Remote medical privacy data security sharing method based on heterogeneous federal learning

The invention relates to the technical field of data processing, in particular to a remote medical privacy data security sharing method based on heterogeneous federal learning, comprising the following steps: S10, a first terminal obtains local model update based on local medical privacy data training, signs the local model update based on an encryption algorithm, and sends the signed local model update to a second terminal; the signed information is sent to the second terminal; s20, the second terminal verifies the signature, updates and aggregates the local models after the verification is passed to obtain a global model, and decomposes the global model into a plurality of local models, one local model corresponds to one local model, the local models are correspondingly broadcasted to the first terminals, and the local models are broadcasted to the second terminals; after the first terminal receives the local model, training based on local medical privacy data to obtain local model update; and circularly ex