公开(公告)号：WO2023078676A1

公开(公告)日：2023-05-11

申请号：PCT/EP2022/079005

申请日：2022-10-19

Applicant: AGFA HEALTHCARE NV ,  VRVIS ZENTRUM FÜR VIRTUAL REALITY UND VISUALISIERUNG FORSCHUNGS-GMBH

Inventor： WIMMER, Maria ,  MAJOR, David ,  LENIS, Dimitrios ,  BERG, Astrid ,  BUEHLER, Katja

IPC: G16H50/20 ,  G16H30/40

Abstract: A computer-implemented method for developing a mammography deep learning model wherein a set of task-specific mammography deep learning models is developed, each trained for performing a different task on a mammography dataset and each generating one or more feature vectors and wherein the task-specific models are combined to a patient model to obtain a patient prediction by fusing said feature vectors.

公开(公告)号：WO2022175148A1

公开(公告)日：2022-08-25

申请号：PCT/EP2022/053187

申请日：2022-02-10

Applicant: AGFA NV

Inventor： EXELMANS, Walter

IPC: A61B6/00 ,  A61B5/00

Abstract: A first visual image of a region of interest is generated by means of an optical camera at the time when a first radiation image is recorded, a second visual image is generated by means of an optical camera of said same region of interest, first and second visual images are compared, and a second radiation image is recorded when said first and second visual images are substantially identical.

公开(公告)号：WO2022073856A1

公开(公告)日：2022-04-14

申请号：PCT/EP2021/077050

申请日：2021-10-01

Applicant: AGFA NV ,  VRVIS ZENTRUM FÜR VIRTUAL REALITY UND VISUALISIERUNG FORSCHUNGS-GMBH

Inventor： BERG, Astrid ,  VANDERSMISSEN, Eva ,  BUEHLER, Katja

IPC: G06T5/40 ,  G06T7/00 ,  G06T7/10 ,  G06N3/08

Abstract: A method of augmenting the number of labeled images for training a neural network comprising the steps of - Starting from a dataset of labeled images with corresponding segmentation masks and a dataset of unlabeled images, gathering for a given image i in a data set of labeled images a number of images with similar metadata in said dataset of unlabeled images so as to form data sub-set Sim i, - Training a multiclass segmentation neural network on said labeled images thereby generating segmentation masks for the images in subset Sim i, - On the basis of these segmentation masks judging similarity between images of Sim i and image i and finding the most similar image(s) in Sim i by computing and comparing histograms of segmentation masks of image i and images in Sim i - Transferring the histogram of the most similar images in Sim i to given image i.

公开(公告)号：WO2021009024A1

公开(公告)日：2021-01-21

申请号：PCT/EP2020/069504

申请日：2020-07-10

Applicant: AGFA HEALTHCARE NV

Inventor： HUSSAIN, Mohannad ,  WALLACE, William

IPC: G16H40/63 ,  G16H40/67

Abstract: Systems and methods for integrating a plurality of healthcare software applications. A user initiates one or more software applications required to perform different parts of a healthcare task. The software applications include a local application installed on the user's device and a web-based application accessed through a browser. A host application is launched in order to facilitate local two-way communication between the local application and the web-based application. The host application generates a local communication server on the user device through which the local application and web application communicate. This allows a user to easily and efficiently perform multiple parts of the healthcare tasks. This also allows different healthcare software applications to be integrated without requiring any plug-ins or other integration components to be installed on the user's device.

公开(公告)号：WO2019011692A9

公开(公告)日：2019-01-17

申请号：PCT/EP2018/067786

申请日：2018-07-02

Applicant: AGFA NV

Inventor： NEBOSIS, Rainer ,  LEBLANS, Paul

IPC: A61B6/00 ,  G06T7/00 ,  G06T11/00 ,  H04N5/335 ,  G21K1/02 ,  G21K1/10

Abstract: A phase contrast imaging (PCI) method in which instead of using an analyzer grid detector pixels are grouped and only a part of the total pixels are used to calculate a phase contrast image. In a second, third... step the pixels which were not used in the previous recalculation are used additionally to recalculate a second, third.... phase contrast image. Finally the different phase contrast images are fused to result in a full image.

公开(公告)号：WO2018108569A1

公开(公告)日：2018-06-21

申请号：PCT/EP2017/081111

申请日：2017-12-01

Applicant: AGFA HEALTHCARE

Inventor： OUJI, Asma

IPC: G06T7/174 ,  G06T7/11

Abstract: On the basis of user input a set of contour points of the 3D object is detected in a number of 2D slice images representing the 3D object. Next a 2D object is segmented in each of the slice images by means of the set of contour points so as to obtain segmentation masks. Finally by means of interpolation between computed segmentation masks, the 3D object is segmented.

