Provided is a chromosome abnormality detection model, wherein steps for establishing the model comprise: obtaining a reference database; using an unsupervised learning method classifier to perform image transformation so as to obtain a plurality of reference chromosome karyotype images; performing preliminary classification according to the number of chromosomes in the plurality of reference chromosome karyotype images; analyzing the plurality of reference chromosome karyotype images to obtain at least one image feature value; and performing training by means of a convolutional neural network learning classifier to obtain the chromosome abnormality detection model. Further provided are a chromosome abnormality detection method and a chromosome abnormality detection system.
展开▼