Cardiac magnetic resonance image segmentation method, system, equipment and medium

The invention discloses a heart magnetic resonance image segmentation method, system and device and a medium, and relates to the technical field of image segmentation, and the method comprises the steps: employing a training set to train a semi-supervised learning framework, and obtaining a heart magnetic resonance image segmentation model; the semi-supervised learning framework comprises a teacher model and a student model, and the training set comprises a first training subset formed by sample images carrying labels and a second training subset formed by sample images not carrying labels; in the process of training the semi-supervised learning framework by adopting the training set, taking the index moving average value of the weight in the student model as the weight of the teacher model; and inputting a to-be-segmented heart magnetic resonance image into the heart magnetic resonance image segmentation model to obtain a segmentation result. According to the method, the generalization ability of segmentatio