公开(公告)号：US20180232882A1

公开(公告)日：2018-08-16

申请号：US15750990

申请日：2016-09-22

申请人： Siemens Healthcare GmbH

发明人： Saikiran RAPAKA ,  Ali KAMEN ,  Noha EL-ZEHIRY ,  Bogdan GEORGESCU ,  Anton SCHICK ,  Uwe PHILIPPI ,  Oliver HAYDEN ,  Lukas RICHTER ,  Matthias UGELE

IPC分类号： G06T7/00 ,  G06T7/11 ,  G06K9/62 ,  G06T5/00 ,  G06T7/55 ,  G06K9/00 ,  G03H1/00 ,  G03H1/04 ,  G03H1/16 ,  G03H1/08 ,  G01N15/14

CPC分类号： G06T7/0014 ,  G01N15/1429 ,  G01N15/1434 ,  G01N15/1475 ,  G01N2015/1006 ,  G01N2015/1454 ,  G03H1/0005 ,  G03H1/0443 ,  G03H1/0866 ,  G03H1/16 ,  G03H2001/005 ,  G03H2001/0452 ,  G03H2001/0883 ,  G03H2210/55 ,  G06K9/0014 ,  G06K9/6255 ,  G06K2209/403 ,  G06T5/005 ,  G06T7/11 ,  G06T7/55 ,  G06T2207/10056 ,  G06T2207/20081 ,  G06T2207/30024

摘要： A computer-implemented method for analyzing digital holographic microscopy (DHM) data for hematology applications includes receiving a DHM image acquired using a digital holographic microscopy system. The DHM image comprises depictions of one or more cell objects and background. A reference image is generated based on the DHM image. This reference image may then be used to reconstruct a fringe pattern in the DHM image into an optical depth map.

公开(公告)号：US20210334970A1

公开(公告)日：2021-10-28

申请号：US17228813

申请日：2021-04-13

申请人： Siemens Healthcare GmbH

发明人： Siqi LIU ,  Yuemeng LI ,  Arnaud Arindra ADIYOSO ,  Bogdan GEORGESCU ,  Sasa GRBIC ,  Ziming QIU ,  Zhengyang SHEN

IPC分类号： G06T7/00 ,  G06T7/62

摘要： A computer-implemented method is for classifying a lesion. In an embodiment, the method includes receiving a first medical image of an examination volume, the first medical image corresponding to a first examination time; receiving a second medical image of the examination volume, the second medical image corresponding to a second examination time, different from the first examination time; determining a first lesion area corresponding to a lesion within the first medical image; determining a registration function based on a comparison of the first medical image and the second medical image; determining a second lesion area within the second medical image based on the registration function and the first lesion area; and classifying the lesion within the first medical image based on the second lesion area. A computer-implemented method for providing a trained classification function, a classification system, and computer program products and computer-readable media are also disclosed.