Mammographic mass detection by adaptive thresholding and region growing

We present an efficient method to detect mass lesions on digitized mammograms, which consists of breast region extraction, region partitioning, automatic seed selection, segmentation by region growing, feature extraction, and neural network classification. The method partitions the breast region into a fat region, a fatty and glandular region, and a dense region, so that different threshold values can be applied to each partitioned region during processes of the seed selection and segmentation. The mammographic masses are classified by using four features representing shape, density, and margin of the segmented regions. The method detects subtle mass lesions with various contrast ranges and can facilitate a procedure of mass detection in computer‐aided diagnosis systems. © 2001 John Wiley & Sons, Inc. Int J Imaging Syst Technol, 11, 340–346, 2000     

深度学习在材料显微图像分析中的应用与挑战

材料的组织结构主要受成分和制备加工工艺的影响,是决定材料性能的关键因素,在材料研发的全周期内具有重要作用。材料组织结构以非结构化图像数据的形式呈现,利用人工经验性的手段进行分析和信息抽取,遗漏了大量的材料学信息和隐含知识。深度学习技术的发展和应用,为材料显微图像中信息的精准、快速、自动获取提供了重要的研究手段。本文从图像处理、图像分析和图像理解3个方面概述了材料显微图像处理与信息挖掘的主要研究内容和关键技术,详细介绍了深度学习在图像分析中的图像识别、图像分割和图像生成3个任务中的研究进展,讨论了深度学习在材料显微图像分析和信息挖掘中的发展方向和挑战。     

SS-OCTA对黑色素瘤皮肤结构和血管的成像实验

扫频源光学相干层析血管成像(SS-OCTA)是一种基于分频幅去相关血管造影法(SSADA)的新型血管成像技术,在肿瘤等疾病的早期诊断方面拥有较大前景。本文在5.12 mm×5.12 mm成像视场、标准图像最大信噪比34.3dB的SS-OCTA成像平台,对黑色素瘤C57BL6小鼠进行皮肤结构和血管成像采集。结果表明在皮肤科疾病的早期诊断方面,利用SS-OCTA系统进行血管成像优于结构成像。     

Microstructural characterisation of 3D printed and injection-moulded glass fibre-reinforced polypropylene by image analysis,simulation and experimental methods

The mechanical properties of fibre-reinforced thermoplastics and their dependencies on the manufacturing process, fibre properties, fibre concentration and strain rate have been researched intensively for years in order to predict their macroscopic behaviour by numerical simulations as precisely as possible. Including the microstructure in both real and virtual experiments has improved prediction precision for injection-moulded glass fibre-reinforced thermoplastics significantly. In this work, we apply three established methods for characterisation and modelling to an injection-moulded and to a 3D printed material. The geometric properties of the fibre component as fibre orientation, fibre length and fibre diameter distributions are identified by analysing reconstructed tomographic images. For comparing the fibre lengths, a recently suggested new method is applied. Based on segmentations of the tomographic images, we calculate the elastic stiffness of both composites numerically on the microscale. Finally, the mechanical behaviour of both materials is experimentally characterised by micro tensile tests. The simulation results agree well with the measured stiffness in case of the injection-moulded material. However, for the 3D printed material, measurement and simulation differ strongly. The prediction from the simulation agrees with the values expected from the image analytic findings on the microstructure. Therefore, the differences in the measured behaviour have to be contributed to the matrix material. This proves demand for further research for 3D printed materials for predictable prototypes, preproduction series and possible serial application.