Sequential transmission of compressed medical image data

A medical system comprising:

• a remote memory storing remote machine executable instructions and a decoder neural network being a decoder portion of an autoencoder, wherein the decoder neural network is configured to output an approximation of a medical image data in response to receiving at least a part of a feature vector descriptive of the medical image data as input, wherein the feature vector comprises a ranking assigning an importance to elements of the feature vector and the feature vector and the ranking is determined in response to inputting the medical image data into an encoder neural network being an encoder portion of the autoencoder in a local computational system;
• and
• a remote computational system, wherein execution of the remote machine executable instructions causes the remote computational system to:
• sequentially receive portions of the feature vector via the network connection;
• assemble the received portions of the feature vector into the part of the feature vector;
• receive an approximation of the medical image data in response to inputting to the part of the feature vector into the decoder neural network of the autoencoder, and wherein the part of the feature vector is input into the decoder neural network before all portions of the feature vector have been received.

A computer program comprising local machine executable instructions for execution by a local computational system, wherein execution of the machine executable instructions further causes the local computational system to:

• receive medical image data;
• receive a feature vector in response to inputting the medical image data into an encoder neural network being an encoder portion of an autoencoder;
• determine and assign a ranking to elements of the feature vector;
• receive the feature vector descriptive of the medical image data, wherein the feature vector is configured to be input into a decoder neural network being a decoder portion of the autoencoder, wherein the decoder neural network is configured to output an approximation of the medical image data when receiving at least a part of the feature vector as input, wherein the feature vector comprises the ranking assigning an importance to the elements of the feature vector;
• and
• sequentially transmit portions of the feature vector to a remote computational system via a network connection, wherein the portions of the features vector with a higher importance are transmitted first.

A medical imaging method, wherein the method comprises:

• receiving medical image data;
• receiving a feature vector in response to inputting the medical image data into an encoder neural network being an encoder portion of an autoencoder;
• determining and assigning a ranking to elements of the feature vector;
• receiving the feature vector descriptive of the medical image data by a local computational system, wherein the feature vector is configured to be input into a decoder neural network being a decoder portion of the autoencoder, wherein the decoder neural network is configured to output an approximation of the medical image data when receiving at least a part of the feature vector as input, wherein the feature vector comprises the ranking assigning an importance to elements of the feature vector;
• and
• sequentially transmitting portions of the feature vector to a remote computational system via a network connection by the local computational system, wherein the portions of the features vector with a higher importance are transmitted first,
• sequentially receiving the portions of the feature vector via the network connection by the remote computational system;
• assemble received portions of the feature vector into the part of the feature vector by the remote computational system;
• and
• receiving the approximation of the medical image data in response to inputting to the part of the feature vector into the decoder neural network by the remote computational system, and wherein the received portions of the feature vector is input into the decoder neural network before all portions of the feature vector have been received.