Medical image pre-processing at the scanner for facilitating joint interpretation by radiologists and artificial intelligence algorithms

1. A method for medical image acquisition and pre-processing, comprising:

acquiring raw medical image data by performing a medical image scan of a patient using a medical image scanner;

acquiring input data associated with the medical image scan of the patient and available downstream automated image analysis algorithms;

selecting a set of pre-processing algorithms to apply to the raw medical image data based on the input data associated with the medical image scan of the patient and the available downstream automated image analysis algorithms using a trained machine learning based model, wherein selecting a set of pre-processing algorithms to apply to the raw medical image data based on the input data associated with the medical image scan of the patient and the available downstream automated image analysis algorithms using a trained machine learning based model comprises:

selecting a first set of one or more pre-processing algorithms for generating one or medical images for human visual interpretation, and

selecting a second set of one or more pre-processing algorithms for generating one or more medical images for automated image analysis using one or more of the available downstream automated image analysis algorithms; and

generating one or more medical images from the raw medical image data by applying the selected set of pre-processing algorithms to the raw medical image data, wherein generating one or more medical images from the raw medical image data by applying the selected set of pre-processing algorithms to the raw medical image data comprises:

generating one or more medical images for human visual interpretation by applying the first set of one or more pre-processing algorithms to the raw medical image data, and

generating one or more medical images for automated image analysis by applying the second set of pre-processing algorithms to the raw medical image data.

2. The method of claim 1 , further comprising:

displaying the one or more medical images for human visual interpretation generated using the first set of pre-processing algorithms on a display device; and

providing the one or more medical images for automated image analysis generating using the second set of pre-processing algorithms as input to one or more of the available downstream automated image analysis algorithms.

3. The method of claim 2 , further comprising:

displaying results of the automated image analysis performed on the one or more medical images generated using the second set of pre-processing algorithms by the one or more of the available downstream automated image analysis algorithms on the one or more medical images generated using the first set of pre-processing algorithms.

4. The method of claim 1 , wherein the first set of pre-processing algorithms and the second set of pre-processing algorithms include reconstruction algorithms with different settings.

5. The method of claim 1 , wherein the second set of pre-processing algorithms includes one or more pre-processing algorithms corresponding to each of a plurality of available downstream automated image analysis algorithms predicted to be applied to the medical image scan, and generating one or more medical images for automated image analysis by applying the second set of pre-processing algorithms to the raw medical image data comprises:

generating a respective medical image suited for each of the plurality available downstream automated image analysis algorithms predicted to be applied to the medical image scan using the corresponding one or more pre-processing algorithms in the second set of pre-processing algorithms.

6. The method of claim 1 , wherein the trained machine learning based model is trained based on a training database of training images acquired using a variety of pre-processing algorithms by comparing output interpretations of the training images determined using one or more available downstream automated image processing algorithms with ground truth interpretations of the training images.

7. An apparatus for medical image acquisition and pre-processing, comprising:

means for acquiring raw medical image data by performing a medical image scan of a patient using a medical image scanner;

means for acquiring input data associated with the medical image scan of the patient and available downstream automated image analysis algorithms;

means for selecting a set of pre-processing algorithms to apply to the raw medical image data based on the input data associated with the medical image scan of the patient and the available downstream automated image analysis algorithms using a trained machine learning based model, wherein the means for selecting a set of pre-processing algorithms to apply to the raw medical image data based on the input data associated with the medical image scan of the patient and the available downstream automated image analysis algorithms using a trained machine learning based model comprises:

means for selecting a first set of one or more pre-processing algorithms for generating one or medical images for human visual interpretation, and

means for selecting a second set of one or more pre-processing algorithms for generating one or more medical images for automated image analysis using one or more of the available downstream automated image analysis algorithms; and

means for generating one or more medical images from the raw medical image data by applying the selected set of pre-processing algorithms to the raw medical image data, wherein the means for generating one or more medical images from the raw medical image data by applying the selected set of pre-processing algorithms to the raw medical image data comprises:

means for generating one or more medical images for human visual interpretation by applying the first set of one or more pre-processing algorithms to the raw medical image data, and

means for generating one or more medical images for automated image analysis by applying the second set of pre-processing algorithms to the raw medical image data.

