Automatic slice selection in medical imaging

A method for operating a medical imaging system, wherein the method comprises

• receiving three-dimensional medical image data comprising multiple slices;
• receiving an imaging modality of the three-dimensional medical image data;
• receiving an anatomical view classification of the three-dimensional medical image data;
• selecting a chosen abnormality detection module from a set of abnormality detection modules using the imaging modality and the anatomical view classification, wherein each of the abnormality detection modules is a neural network trained for identifying at least a portion of the multiple slices as either normal or abnormal;
• classifying the at least a portion of the multiple slices as normal or abnormal using the abnormality detection module;
• and
• choosing a set of selected slices from the multiple slices according to a predetermined selection criteria if a predetermined number of the multiple slices are classified as abnormal.

A computer program product comprising machine executable instructions for execution by a processor controlling a medical imaging system, wherein execution of the machine executable instructions causes the processor to:

• receive three-dimensional medical image data comprising multiple slices;
• receive an imaging modality of the three-dimensional medical image data;
• receive an anatomical view classification of the three-dimensional medical image data;
• select a chosen abnormality detection module from a set of abnormality detection modules using the imaging modality and the anatomical view classification for identifying at least a portion of the multiple slices as either normal or abnormal;
• classify the at least a portion of the multiple slices as normal or abnormal using the abnormality detection module;
• and
• choose a set of selected slices from the multiple slices according to a predetermined selection criteria if a predetermined number of the multiple slices are classified as abnormal,
• wherein the instructions cause the processor to provide suggestions for other clinical findings based on the selected slices and prior history of a same patient.