公开(公告)号：WO2018029237A1

公开(公告)日：2018-02-15

申请号：PCT/EP2017/070150

申请日：2017-08-09

Applicant: AGFA HEALTHCARE ,  VRVIS ZENTRUM FÜR VIRTUAL REALITY UND VISUALISIERUNG

Inventor： NOVIKOV, Alexey ,  MAJOR, David ,  WIMMER, Maria ,  BUEHLER, Katja

IPC: G06T7/187

Abstract: Method and system to fully-automatically analyze a medical image represented by a digital image representation. A method is disclosed to automatically detect systemic arteries in arbitrary field-of-view computed tomography angiography (CTA).

Abstract translation: 用于全自动分析由数字图像表示表示的医学图像的方法和系统。 公开了一种在任意视野计算机断层摄影血管造影术（CTA）中自动检测全身动脉的方法。

公开(公告)号：WO2017050733A1

公开(公告)日：2017-03-30

申请号：PCT/EP2016/072249

申请日：2016-09-20

Applicant: AGFA HEALTHCARE

Inventor： CRESENS, Marc ,  VAN GOUBERGEN, Herman

IPC: G06T5/00

CPC classification number: G06T5/005 ,  G06T2207/10116 ,  G06T2207/20012

Abstract: An adaptive reconstruction kernel taking up to four main and diagonal, defect-free sub-kernel directions into account is composed for each defective image pixel. The defective pixels impacted image is real-time corrected by statistical filtering or by a weighed directional convolution of kernel-associated replacement values, calculated by means of an advanced multi-parabolic reconstruction algorithm, for each contributing sub-kernel direction based on 5 x 5 pixels neighborhood image data readily accessible via a predetermined AMP kernels image-offsets structure.

Abstract translation: 针对每个缺陷图像像素构成了考虑到四个主要和对角线，无缺陷的子核方向的自适应重建内核。 通过统计过滤或通过基于5×5的每个贡献子核方向的借助于高级多抛物线重建算法计算的核相关替换值的加权方向卷积来实时校正受影响图像的缺陷像素 通过预定的AMP内核图像偏移结构容易地访问像素邻域图像数据。

公开(公告)号：WO2017005838A1

公开(公告)日：2017-01-12

申请号：PCT/EP2016/066088

申请日：2016-07-07

Applicant: AGFA HEALTHCARE ,  BOONE, Marc

Inventor： BOONE, Marc ,  DHAENENS, Frans

IPC: G06T7/00

CPC classification number: G06T7/0012 ,  A61B5/0066 ,  A61B5/444 ,  A61B2576/00 ,  G06T7/50 ,  G06T2207/10101 ,  G06T2207/30004

Abstract: A depth profile of a region of interest in a 3D image of a biological tissue obtained through full field time domain high definition optical coherence tomography imaging is subjected to curve fitting. Parameters linked to optical anisotropy characteristics of this tissue are deduced from fitted curves and are compared to set values of these parameters, the set values being indicative of normal, precancerous or cancerous tissue characteristics.

Abstract translation: 通过全场时域高清晰度光学相干断层成像获得的生物组织的3D图像中的感兴趣区域的深度轮廓进行曲线拟合。 从拟合曲线推导出与该组织的光学各向异性特征相关的参数，并与这些参数的设定值进行比较，该设定值表示正常，癌前或癌组织特征。

公开(公告)号：WO2016091999A1

公开(公告)日：2016-06-16

申请号：PCT/EP2015/079213

申请日：2015-12-10

Applicant: AGFA HEALTHCARE

Inventor： CRESENS, Marc ,  VAN GOUBERGEN, Herman

IPC: G06T5/00

CPC classification number: G06T5/005 ,  G06T2207/10116 ,  G06T2207/20012 ,  G06T2207/20028 ,  G06T2207/20032

Abstract: A method for reducing image disturbances caused by reconstructed defective pixel clusters located in signal-gradient affected diagnostic image regions. An individually adapted central symmetrical pair reconstruction (CSP) kernel is composed for a defective image pixel based on a kernel-pair candidate order encoded in a model thereby using the pixel's validity state. The image impacted by defective pixels is corrected in real-time by statistical filtering or spatial convolution of the kernel-associated image data accessible via a predetermined CSP kernels image-offsets structure.

Abstract translation: 一种用于减少由位于信号梯度影响的诊断图像区域中的重建缺陷像素簇引起的图像干扰的方法。 基于在模型中编码的内核对候选顺序，针对缺陷图像像素构成单独适应的中心对称重构（CSP）内核，从而使用像素的有效状态。 由缺陷像素影响的图像通过经由预定的CSP内核图像偏移结构可访问的内核相关图像数据的统计滤波或空间卷积来实时校正。