A device for predicting embryo pregnancy outcome based on segmented sampling video features

Abstract

The invention discloses an embryo pregnancy result prediction device based on segmented sampling video features, comprising a memory, a processor and a computer program, wherein the memory stores an embryo pregnancy result prediction model including n feature extraction modules and fusion classification modules, When the processor executes the computer program, the following steps are implemented: after receiving the embryonic development video and denoising the frame images, dividing the remaining frame images into n segments, and extracting m frames of images at medium time intervals in each segment; The feature extraction module inputs m frames of images. The extracted n*m frame images are respectively input to n feature extraction modules. After feature extraction, the extracted n-type feature maps are merged and input into the fusion classification module, and the output is calculated. The predicted probability of embryo pregnancy outcome; where n is a natural number greater than or equal to 4, and m is a natural number less than 20. The device can accurately output the predicted probability of embryo pregnancy outcome based on embryo development video.

Description

technical field [0001] The invention belongs to the field of medical video processing, in particular to a device for predicting embryo pregnancy results based on segmented sampling video features. Background technique [0002] Assisted reproductive technology is the abbreviation of assisted reproductive technology (Assisted Reproductive Technology, ART), which refers to the technology of using medical assistance to make infertile couples pregnant, including artificial insemination (AI) and in vitro fertilization-embryo transfer (In Vitro Fertilization and Embryo Transfer). EmbryoTransfer, IVF-ET) and its derivative technologies. In the past 40 years, IVF test-tube baby technology has undergone three generations of changes. The third-generation test-tube baby technology can not only solve female infertility and male infertility, but also meet people's needs for prenatal and postnatal care through pre-transplant screening. [0003] IVF technology uses artificial methods to fe...

AI Technical Summary

Problems solved by technology

It takes three to four days from in vitro fertilization to fertilized eggs to develop into blastocysts. Although the sampling interval is relatively large, a video still has about 800 frames of data, which is very difficult for existing machine learning methods or Inaccessible to deep learning methods

On the one hand, the current computing power cannot support the simultaneous processing of hundreds of frames of images; on the other hand, the existing models for processing time series, such as rnn, lstm, etc., have certain "forgetfulness" and cannot learn such long-span time information

Images