Unlock instant, AI-driven research and patent intelligence for your innovation.

Multi-target sperm real-time monitoring method based on deep learning

A real-time monitoring and deep learning technology, applied in the field of sperm monitoring, can solve the problems of slow monitoring speed, low monitoring accuracy, unfavorable real-time tracking of sperm with multiple targets, etc., and achieve the effect of high accuracy and good adaptability

Pending Publication Date: 2020-12-29
TSINGHUA UNIV
View PDF9 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] (2) At present, the main sperm monitoring methods are based on traditional image processing methods, with low monitoring accuracy and slow monitoring speed, which is not conducive to real-time multi-target tracking of sperm

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Multi-target sperm real-time monitoring method based on deep learning
  • Multi-target sperm real-time monitoring method based on deep learning
  • Multi-target sperm real-time monitoring method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] The present invention will be described in detail below in conjunction with the accompanying drawings.

[0027] (1) Obtain real-time image data of sperm samples

[0028] Semen samples were obtained and incubated at 37°C for 30 minutes to liquefy the semen. Before capacitation, semen samples were separated by two-layer density centrifugation (300 × g, 20 min) with with (Nidacon, Sweden) for purification. Remove the supernatant, resuspend the sperm pellet and dissolve in 4.5 ml It was centrifuged again 38 times (500 x g, 10 min) in wash buffer (Nidacon, Sweden).

[0029] Use a high-speed camera to record the movement of sperm under the microscope, and get several groups of 50FPS sample videos. Then 1000 images are extracted from 6 different sets of videos, resulting in 6000 samples. Use LabelImagine software to label the sample and get the XML file corresponding to the image. The file records the position of the four corners of the sperm box in the image, and th...

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

No PUM Login to View More

Abstract

The invention discloses a multi-target sperm real-time monitoring method based on deep learning, and belongs to the technical field of sperm detection. The method comprises the following steps: (1) acquiring real-time image data of a sperm sample; (2) sampling a feature map of the multi-layer sperm sample image through an SSD network model, extracting features by using a CNN, and performing classification and regression; (3) transmitting the data to a KCF tracker; (4) for the lost part of the sperm target monitored by the SSD, monitoring the target in a motion state by using a KCF tracker, andupdating a result; (5) allocating a tracker by using a Kalman filter; and (6) estimating the distance between the targets according to the position monitored by the SSD and a Kalman filter, and matching the position with a KCF tracker to obtain real-time monitoring data of the multi-target sperms. According to the invention, real-time monitoring and tracking of a plurality of sperms are realized,and the method can be used for solving the black box problem of computer-assisted sperm analysis and can be applied to sperm tracking and extraction in a sperm plasma injection link.

Description

technical field [0001] The invention belongs to the technical field of sperm monitoring, and relates to a multi-target sperm real-time monitoring method based on deep learning. Background technique [0002] Human semen samples carry various degrees of accessibility information, from traditional counts and visual measures of motility to information requiring more computational methods. Human semen samples contain a wealth of information about reproductive potential and general health. With the aid of a microscope, fast-moving sperm can be observed. In addition to observing their motility characteristics, minute changes in sperm morphology can occur from moment to moment, ranging from gross features to molecular damage to molecular chromosomes. It is simply unreasonable to expect this complexity to be integrated through simple visual assessment. Although Computer Assisted Sperm Analysis (CASA) is increasingly used in assisted reproduction laboratories, the gold standard for ...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00G06T7/246G06T7/73G06N3/08G06N3/04
CPCG06T7/0012G06T7/246G06T7/73G06N3/08G06T2207/10056G06T2207/30024G06T2207/10016G06N3/045
Inventor 刘冉闫一默王卓然张博翾肖剑
Owner TSINGHUA UNIV