An Open Dataset of Annotated Metaphase Cell Images for Chromosome Identification

Abstract

Chromosomes are a principal target of clinical cytogenetic studies. While chromosomal analysis is an integral part of prenatal care, the conventional manual identification of chromosomes in images is time-consuming and costly. This study developed a chromosome detector that uses deep learning and that achieved an accuracy of 98.88% in chromosomal identification. Specifically, we compiled and made available a large and publicly accessible database containing chromosome images and annotations for training chromosome detectors. The database contains five thousand 24 chromosome class annotations and 2,000 single chromosome annotations. This database also contains examples of chromosome variations. Our database provides a reference for researchers in this field and may help expedite the development of clinical applications.

Fig. 1

Example of a raw chromosome image with three annotated datasets. (a) Original chromosome image taken from fetal amniotic fluid; (b) annotation of single chromosomes; (c) annotations of 24 chromosome categories.

Fig. 2

Karyotype (46, XY) produced by an expert processing from the original chromosome map.

Fig. 3

Examples of difficult image according to three definitions. (a) Multiple chromosomal overlaps; (b) suboptimal dark and light banding; (c) excessively elongated chromosomes.

Fig. 4

Two images containing detected chromosomes. (a) Simple and (b) difficult images. Detection accuracy was 100% with a simple image. Multiple overlapping and adherent chromosomes make detection more difficult. Chromosomes not captured correctly were those that fell between three overlapping chromosomes.

Fig. 5

Curve of number of images and model accuracy (%).