Searching of data structures in pre-processing data for a machine learning classifier

A data processing system configured to pre-process data for a machine learning classifier, the data processing system comprising:

• an input port that receives one or more radiological images;
• a shared memory data store that stores the one or more radiological images, with each of the one or more radiological images being written to the shared memory data store;
• and
• at least one processor configured to perform operations comprising:
• extracting image data representing a biological structure by image processing one of the one or more radiological images;
• identifying one or more portions of the biological structure each having a biological signature based on comparing the biological structure to a library of specified biological signatures, wherein a given biological signature is identified based on a fuzzy matching of features of the biological structure to features of the specified biological signatures;
• generating a data structure from the biological structure, wherein the data structure includes a plurality of nodes connected with edges, each node in the data structure represents one of the biological signatures, and wherein each edge specifies a relationship between a first node and a second node;
• selecting a particular biological signature of the biological structure in the data structure;
• for the particular biological signature that is selected,
• identifying each instance of the particular biological signature in the data structure;
• segmenting the data structure around instances of the particular biological signature;
• and
• identifying, based on the segmenting, one or more sequences of biological signatures connected to the particular biological signature in the data structure, each of the one or more sequences identified being different from other identified sequences of biological signatures connected to the particular biological signature in the data structure, each of the one or more sequences including a connected set of biological data signatures in the data structure that are connected to the particular biological data signature, wherein each connection represents a relationship between a first biological data signature and a second biological data signature of the set of data signatures;
• generating, independent of precoding any sequence of biological data signatures into a rule, a logical ruleset, wherein each logical rule of the logical ruleset is a sequence of biological signatures of the one or more sequences of biological signatures that are identified, and wherein a logical rule of the logical ruleset comprises a shape rule specifying one or a sequence of more shapes that are either permitted to be adjacent in the sequence of data signatures, restricted from being adjacent in the sequence of data signatures, or replacing one or more prior shapes of the sequence of data signatures;
• executing one or more classifiers against the logical ruleset to classify the one or more radiological images received by the input port;
• and
• generating, based on the executing, one or more additional logical rules for the logical ruleset, the one or more additional logical rules specifying an additional relationship between at least two shapes of the plurality of shapes.