Transforming visual preference terminology for radiographic images
View Patent ↗A method for transforming visual preference terminology into a set of fundamental image-quality attributes, and using these attributes to produce a display ready radiographic image that meets the desired appearance. The visual preference terminology can be specified in a hierarchical structure providing a flexible and configurable interface to a radiographic system.
1. A method of processing a digital medical image, comprising:
determining an exam-type probability corresponding to the digital medical image;
accessing a first prediction model corresponding to a selected visual preference;
calculating a second prediction model using the first prediction model and the exam-type probability;
determining a plurality of image quality attributes based on at least the second prediction model; and
generating a processed digital medical image based on the plurality of image quality attributes.
2. The method of claim 1 , wherein the first prediction model is obtained from a database.
3. The method of claim 1 , wherein the exam-type probability corresponds to a probability of 1) body part and 2) projection of the digital medical image.
4. The method of claim 1 , wherein the exam-type probability corresponds to a probability of 1) body part, 2) projection, and 3) examination purpose of the digital medical image.
5. The method of claim 1 , wherein the second prediction model comprises a weighted sum of the first prediction model and the exam-type probability.
6. The method of claim 1 , further including generating the plurality of image quality attributes based on a plurality of features extracted from the digital medical image.
7. The method of claim 6 , wherein the type and number of the plurality of features extracted are based on the selected visual preference.
8. The method of claim 1 , wherein a distinct prediction model is generated for each image quality attribute of the plurality of image quality attributes.
9. The method of claim 1 , wherein the plurality of image quality attributes comprises at least one of brightness, global contrast, detail contrast, and sharpness.
10. A method of processing a digital medical image, comprising:
determining an exam-type probability corresponding to the digital medical image;
accessing a first prediction model corresponding to a selected visual preference;
calculating a second prediction model using the first prediction model and the exam-type probability;
determining image quality attributes representative of the selected visual preference;
generating a processed digital medical image based on the image quality attributes; and
displaying, transmitting, or printing the processed digital medical image.
11. A method of processing a digital medical image, comprising the steps of:
accessing the digital medical image;
allowing a user to select a visual preference;
determining an exam-type probability;
accessing a first prediction model corresponding to the selected visual preference;
determining a second prediction model using the first prediction model and the exam-type probability;
extracting features from the accessed digital medical image;
using the extracted features and the second prediction model to generate image quality attributes; and
using the image quality attributes to generate a processed digital medical image.
12. The method of claim 11 , further comprising the step of displaying, transmitting, or printing the processed digital medical image.
13. computer-readable medium encoded with computer-executable instructions to perform the method of claim 11 .