-
1.
公开(公告)号:US20160048957A1
公开(公告)日:2016-02-18
申请号:US14823007
申请日:2015-08-11
Inventor: Dong-Joo Kim
CPC classification number: G06T11/60 , A61B6/032 , A61B6/501 , A61B6/505 , A61B6/5205 , A61B6/5258 , G06T5/005 , G06T7/12 , G06T11/008 , G06T2207/10081 , G06T2207/30008 , G06T2207/30016 , G06T2211/40
Abstract: A method for eliminating an artifact of a computed tomography image is provided. The method includes eliminating a machine artifact from a cross-sectional computed tomography scans of the brain; setting boundary points positioned on a horizontal axis and a vertical axis in the computed tomography image, from which the machine artifact has been eliminated; resetting a boundary point positioned at a boundary between the skull of the brain and the brain tissue based on the boundary points; detecting a position of a fracture of the brain in the computed tomography image, in which the boundary point has been reset, and overlaying a fracture region with a skull pixel; and eliminating an artifact from the computed tomography image, which has been overlaid with the skull pixel.
Abstract translation: 提供了一种消除计算机断层摄影图像伪像的方法。 该方法包括从脑的横断面计算机断层扫描中消除机器伪影; 设置位于计算机断层摄影图像中的水平轴和垂直轴上的边界点,从其中消除了机器人造物; 基于边界点重置位于脑颅骨与脑组织之间边界的边界点; 检测计算机断层摄影图像中的骨折位置,其中边界点已被复位,并且覆盖具有颅骨像素的骨折区域; 并从计算机断层摄影图像中消除了与颅骨像素重叠的伪像。
-
公开(公告)号:US20180110473A1
公开(公告)日:2018-04-26
申请号:US15564778
申请日:2016-04-07
Inventor: Dong-Joo Kim
CPC classification number: A61B5/7278 , A61B5/021 , A61B5/02116 , A61B5/4064 , A61B5/4842 , A61B5/7225 , A61B5/7264 , A61B5/7275 , A61B5/742
Abstract: The present invention relates to a device and a method for detecting a feature of an arterial blood pressure (ABP) waveform. The feature detecting method of an ABP waveform according to an exemplary embodiment of the present invention includes: searching a peak and a trough; detecting a systolic peak based on a time interval between a peak and a neighboring peak and an average pressure value; detecting a pulse onset with respect to a trough directly before the systolic peak; extracting a candidate of a dicrotic notch based on a magnitude of the systolic peak and a measurement time; detecting a point having a lowest pressure value among candidates of the dicrotic notch as a dicrotic notch and detecting a dicrotic peak based on the dicrotic notch; and classifying the detected dicrotic notches into a normal notch and a transient notch.
-
3.
公开(公告)号:US20180103862A1
公开(公告)日:2018-04-19
申请号:US15567858
申请日:2016-04-20
Inventor: Dong-Joo Kim
IPC: A61B5/03 , A61B5/0205 , A61B5/00
CPC classification number: A61B5/031 , A61B5/0205 , A61B5/021 , A61B5/024 , A61B5/4842 , A61B5/7278 , A61B5/7425 , A61B5/743 , G06K9/46
Abstract: The present invention relates to a device and a method for detecting a peak of an intracranial pressure (ICP) waveform using a morphological feature of an arterial blood pressure waveform. A peak extracting method of an ICP waveform using a morphological feature of an arterial blood pressure waveform according to an aspect of the present invention includes: extracting a pulse onset from a continuous ICP waveform based on systolic peak from arterial blood pressure waveform; dividing individual ICP waveforms in the continuous ICP waveform based on the pulse onset; deriving a derivative value from each of the ICP waveforms to extract a peak, a trough, and a flat; calculating latencies from the pulse onset extracted in each of the ICP waveforms to the extracted peaks to cluster peaks with a similar time interval and generate a peak cluster; searching a notch from each of the ICP waveforms based on the latency of a dicrotic notch of the arterial blood pressure waveform; and extracting P1, P2, and P3 peaks from each of the ICP waveforms by referring to the searched notch of the ICP.
