胃镜下早期胃癌计算机辅助分析研究综述

胃癌是全世界癌症死亡的第三大主要原因,胃癌的早期检测会对胃癌患者的后期治疗起到至关重要的作用。随着人工智能的发展,可以利用计算机视觉领域的机器学习模型辅助检测早期胃癌,有研究发现一些计算机辅助诊断模型的筛查率接近甚至高于医生。利用计算机辅助诊断可以及早发现胃癌以减少胃癌患者的后期治疗成本。报告了基于机器学习在胃镜下早期胃癌辅助诊断的研究现状,介绍了胃镜下早期胃癌的临床诊断方式,并基于此提出了计算机辅助诊断该疾病的技术路线,分析了不同诊断技术路线的研究特点,为计算机辅助诊断早期胃癌提供不同的切入点。总结了用于早期胃癌检测的机器学习、深度学习、目标检测模型,讨论了其应用于计算机辅助诊断的问题及挑战。     

基于双正交小波的乳腺图像增强

乳腺癌的早期诊断是降低死亡率的主要手段。该文利用双正交小波变换,将乳腺图像进行分解和非线性增强,然后重构图像,使得原图像中可见性差的病变特征(如肿块)更清楚地显现在重构图像中。实验结果说明该方法的有效性。     

Markers of breast and gynecological malignancies: The clinical approach of proteomics-based studies

Breast, ovarian, endometrial and cervical cancers are diseases with a relatively high lethal outcome, despite recent development of surgical, chemotherapeutic, radiotherapeutic and anti-hormonal treatments. The mortality of women with breast cancer is at the level of 20%, while for ovarian cancer it is up to 75%. For cervical cancer it is 38% and for endometrial cancer 17%. To enhance survival of patients, better diagnostics of these cancers in early and non-invasive stages is essential. Proteomics allows a comprehensive analysis of hundreds of proteins in clinical samples. Studies of tissue samples and body fluids, e.g., blood and plasma, have shown their usefulness in discovery of markers of cancers at early stages. Searches for markers of breast and ovarian cancers have been performed with promising results. Studies of endometrial and cervical cancer markers have been less extensive. This review describes recent reports of markers for early detection of breast, ovarian, endometrial and cervical cancers. A clinical evaluation of the markers discovered with the use of proteomics techniques is discussed.     

Revealing determinant factors for early breast cancer recurrence by decision tree

Early breast cancer recurrence is indicative of poor response to adjuvant therapy and poses threats to patients’ lives. Most existing prediction models for breast cancer recurrence are regression-based models and difficult to interpret. We apply a Decision Tree algorithm to the clinical information of a cohort of non-metastatic invasive breast cancer patients, to establish a classifier that categorizes patients based on whether they develop early recurrence and on similarities of their clinical and pathological diagnoses. The classifier predicts for whether a patient developed early disease recurrence; and is estimated to be about 70% accurate. For an independent validation cohort of 65 patients, the classifier predicts correctly for 55 patients. The classifier also groups patients based on intrinsic properties of their diseases; and for each subgroup lists the disease characteristics in a hierarchal order, according to their relevance to early relapse. Overall, it identifies pathological nodal stage, percentage of intra-tumor stroma and components of TGFβ-Smad signaling pathway as highly relevant factors for early breast cancer recurrence. Since most of the disease characteristics used by this classifier are results of standardized tests, routinely collected during breast cancer diagnosis, the classifier can easily be adopted in various research and clinical settings.     

Approaches for automated detection and classification of masses in mammograms

Breast cancer continues to be a significant public health problem in the world. Early detection is the key for improving breast cancer prognosis. Mammography has been one of the most reliable methods for early detection of breast carcinomas. However, it is difficult for radiologists to provide both accurate and uniform evaluation for the enormous mammograms generated in widespread screening. The estimated sensitivity of radiologists in breast cancer screening is only about 75%, but the performance would be improved if they were prompted with the possible locations of abnormalities. Breast cancer CAD systems can provide such help and they are important and necessary for breast cancer control. Microcalcifications and masses are the two most important indicators of malignancy, and their automated detection is very valuable for early breast cancer diagnosis. Since masses are often indistinguishable from the surrounding parenchymal, automated mass detection and classification is even more challenging. This paper discusses the methods for mass detection and classification, and compares their advantages and drawbacks.     