8. The apparatus of claim 7 , further comprising:

means for displaying the one or more medical images for human visual interpretation generated using the first set of pre-processing algorithms on a display device; and

means for providing the one or more medical images for automated image analysis generating using the second set of pre-processing algorithms as input to one or more of the available downstream automated image analysis algorithms.

9. The apparatus of claim 7 , wherein the second set of pre-processing algorithms includes one or more pre-processing algorithms corresponding to each of a plurality of available downstream automated image analysis algorithms predicted to be applied to the medical image scan, and the means for generating one or more medical images for automated image analysis by applying the second set of pre-processing algorithms to the raw medical image data comprises:

means for generating a respective medical image suited for each of the plurality available downstream automated image analysis algorithms predicted to be applied to the medical image scan using the corresponding one or more pre-processing algorithms in the second set of pre-processing algorithms.

10. The apparatus of claim 7 , wherein the trained machine learning based model is trained based on a training database of training images acquired using a variety of pre-processing algorithms by comparing output interpretations of the training images determined using one or more available downstream automated image processing algorithms with ground truth interpretations of the training images.

11. A non-transitory computer readable medium storing computer program instructions for medical image acquisition and pre-processing, the computer program instructions when executed by a processor cause the processor to perform operations comprising:

acquiring raw medical image data by controlling a medical image scanner to perform a medical image scan of a patient;

acquiring input data associated with the medical image scan of the patient and available downstream automated image analysis algorithms;

selecting a set of pre-processing algorithms to apply to the raw medical image data based on the input data associated with the medical image scan of the patient and the available downstream automated image analysis algorithms using a trained machine learning based model, wherein selecting a set of pre-processing algorithms to apply to the raw medical image data based on the input data associated with the medical image scan of the patient and the available downstream automated image analysis algorithms using a trained machine learning based model comprises:

selecting a first set of one or more pre-processing algorithms for generating one or medical images for human visual interpretation, and

selecting a second set of one or more pre-processing algorithms for generating one or more medical images for automated image analysis using one or more of the available downstream automated image analysis algorithms; and

generating one or more medical images from the raw medical image data by applying the selected set of pre-processing algorithms to the raw medical image data, wherein generating one or more medical images from the raw medical image data by applying the selected set of pre-processing algorithms to the raw medical image data comprises:

generating one or more medical images for human visual interpretation by applying the first set of one or more pre-processing algorithms to the raw medical image data, and

generating one or more medical images for automated image analysis by applying the second set of pre-processing algorithms to the raw medical image data.

12. The non-transitory computer readable medium of claim 11 , further comprising:

displaying the one or more medical images for human visual interpretation generated using the first set of pre-processing algorithms on a display device; and

providing the one or more medical images for automated image analysis generating using the second set of pre-processing algorithms as input to one or more of the available downstream automated image analysis algorithms.

13. The non-transitory computer readable medium of claim 11 , wherein the second set of pre-processing algorithms includes one or more pre-processing algorithms corresponding to each of a plurality of available downstream automated image analysis algorithms predicted to be applied to the medical image scan, and generating one or more medical images for automated image analysis by applying the second set of pre-processing algorithms to the raw medical image data comprises:

generating a respective medical image suited for each of the plurality available downstream automated image analysis algorithms predicted to be applied to the medical image scan using the corresponding one or more pre-processing algorithms in the second set of pre-processing algorithms.

14. The non-transitory computer readable medium of claim 11 , wherein the trained machine learning based model is trained based on a training database of training images acquired using a variety of pre-processing algorithms by comparing output interpretations of the training images determined using one or more available downstream automated image processing algorithms with ground truth interpretations of the training images.