Human oocytes image classification method based on deep neural networks

Table 3 Mean accuracy coefficient \(\overline{Acc}\)

ID	Input layer	Accuracy
1	Oocyte image	0.93
2	Clear cytoplasm	0.93
3	Diffuse cytoplasmic granularity	0.90
4	Cytoplasmic granular area	0.93
5	Smooth endoplasmic reticulum cluster	0.94
6	Vacuoles	0.92
7	Dark cytoplasm	0.93
9	Multi-polar body	0.92
10	Fragmented first polar body	0.99
11	Second polar body	0.93
12	Perivitelline space	0.95
13	Zona pellucida	0.91

Network input: first polar body, germinal vesicle + class specified in the table. Output: oocyte type (PI/MI/MII)

ISSN: 1475-925X