基于NRU网络的肺结节检测方法

肺癌的早期发现和早期诊断是提高肺癌患者生存率的关键。由于肺癌早期结节很小,目前已有的肺结节检测系统在检测这些结节时很容易漏诊。准确检测早期肺癌结节对于提高肺癌治愈率至关重要,为了降低检测系统对早期结节的漏诊率,需要优化候选结节的提取步骤。在U-Net网络中引入残差网络的捷径,有效解决了传统U-Net网络由于缺乏深度而导致结果较差的问题。在此改进的基础上提出了一种U型噪声残差网络NRU（Noisy Residual U-Net）,通过利用跳跃层连接的特性和向卷积层添加噪声来增强神经网络对小结节的灵敏度。使用Lung Nodule Analysis 2016和阿里巴巴天池肺癌检测竞赛数据集训练神经网络。U-Net和NRU之间的比较实验表明,该算法对直径为3~5 mm（97.1％）的小结节的灵敏度大于U-Net值（90.5％）。     

Medical imaging technique using curvelet transform and machine learning for the automated diagnosis of breast cancer from thermal image

Pattern Analysis and Applications - Thermography is a useful imaging tool using infrared for the early diagnosis of breast cancer. Screening cancer aims to outstrip prognosis by seeing the...     

基于BP神经网络的肺结节自动检测算法研究

长期监测发现近年来我国肺癌发病率上升至原先的4倍,气象等专家经过研究发现,灰霾是致肺癌高发的一个根本原因,特别是在城市。肺癌的早期诊断十分重要。利用计算机图像处理技术检测肺癌早期标志物——肺结节,可以提高肺癌诊断的准确率。据此设计了一个CAD系统,尝试通过四个步骤实现肺结节的检测:肺实质分割、感兴趣区(ROI)的提取、特征的提取与计算、肺结节检测。     

基于磁感应QCM在早期肺癌检测上的研究

利用压电石英晶体的振动频率对其电极表面负荷敏感的特点和现场可编程（逻辑）门阵列（FP-GA）与单片机的控制功能,同时,利用纳米免疫磁微珠具有增大负载响应和磁分离的能力,结合特异性生物化学反应,设计了一种用于早期肺癌诊断的纳米生物电子微系统。通过该系统对蒸馏水和CPA—Y2肺癌细胞溶液的测量结果表明：该系统具有较好的线性变化规律,能准确反映被测生物物质的质量或含量,对实现肺癌的早期诊断和筛查具有重要的参考价值。     

基于多分类算法混合比较的乳腺癌预测

乳腺癌具备易于复发性和高死亡率等特点, 已成为女性癌症患者死亡的重要原因. 乳腺癌的早期诊断可增加癌 症治愈的可能性, 因此, 提高早期诊断的准确性尤为重要. 传统的早期诊断主要依靠人类经验, 通过分析临床或检查数 据来判断乳腺癌, 无法保证足够的准确性. 许多研究人员提出了各种机器学习方法, 以提高预测的准确性和效率. 但现 有的算法计算复杂性很高, 并且难以从多种算法中直接确定最适合的算法. 本文尝试了10种流行的分类算法, 比较了它 们之间的差异, 并应用了量子支持向量机来加速计算过程. 数值实验显示支持向量机和人工神经网络的预测效果最佳, 表明了多种分类算法混合比较的有效性     

应用于癌症基因表达数据的OMB双向聚类算法

癌症基因表达数据的聚类分析可以为癌症的早期诊断和精确的癌症亚型分型提供依据。针对癌症基因表达数据的特点,提出一种称为OMB（Override Matrix Bicluster）的双向聚类算法。OMB算法分别在基因表达数据矩阵的行和列上搜索低于阈值的行和列,用删除添加算法产生一个子矩阵;构建与基因表达矩阵大小相同的覆盖矩阵,标识矩阵中上一次迭代产生的子矩阵的位置;在标识出来的矩阵中,重复贪婪迭代搜索找到K个聚类结果。Matlab实验结果表明OMB算法对具有重叠结构的癌症基因表达数据具有更好的聚类效果。     

Computer-aided detection and classification of microcalcifications in mammograms: a survey

