Infertility is becoming an issue for an increasing number of couples. The most common solution, in vitro fertilization, requires embryologists to carefully examine light microscopy images of human oocytes to determine their developmental potential. We propose an automatic system to improve the speed, repeatability, and accuracy of this process. We first localize individual oocytes and identify their principal components using CNN (U-Net) segmentation. We calculate several descriptors based on geometry and texture. The final step is an SVM classifier. Both the segmentation and classification training are based on expert annotations. The presented approach leads to the classification accuracy of 70%.

The aim of this study was to evaluate the efficiency of using meiotic spindle (MS) visibility and relative position to the polar body (PB) as indicators of oocyte maturation in order to optimize intracytoplasmic sperm injection (ICSI) timing. This was a cohort study of patients younger than 40 years with planned ICSI, the timing of which was determined by MS status, compared with those without MS evaluation. The angle between PB and MS and MS visibility were evaluated by optical microscope with polarizing filter. Oocytes with MS evaluation were fertilized according to MS status either 5–6 h after ovum pick-up (OPU) or 7–8 h after OPU. Oocytes without MS evaluation were all fertilized 5–6 h after OPU. For patients over 35 years visualization of MS influenced pregnancy rate (PR): 182 patients with MS visualization had 32% PR (58/182); while 195 patients without MS visualization had 24% PR (47/195). For patients under 35 years, visualization of MS did not influence PR: 140 patients with MS visualization had 41% PR (58/140), while 162 patients without MS visualization had 41% PR (66/162). Visualization of MS therefore appears to be a useful parameter for assessment of oocyte maturity and ICSI timing for patients older than 35.

No abstract

No abstract