Hardware apparatus for selecting radiomic features, method for building a learning model using significant radiomic features, and method for supporting patient treatment using a learning model

A method for building a learning model using radiomic features, the method comprising:

• receiving, by a hardware apparatus, a plurality of radiomic features extracted from medical images;
• generating, by the hardware apparatus, a feature map including one or more statistical relevance indicators and one or more collinearity indicators for each of the radiomic features;
• selecting, by the hardware apparatus, significant radiomic features by removing some of the radiomic features using the feature map;
• and
• training, by the hardware apparatus, a learning model using the significant radiomic features,
• wherein selecting the significant radiomic features comprises:
• performing, by the hardware apparatus, univariate statistical analysis on each of the radiomic features with respect to an output value of the learning model to compute the one or more statistical relevance indicators;
• performing, by the hardware apparatus, a first removal of features whose relevance to the output value is below a threshold, from a candidate radiomic feature set including the radiomic features, based on at least one first statistical relevance indicator selected from the one or more statistical relevance indicators;
• performing, by the hardware apparatus, a second removal of redundant features from the candidate radiomic feature set remaining after the first removal based on the one or more collinearity indicators;
• performing, by the hardware apparatus, a third removal of features whose relevance to the output value is below a threshold, from the candidate radiomic feature set remaining after the second removal, based on at least one second statistical relevance indicator selected from the one or more statistical relevance indicators;
• and
• applying, by the hardware apparatus, an embedded method to the candidate radiomic feature set remaining after the third removal to select final radiomic features.

A hardware apparatus comprising:

• a storage device configured to store medical images;
• a memory configured to store instructions for radiomic feature filtering;
• and
• a processor configured to execute the instructions to:
• (i) extract a plurality of radiomic features from the medical images;
• (ii) perform univariate statistical analysis on each of the radiomic features with respect to an output value of a learning model to compute a plurality of statistical relevance indicators;
• (iii) perform a first removal of features whose relevance to the output value is below a threshold, from a candidate radiomic feature set including the radiomic features, based on at least one first statistical relevance indicator selected from the plurality of statistical relevance indicators;
• (iv) perform a second removal of redundant features from the candidate radiomic feature set remaining after the first removal based on collinearity criteria;
• (v) perform a third removal of features whose relevance to the output value is below a threshold, from the candidate radiomic feature set remaining after the second removal, based on at least one second statistical relevance indicator selected from the plurality of statistical relevance indicators, wherein the at least one second statistical relevance indicator comprises the at least one first statistical relevance indicator;
• and
• (vi) apply an embedded method to the candidate radiomic feature set remaining after the third removal to select final radiomic features.