Method for automatically acquiring qualified four-cavity heart tangent image in fetal heart ultrasonic video
A fetal heart and section image technology, applied in the field of fetal heart ultrasound imaging, can solve the problems that the target detection model cannot accurately reflect the clarity of the four-chamber heart area, anatomical structure positioning, and multiple missed detections
Pending Publication Date: 2022-05-10
CHINA THREE GORGES UNIV
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0013] The classification model cannot locate the specified anatomical structure, so it cannot focus all the attention of the model on the four-chamber heart area; while the target detection model cannot accurately reflect the clarity of the four-chamber heart area, resulting in multiple missed detections at high thresholds , the problem of multiple false detections at low thresholds
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0065] A method for automatically acquiring qualified four-chamber view images in fetal heart ultrasound video, including:
[0066] Model training stage, described model training stage comprises the following steps, as shown in Figure 3 (a):
[0067] Step 1: Framing the fetal cardiac ultrasound video;
[0068] Step 2: Manually select qualified four-chamber heart view images, and mark the four-chamber heart and descending aorta;
[0069] Step 3: Build and train the YOLOv5 target detection network;
[0070] Step 4: Build and train the Darknet53 classification network;
[0071] The model use stage, the model use stage includes the following steps, as shown in Figure 3 (b):
[0072] Step 5: Framing processing, and using the YOLOv5 target detection network to detect the four-chamber heart and descending aorta in the fetal heart ultrasound video of the section to be extracted frame by frame;
[0073] Step 6: Record whether there is a descending aorta in the video and extract the...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for automatically acquiring a qualified four-cavity heart tangent image in a fetal heart ultrasonic video. The method comprises the following steps: framing the fetal heart ultrasonic video; qualified four-cavity heart section images are screened out manually, and four-cavity heart and descending aorta areas are marked; building and training a YOLOv5 target detection network; a Darknet53 classification network is established and trained; framing processing is carried out, and a four-cavity heart and a descending aorta in the fetal heart ultrasonic video of the section to be extracted are detected frame by frame by using a YOLOv5 target detection network; recording whether a descending aorta exists in the video and extracting a detected four-cavity heart area image; and a Darknet53 classification network is utilized to classify the extracted quasi four-cavity heart area, meanwhile, scores of four-cavity heart tangent plane images classified to be qualified are calculated, the tangent plane images are sorted according to the scores, and the optimal four-cavity heart tangent plane image in the video is automatically given. According to the method, the qualified four-cavity heart tangent plane image in the fetal heart ultrasonic video can be effectively obtained, and meanwhile, the optimal tangent plane image can be given.
Description
technical field [0001] The invention relates to the technical field of fetal heart ultrasound imaging, in particular to a method for automatically acquiring qualified four-chamber heart section images in fetal heart ultrasound video. Background technique [0002] Ultrasound imaging technology has been widely used in the screening and diagnosis of prenatal fetal congenital heart disease due to its painless, non-invasive, non-ionizing radiation, and real-time imaging advantages. Compared with CT, MRI and other imaging methods, the quality of ultrasound images contains more noise, and is also affected by fetal activity. In addition, there are subjective differences in the operation techniques of ultrasound doctors and the process of reading pictures. These factors lead to a relatively low detection rate of fetal congenital heart disease. About 300,000 fetuses die of congenital heart disease during pregnancy or childbirth every year. [0003] As documented in [1]: Vullings R, F...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00G06V10/764G06V10/82G06K9/62
CPCG06T7/0012G06T2207/10132G06T2207/10016G06T2207/30048G06T2207/20081G06T2207/20084G06T2207/30101G06F18/241
Inventor 徐光柱钱奕凡王阳刘蓉周军雷帮军
Owner CHINA THREE GORGES UNIV