Method and system for data extraction from echocardiographic strain graphs
A system to analyze heart data includes a memory configured to store echocardiographic strain image data. The system also includes a processor operatively coupled to the memory and configured to segment the echocardiographic strain image data. The processor is also configured to analyze the segmented echocardiograph strain image data to generate a point cluster of pixels. The processor is also configured to stitch together points in the point cluster to generate a continuous curve that represents the echocardiographic strain image data. The processor is further configured to analyze the continuous curve to identify a heart condition.
1. A system to analyze heart data, the system comprising:
a memory configured to store echocardiographic strain image data; and
a processor operatively coupled to the memory and configured to:
segment the echocardiographic strain image data, wherein segmentation of the echocardiographic strain image data extracts a scale and graph data from the echocardiographic strain image data, wherein the processor is configured to convert the graph data into a two-dimensional array of hue saturation values;
analyze the segmented echocardiograph strain image data to generate a point cluster of pixels;
stitch together points in the point cluster to generate a continuous curve that represents at least a portion of the echocardiographic strain image data; and
analyze the continuous curve to identify a heart condition.
2. The system of claim 1 , wherein the echocardiographic strain image data includes one or more visual graphs depicting a plurality of apical echo views of a heart.
3. The system of claim 2 , wherein each of the apical echo views of the heart contain a plurality of curves that are coded to specify locations within the heart.
4. The system of claim 1 , wherein the continuous curve is formed by use of an interpolated univariate spline on the point cluster.
5. The system of claim 1 , wherein the processor is configured to identify all pixels in the graph data that represent a specific location of the heart.
6. The system of claim 5 , wherein the processor is configured to convert the identified pixels into the point cluster of pixels.
7. The system of claim 6 , wherein the point cluster is based on row and column indices of the identified pixels.
8. The system of claim 1 , wherein the processor analyzes the continuous curve to identify one or more patterns present in the echocardiographic strain image data.
9. The system of claim 8 , wherein the processor compares the one or more patterns to previously analyzed strain data to identify the heart condition.
10. The system of claim 8 , wherein the one or more patterns relate to a timing of peak strain in the echocardiograph strain image data.
11. The system of claim 8 , wherein the one or more patterns relate to early systolic lengthening and post systolic shortening.
12. The system of claim 8 , wherein the one or more patterns relate to myocardial stretch with atrial contraction.
13. A method of analyzing heart data, the method comprising:
storing, in a memory, echocardiographic strain image data; and
segmenting, by a processor operatively coupled to the memory, the echocardiographic strain image data, wherein the segmenting includes segmenting the echocardiographic strain image data into a scale and graph data, and further comprising converting, by the processor, the graph data into a two-dimensional array of hue saturation values;
analyzing the segmented echocardiograph strain image data to generate a point cluster of pixels;
stitching together, by the processor, points in the point cluster to generate a continuous curve that represents at least a portion of the echocardiographic strain image data; and analyzing, by the processor, the continuous curve to identify a heart condition.
14. The method of claim 13 , further comprising identifying, by the processor, all pixels in the graph data that represent a specific location of the heart, and converting the identified pixels into the point cluster of pixels.
15. The method of claim 13 , further comprising analyzing, by the processor, the continuous curve to identify one or more patterns present in the echocardiographic strain image data.
16. The method of claim 15 , further comprising comparing, by the processor, the one or more patterns to previously analyzed strain data to identify the heart condition.