Breast cancer continues to be a significant public health problem in the world. Approximately, 182,000 new cases of breast cancer are diagnosed and 46,000 women die of breast cancer each year in the United States. Even more disturbing is the fact that one out of eight women in US will develop breast cancer at some point during her lifetime. Primary prevention seems impossible since the causes of this disease still remain unknown. Early detection is the key to improving breast cancer prognosis. Mammography is one of the reliable methods for early detection of breast carcinomas. There are some limitations of human observers, and it is difficult for radiologists to provide both accurate and uniform evaluation for the enormous number of mammograms generated in widespread screening. The presence of microcalcification clusters (MCCs) is an important sign for the detection of early breast carcinoma. An early sign of 30–50% of breast cancer detected mammographically is the appearance of clusters of fine, granular microcalcification, and 60–80% of breast carcinomas reveal MCCs upon histological examinations. The high correlation between the appearance of the microcalcification clusters and the diseases show that the CAD (computer aided diagnosis) systems for automated detection/classification of MCCs will be very useful and helpful for breast cancer control. In this survey paper, we summarize and compare the methods used in various stages of the computer-aided detection systems (CAD). In particular, the enhancement and segmentation algorithms, mammographic features, classifiers and their performances are studied and compared. Remaining challenges and future research directions are also discussed.     

协同肿瘤研究临床试验信息系统SOA 建模

开放标准的肿瘤信息系统在不断发展,以此应对现代肿瘤临床试验带来的挑战。本文提出面向服务的软件范例,用于衍生临床试验信息管理系统,支持多机构协同肿瘤研究。提出的方案将临床试验(元)模型和基于WSRF(Web Services Resource Framework)的面向服务体系结构(Service-Oriented Architecture,SOA)相结合,并且应用早期临床试验进行评估。尽管主要目的是针对肿瘤研究,但也适用于具有相似信息模型的其它领域。     

Automatic classification with concatenation of deep and handcrafted features of histological images for breast carcinoma diagnosis

Multimedia Tools and Applications - The second leading cause of death from cancer among women is breast cancer. In order to prevent avoidable deaths, early detection is extremely necessary....     

Pulmonary nodules recognition based on parallel cross-convolution

Multimedia Tools and Applications - Lung cancer is the highest incidence rate and mortality rate in human beings. Pulmonary nodules are the early manifestation of lung cancer. The accurate...     

Application of proteomics in the study of rodent models of cancer

The molecular and cellular mechanisms underlying the multistage processes of cancer progression and metastasis are complex and strictly depend on the interplay between tumor cells and surrounding tissues. Identification of protein aberrations in cancer pathophysiology requires a physiologically relevant experimental model. The mouse offers such a model to identify protein changes associated with tumor initiation and progression, metastasis development, tumor/microenvironment interplay, and treatment responses. Furthermore, the mouse model offers the ability to collect samples at any stage in tumor development from highly matched disease cases and controls with identical environmental and genetic backgrounds, thus providing an excellent method for biomarker discovery. Xenograft and genetically engineered mouse models have been widely used to identify proteomic patterns in tumor tissues and plasma samples associated with different stages of human cancer, including early cancer detection and development of metastasis. Here, we review proteomic strategies to identify proteins involved in key cancer processes within such animal models as well as biomarkers for diagnosis, prognosis, and monitoring of cancer progression and treatment response. Central to such studies is the ability to ensure at an early stage that the identified proteins are of clinical relevance by examining relevant specimens from larger cohorts of cancer patients.     

Mammogram pectoral muscle removal and classification using histo-sigmoid based ROI clustering and SDNN

Multimedia Tools and Applications - Mammograms are the images used by radiologists to diagnose breast cancer. Breast cancer is one of the most common cancers in women. The early detection of breast...     

A survey on cancer prediction and detection with data analysis

Innovations in Systems and Software Engineering - World Health Organization reports cancer as a leading cause worldwide in mortality and morbidity. Accurate and early cancer risk assessment in...     

Intelligent breast cancer recognition using particle swarm optimization and support vector machines

Breast cancer is the most common cancer among women, except for skin cancer, but early detection of breast cancer improves the chances of survivability. Data mining is widely used for this purpose. As technology develops, large number of breast tumour features are being collected. Using all these features for cancer recognition is expensive and time-consuming. Feature extraction is necessary for increasing the classification accuracy. The goal of this work is to recognise breast cancer using extracted features. To reach this goal, a combination of clustering and classification is used. Particle swarm optimization is used to recognise tumour patterns. The membership degree of each tumour to the patterns is calculated and considered as a new feature. Support vector machine is then employed to classify tumours. Finally this method is analysed in terms of its accuracy, specificity, sensitivity and CPU time consuming using Wisconsin Diagnostic Breast Cancer data set.     

Breast cancer detection in mammography using spatial diversity,geostatistics, and concave geometry

Multimedia Tools and Applications - Breast cancer is a global health problem which mainly affects the female population. It is known that early detection increases the chances of effective...