Automatic feature extraction for improved validation of image data

A computer program product for improved validation of image data, the computer program product comprising a non-transitory computer-readable medium comprising code causing an apparatus to:

• receive an image file;
• apply one or more pre-processing steps to the image file to generate a standardized image file;
• process the standardized image file using a hierarchical stacked deformable convolutional neural network (HSDCNN) to identify at least one image feature, wherein the at least one image feature is associated with a correlation score;
• input the at least one image feature and the correlation score into a large language model, wherein the large language model is trained to output a decision associated with the image file;
• and
• based on the decision, execute a validation process associated with the image file.

A method for improved validation of image data, the method comprising:

• receiving an image file;
• applying one or more pre-processing steps to the image file to generate a standardized image file;
• processing the standardized image file using a hierarchical stacked deformable convolutional neural network (HSDCNN) to identify at least one image feature, wherein the at least one image feature is associated with a correlation score;
• inputting the at least one image feature and the correlation score into a large language model, wherein the large language model is trained to output a decision associated with the image file;
• and
• based on the decision, executing a validation process associated with the image file.