CARDIAC FLOW DETECTION BASED ON MORPHOLOGICAL MODELING IN MEDICAL DIAGNOSTIC ULTRASOUND IMAGING
For cardiac flow detection in echocardiography, by detecting one or more valves, sampling planes or flow regions spaced from the valve and/or based on multiple valves are identified. A confidence of the detection may be used to indicate confidence of calculated quantities and/or to place the sampling planes.
1 . A method for detecting cardiac flow in echocardiography, the method comprising:
detecting, by an image processor, two or more heart valves over time from B-mode data of one or more volumetric ultrasound scans;
determining, by the image processor, a confidence for the detecting of the two or more heart valves;
placing, by the image processor, a measurement area surface over time in a cardiac flow region based on the detected heart valves;
calculating, by the image processor, a cardiac flow value from flow data of the volumetric ultrasound scans for the measurement area surface over time, the calculating of the cardiac flow value limited to avoid flow during a portion of a heart cycle; and
outputting an image of the cardiac flow value
wherein the image indicates the confidence and/or the measurement area surface is placed based on the confidence and the detected heart valves over time.
2 . The method of claim 1 wherein detecting the two or more valves comprises detecting a location, orientation, and scale of each of the two or more valves represented by the B-mode data.
3 . The method of claim 1 wherein detecting comprises fitting morphological models of the two or more valves to the B-mode data.
4 . The method of claim 1 wherein detecting comprises detecting with a machine-learnt classifier, and wherein determining the confidence comprises determining the confidence with the machine-learnt classifier.
5 . The method of claim 1 wherein placing comprises placing the measurement surface area in the cardiac flow region spaced from the two or more valves.
6 . The method of claim 5 wherein the two valves comprise a mitral valve and an aortic valve, and wherein placing comprises placing the measurement surface area at a left ventricle outflow tract.
7 . The method of claim 6 wherein placing comprises placing the measurement surface area to avoid quantifying flow from other valves, the placing comprising determining intersection with valve models.
8 . The method of claim 1 wherein placing comprises placing based on the detecting of the two or more valves at one time and tracking the measurement area surface for other times.
9 . The method of claim 1 wherein calculating comprises calculating limited to avoid regurgitant flow.
10 . The method of claim 1 wherein calculating comprises calculating the cardiac flow value for in-flow, out-flow, and/or stroke volume.
11 . The method of claim 1 wherein outputting comprises outputting the image of models of the detected two or more valves and a quantity for the cardiac flow value.
12 . The method of claim 1 wherein outputting comprises outputting with the image including an indication of the confidence.
13 . The method of claim 1 wherein placing comprises placing based on the confidence.
14 . The method of claim 1 further comprising refining the placement of the measurement area surface based on aliased flow, relative flow, maximum flow, and/or a smoothness of outflow and/or inflow as a function of time.
15 . A system for detecting cardiac flow, the system comprising:
an ultrasound scanner configured to scan a heart volume of a patient, the scan providing B-mode and Doppler flow data;
an image processor configured to fit a model of a heart valve over a heart cycle to the B-mode data with a machine-learnt classifier, to use the model to locate a cardiac flow area, and to calculate the cardiac flow from the Doppler flow data for the cardiac flow area; and
a display configured to generate a visualization of the model over time as fit to the B-mode data, highlight the cardiac flow area, and indicate a value of the calculated cardiac flow;
wherein the location of the cardiac flow area is based, in part, on a confidence of the fit output by the machine-learnt classifier and/or wherein the display is configured to indicate a value of the confidence.
16 . The system of claim 15 wherein the location of the cardiac flow is based, in part, on the confidence.
17 . The system of claim 15 wherein the display is configured to indicate the value of the confidence.
18 . The system of claim 15 wherein the image processor is configured to limit the calculation of the cardiac flow to a portion of the heart cycle based on timing from the fit model.
19 . The system of claim 15 wherein the image processor is configured to fit the model of the heart valve and another model for another heart valve and is configured to locate the cardiac flow area to flow in a non-valvular region based on positions of the heart valve and the other heart valve.
20 . A system for detecting cardiac flow, the system comprising:
an ultrasound scanner configured to scan a heart volume of a patient over at least a heart cycle, the scan providing B-mode and Doppler flow data;
an image processor configured to fit first and second models of first and second heart valves over the heart cycle to the B-mode data, to use the first and second models to locate a cardiac flow region in a non-valvular region, and to calculate the cardiac flow from the Doppler flow data for the cardiac flow region; and
a display configured to generate a visualization of the model over time as fit to the B-mode data, highlight the cardiac flow area, and indicate a value of the calculated cardiac flow.