Abstract

In the field of chromosome karyotype analysis, performing straightening preprocessing on chromosomes is a critical step to improve the accuracy of chromosome identification. Previous studies have typically relied on geometric algorithms; however, during the straightening process, external perturbations caused by geometric factors often result in the distortion or deformation of chromosome banding patterns, leading to the loss of key feature information and the blurring of details. This, in turn, reduces the model’s representation capability and generalization performance. In this paper, we propose a novel disentangled representation learning framework aimed at exploring the semantic disentanglement trajectories within generative models to generate chromosome images with purer straightening semantics. Our framework consists of four main components: the Generator(G), the Semantic Trajectory Explorer(STE), the Disentangled Encoder(DE), and the Classifier(C). The Generator is responsible for generating chromosome images, the Semantic Trajectory Explorer explores disentangled semantic trajectories within the latent space of the generative model, the Disentangled Encoder maps chromosome images into the disentangled feature space, and the Classifier predicts the chromosome type based on these features. The framework employs a disentanglement loss to force the Semantic Trajectory Explorer to find disentangled semantic trajectories within the latent space of the Generator and uses classification loss to ensure that content information remains unchanged along the semantic trajectory, thereby achieving chromosome straightening without altering karyotype features. Evaluation results show that our method achieves superior performance in terms of both the Rectangular Aspect Ratio(RAR) metric and the Downstream Classification Accuracy (DCA) metric.

Citations

Authors (4)