-
公开(公告)号:US11224350B2
公开(公告)日:2022-01-18
申请号:US16122219
申请日:2018-09-05
Inventor: Dong-Joo Kim
Abstract: A method for detecting an artifact of a blood pressure signal includes measuring consecutive signals of blood pressure in real time, dividing the consecutive signals into a plurality of unit waveforms by searching systolic starting points and peak points from the measured signals, determining two outputs by applying N consecutive unit waveforms among the plurality of unit waveforms as an input to a deep belief network (DBN), and determining whether the signal is an artifact or a normal signal by using the two determined outputs. According to the present disclosure, it is possible to minimize various false alarms of bio-signal streams which are measured in real time by automatically eliminating ABP artifacts using a deep belief network (DBN) which is one of deep neural network (DNN) models capable of learning causes and shapes of various types of artifacts together.
-
公开(公告)号:US10799130B2
公开(公告)日:2020-10-13
申请号:US15567858
申请日:2016-04-20
Inventor: Dong-Joo Kim
Abstract: The present invention relates to a device and a method for detecting a peak of an intracranial pressure (ICP) waveform using a morphological feature of an arterial blood pressure waveform. A peak extracting method of an ICP waveform using a morphological feature of an arterial blood pressure waveform according to an aspect of the present invention includes: extracting a pulse onset from a continuous ICP waveform based on systolic peak from arterial blood pressure waveform; dividing individual ICP waveforms in the continuous ICP waveform based on the pulse onset; deriving a derivative value from each of the ICP waveforms to extract a peak, a trough, and a flat; calculating latencies from the pulse onset extracted in each of the ICP waveforms to the extracted peaks to cluster peaks with a similar time interval and generate a peak cluster; searching a notch from each of the ICP waveforms based on the latency of a dicrotic notch of the arterial blood pressure waveform; and extracting P1, P2, and P3 peaks from each of the ICP waveforms by referring to the searched notch of the ICP.
-
公开(公告)号:US09721368B2
公开(公告)日:2017-08-01
申请号:US14823007
申请日:2015-08-11
Inventor: Dong-Joo Kim
CPC classification number: G06T11/60 , A61B6/032 , A61B6/501 , A61B6/505 , A61B6/5205 , A61B6/5258 , G06T5/005 , G06T7/12 , G06T11/008 , G06T2207/10081 , G06T2207/30008 , G06T2207/30016 , G06T2211/40
Abstract: A method for eliminating an artifact of a computed tomography image is provided. The method includes eliminating a machine artifact from a cross-sectional computed tomography scans of the brain; setting boundary points positioned on a horizontal axis and a vertical axis in the computed tomography image, from which the machine artifact has been eliminated; resetting a boundary point positioned at a boundary between the skull of the brain and the brain tissue based on the boundary points; detecting a position of a fracture of the brain in the computed tomography image, in which the boundary point has been reset, and overlaying a fracture region with a skull pixel; and eliminating an artifact from the computed tomography image, which has been overlaid with the skull pixel.
-
公开(公告)号:US09675313B2
公开(公告)日:2017-06-13
申请号:US14838757
申请日:2015-08-28
Inventor: Dong-Joo Kim
CPC classification number: A61B6/5217 , A61B6/032 , A61B6/501 , A61B6/505 , A61B6/506 , A61B6/5205 , A61B6/5258 , G06T5/005 , G06T7/11 , G06T7/136 , G06T7/187 , G06T11/008 , G06T2207/10081 , G06T2207/30016
Abstract: A method for densitometric analysis of a computed tomography image is provided. The method includes converting CT numbers of each pixel of a cross-sectional computed tomography image of the brain into Hounsfield unit (HU) values; eliminating an artifact from the computed tomography image of which the CT numbers of each pixel have been converted into the HU values, and extracting the brain tissues; and extracting a density distribution of the HU values from the computed tomography image, in which the artifact has been removed and the brain tissues have been extracted.
