91.8%±5.9% vs 77.4%±8.5%,P<0.05)。因此,在基于稳态体感诱发电位的脑机接口的特征识别中,CNN学习算法相比传统使用的机器学习分类算法(如共空间模式+支持向量机)能够显著提升分类准确率,提高脑机接口的整体性能。     

基于卷积神经网络的P300事件相关电位分类识别

针对脑机接口系统中P300电位识别正确率不高的问题,提出一种基于改进卷积神经网络的P300事件相关电位分类识别方法。通过将传统卷积神经网络中第二个串行连接的卷积层改为3个并行连接的卷积层,可加大网络宽度,提升网络对P300信号特征提取的能力;将提取的特征经全互连层组合后,采用sigmoid函数构建P300事件相关电位分类器。针对脑机接口竞赛数据中靶刺激与非靶刺激数据量不平衡的问题,采用过抽样方式,对含有P300事件相关电位的脑电数据做部分平均来增加数据量,其训练集和测试集样本量分别为25 500和18 000。采用Adam优化方法,有监督地训练这种改进的卷积神经网络。结果表明,相比传统的卷积神经网络,该方法在实验次数大于11次时,字符识别正确率均高于95%,这对于脑机接口的应用具有重要的意义。     

健康儿童同足异趾甲襞微循环对比分析

目的：探讨儿童同足异趾甲襞微循环检测时质量控制。方法：用WX－753B微循环显微镜和图像处理系统，观察52名6岁健康儿童的左足五趾甲襞微循环十九项指标，并进行五趾甲襞微循环之间的两两比较。结果：同足异趾间的甲襞微循环形态学指标存在显著性差异，而流态和襻周状态指标各趾间无显著性差异。拇趾甲襞管襻清晰度较好、数目最多、襻长最长、直径最大、而小趾则相反。结论：同足异趾间甲襞微循环存在一定差异，因此，观测儿童足甲襞微循环时，特别是动态监测病例，应选择同一足趾甲襞观测，观测部位则以拇趾甲襞最佳。     

Application of Super-Resolution Convolutional Neural Network for Enhancing Image Resolution in Chest CT

In this study, the super-resolution convolutional neural network (SRCNN) scheme, which is the emerging deep-learning-based super-resolution method for enhancing image resolution in chest CT images, was applied and evaluated using the post-processing approach. For evaluation, 89 chest CT cases were sampled from The Cancer Imaging Archive. The 89 CT cases were divided randomly into 45 training cases and 44 external test cases. The SRCNN was trained using the training dataset. With the trained SRCNN, a high-resolution image was reconstructed from a low-resolution image, which was down-sampled from an original test image. For quantitative evaluation, two image quality metrics were measured and compared to those of the conventional linear interpolation methods. The image restoration quality of the SRCNN scheme was significantly higher than that of the linear interpolation methods (p < 0.001 or p < 0.05). The high-resolution image reconstructed by the SRCNN scheme was highly restored and comparable to the original reference image, in particular, for a ×2 magnification. These results indicate that the SRCNN scheme significantly outperforms the linear interpolation methods for enhancing image resolution in chest CT images. The results also suggest that SRCNN may become a potential solution for generating high-resolution CT images from standard CT images.     

结合表型信息的阿尔兹海默症图卷积神经网络分类方法研究

阿尔兹海默症(AD)的早期检测与发现具有重要的临床和社会意义.由于AD患者的功能性脑网络拓扑性质存在异常变化,并且不同表型类型人群中阿尔兹海默症的患病率也存在着较大差异,因此将脑网络特征和表型信息结合构建训练特征,用于阿尔兹海默症不同阶段的分类.同时,图卷积神经网络(GCN)分类方法被证明是目前对图数据学习任务的最佳选...     

