System and method for automated gamete selection

Abstract

In variants, a method for automated gamete selection can include: sampling a video of a scene having a plurality of gametes, tracking each gamete across successive images, and determining attribute values for a gamete, and selecting the gamete. The attribute values can be determined using a model trained to predict the attribute values for the gamete based on a video.

We claim: 
     
         1 . A system for gamete selection, comprising:
 a camera configured to sample a video of a gamete; and   a processor configured to:
 receive a selection metric for the gamete, provided by each of a set of specialists, based on the video; 
 determine an overall metric for the gamete based on the selection metrics provided by each specialist in the set; and 
 train a model to predict the overall metric for the gamete based on the video. 
   
     
     
         2 . The system of  claim 1 , wherein the overall metric is a selection probability. 
     
     
         3 . The system of  claim 1 , wherein the selection metric comprises at least one of: a selection rating, a motility rating, or a morphology rating. 
     
     
         4 . The system of  claim 1 , wherein the processor is further configured to train the model using development data collected for an embryo fertilized using the gamete. 
     
     
         5 . The system of  claim 1 , wherein the processor is further configured to receive a secondary metric for the gamete provided by each specialist in the set of specialists, wherein the model is further trained to predict the secondary metric for the gamete based on the video. 
     
     
         6 . The system of  claim 1 , wherein the trained model is used to select a gamete for an assistive reproductive technology. 
     
     
         7 . A method, comprising:
 sampling a timeseries of measurements of a gamete;   automatically extracting a timeseries of feature values for the gamete based on the timeseries of measurements; and   selecting the gamete from a set of gametes based on the timeseries of feature values, using a trained model.   
     
     
         8 . The method of  claim 7 , further comprising determining a timeseries of gamete attribute values based on the timeseries of feature values, wherein the gamete is selected based on the timeseries of gamete attribute values. 
     
     
         9 . The method of  claim 8 , wherein gamete attribute values comprise at least one of:
 a selection probability score, a motility score, a DNA fragmentation score, or a morphology score.   
     
     
         10 . The method of  claim 8 , wherein the timeseries of gamete attribute values are determined using a second trained model. 
     
     
         11 . The method of  claim 7 , further comprising determining attention scores corresponding to frames of the video, wherein the timeseries of feature values are extracted based on the attention scores. 
     
     
         12 . The method of  claim 11 , wherein an attention score corresponding to a frame is positively correlated with the frame depicting a flat side of the gamete. 
     
     
         13 . The method of  claim 7 , wherein the gamete is located within a scene, the method further comprising:
 tracking the gamete across the scene to a new scene segment;   detecting a new gamete within the new scene segment;   sampling a new timeseries of measurements of the new gamete;   automatically extracting a new timeseries of feature values for the new gamete based on the new timeseries of measurements; and   selecting the new gamete from the set of gametes based on the new timeseries of feature values.   
     
     
         14 . The method of  claim 7 , wherein the trained model is a machine learning model, trained by:
 providing a video of a training gamete to a set of specialists;   from each specialist in the set, receiving a selection metric for the training gamete;   aggregating the selection metrics received from each specialist in the set to determine an overall selection metric for the training gamete; and   training the model to predict the overall metric for the training gamete based on the video of the training gamete.   
     
     
         15 . The method of  claim 7 , wherein the gamete is automatically physically isolated from the set of gametes.