-
公开(公告)号:US11690526B2
公开(公告)日:2023-07-04
申请号:US16938243
申请日:2020-07-24
Inventor: Dong-Joo Kim
IPC: A61B5/03 , A61B5/021 , A61B5/024 , A61B5/0205 , A61B5/00
CPC classification number: A61B5/031 , A61B5/021 , A61B5/024 , A61B5/0205 , A61B5/743 , A61B5/7425 , A61B5/4842 , A61B5/7278
Abstract: The present invention relates to a device and a method for detecting a peak of an intracranial pressure (ICP) waveform using a morphological feature of an arterial blood pressure waveform. A peak extracting method of an ICP waveform using a morphological feature of an arterial blood pressure waveform according to an aspect of the present invention includes: extracting a pulse onset from a continuous ICP waveform based on systolic peak from arterial blood pressure waveform; dividing individual ICP waveforms in the continuous ICP waveform based on the pulse onset; deriving a derivative value from each of the ICP waveforms to extract a peak, a trough, and a flat; calculating latencies from the pulse onset extracted in each of the ICP waveforms to the extracted peaks to cluster peaks with a similar time interval and generate a peak cluster; searching a notch from each of the ICP waveforms based on the latency of a dicrotic notch of the arterial blood pressure waveform; and extracting P1, P2, and P3 peaks from each of the ICP waveforms by referring to the searched notch of the ICP.
-
公开(公告)号:US10856817B2
公开(公告)日:2020-12-08
申请号:US15564778
申请日:2016-04-07
Inventor: Dong-Joo Kim
Abstract: The present invention relates to a device and a method for detecting a feature of an arterial blood pressure (ABP) waveform. The feature detecting method of an ABP waveform according to an exemplary embodiment of the present invention includes: searching a peak and a trough; detecting a systolic peak based on a time interval between a peak and a neighboring peak and an average pressure value; detecting a pulse onset with respect to a trough directly before the systolic peak; extracting a candidate of a dicrotic notch based on a magnitude of the systolic peak and a measurement time; detecting a point having a lowest pressure value among candidates of the dicrotic notch as a dicrotic notch and detecting a dicrotic peak based on the dicrotic notch; and classifying the detected dicrotic notches into a normal notch and a transient notch.
-
10.
公开(公告)号:US20160058408A1
公开(公告)日:2016-03-03
申请号:US14838757
申请日:2015-08-28
Inventor: Dong-Joo Kim
CPC classification number: A61B6/5217 , A61B6/032 , A61B6/501 , A61B6/505 , A61B6/506 , A61B6/5205 , A61B6/5258 , G06T5/005 , G06T7/11 , G06T7/136 , G06T7/187 , G06T11/008 , G06T2207/10081 , G06T2207/30016
Abstract: A method for densitometric analysis of a computed tomography image is provided. The method includes converting CT numbers of each pixel of a cross-sectional computed tomography image of the brain into Hounsfield unit (HU) values; eliminating an artifact from the computed tomography image of which the CT numbers of each pixel have been converted into the HU values, and extracting the brain tissues; and extracting a density distribution of the HU values from the computed tomography image, in which the artifact has been removed and the brain tissues have been extracted.
Abstract translation: 提供了一种计算机断层摄影图像的光密度分析方法。 该方法包括将脑的横截面计算机断层摄影图像的每个像素的CT数转换为Hounsfield单位(HU)值; 从计算机断层摄影图像中消除每个像素的CT数已被转换成HU值的神经元,并提取脑组织; 以及从计算机断层摄影图像中提取HU值的密度分布,其中伪像被去除并且脑组织已被提取。
-
-
-
-
-
-
-
-
-