Neural Processing of Threat Cues in Young Children With Attention-Deficit/Hyperactivity Symptoms

A growing literature indicates that attention deficit/hyperactivity disorder (ADHD) involves difficulty processing threat-related emotion faces. This deficit is especially important to understand in young children, as threat emotion processing is related to the development of social skills and related behavioral regulation. Therefore, the current study aimed to better understand the neural basis of this processing in young children with ADHD symptoms. Forty-seven children between 4 and 7 years of age were included in the analysis, 28 typical developing and 19 with clinically significant levels of ADHD hyperactive/impulsive symptoms. Participants completed a passive affective face-viewing task. Event-related potentials were assessed for each emotion, and parental report of child behavior and emotion regulation abilities was assessed. Children with ADHD symptoms showed altered N170 modulation in response to specific emotion faces, such that the N170 was less negative in response to fearful compared to neutral faces, whereas typically developing children showed the opposite pattern. Groups did not differ in reactivity to anger or non-threat-related emotion faces. The N170 difference in fearful compared to neutral faces correlated with reported behavior, such that less fear reactivity predicted fewer prosocial behaviors. Abnormalities in the underlying neural systems for fear processing in young children with ADHD symptoms may play an important role in social and behavioral deficits within this population.     

Linear and Angular Measurements of the Foot of Modern Humans: A Test of Morton'S Foot Types

In his classic research, Morton established two functionally different configurations of the bipedal and non‐bipedal primate foot: one optimized for stability, with a stiff longitudinal arch and adducted first metatarsal, and the other for compliance. Modern human feet were seen as conforming to the bipedal norm and variation from it as pathology, even though clinical evidence has been clear that variation from the norm of a stiff longitudinal arch or adducted first metatarsal exists. This study aims to document the variation in linear and angular measurements of the foot, using weight‐bearing radiographs of 50 randomly selected people (25 men) from an urban US Level 1 trauma center. The radiographs were obtained to “rule‐out” a foot fracture after trauma or as comparison films for a contralateral foot injury. Measurements were made using Osirix and correlations among the angular and length measurements were determined using Stata with P < 0.05 and Bonferroni correction for multiple comparisons. We found that foot length was not correlated with angular measurements, except for the angle between the first and fifth metatarsal. People with lower medial longitudinal arches also had more abducted first metatarsals and more vertical posterior facets, in accordance with Morton's foot types. Whether or not this variation in modern humans is linked to functionally important consequences remains to be determined in future research. With the new evidence of a more variable foot structure in fossil hominins, understanding the relationship between foot morphology and function becomes more urgent. Anat Rec, 296:1526–1533, 2013. © 2013 Wiley Periodicals, Inc.     

Automating Scoliosis Measurements in Radiographic Studies with Machine Learning: Comparing Artificial Intelligence and Clinical Reports

Journal of Digital Imaging - Scoliosis is a condition of abnormal lateral spinal curvature affecting an estimated 2 to 3% of the US population, or seven million people. The Cobb angle is the...     

Breast Tumor Classification in Ultrasound Images by Fusion of Deep Convolutional Neural Network and Shallow LBP Feature

Journal of Digital Imaging - Breast cancer is one of the most dangerous and common cancers in women which leads to a major research topic in medical science. To assist physicians in pre-screening...     

基于3D深度残差全卷积网络的头颈CT放疗危及器官自动勾画

头颈部肿瘤放射治疗危及器官的准确勾画是放疗计划的关键步骤,然而头颈部放疗危及器官的精确分割挑战性很大,目前临床医生手动勾画危及器官非常繁琐、耗时且缺乏一致性。提出基于3D深度残差全卷积网络的头颈部肿瘤放疗危及器官自动分割方法,通过改进的V-Net网络分割模型,有效地结合危及器官CT影像的深层特征和浅层特征,同时根据特别设计的端到端监督学习确定危及器官分割模型参数。为了解决小器官类分布极不平衡问题,提出利用器官位置先验约束采样区域与随机采样相结合的训练样本选择策略,同时采用Dice损失函数对网络进行训练。该策略不仅可加速训练过程,提升分割性能,而且可保证小器官的分割准确率。该方法在2015年MICCAI头颈自动分割挑战赛数据集PDDCA上验证,各器官分割的Dice系数平均值分别为:颌下骨0.945、左腮腺0.884、右腮腺0.882、脑干0.863、左颌下腺0.825、右颌下腺0.842、左视神经0.807、右视神经0.847、视交叉0.583。大多数器官的95% Hausdorff距离小于3 mm,所有器官的勾画平均距离均小于1.2 mm。实验结果表明,该方法在除脑干以外的危及器官分割中性能比其他对比方法更优。