Scoring and ranking angiograms
Techniques for processing one or more frames of an angiogram are disclosed. The processing may take place during or after an angiography exam. The one or more frames of the angiogram are acquired during the angiography exam. The one or more frames are processed to determine, based on at least one pre-defined criterion, whether the angiogram at least comprises one frame with a diagnostic value among the one or more frames. If the angiogram comprises at least one frame with the diagnostic value, based on the angiogram, a score quantifying the diagnostic value of the angiogram is determined using a trained machine-learning (ML) algorithm. Techniques for processing, e.g., ranking/sorting, multiple angiograms associated with an anatomical region of interest of a patient are also provided, by which a respective score for each of the multiple angiograms is determined using the techniques for processing one or more frames of an angiogram.
1 . A computer-implemented method comprising:
obtaining one or more frames of an angiogram, which angiogram is acquired during an angiography exam of an anatomical region of interest;
based on at least one pre-defined criterion, determining, among the one or more frames, whether the angiogram at least comprises one frame with a diagnostic value, wherein the at least one pre-defined criterion comprises whether at least one stenosis in a segment of a blood vessel within the anatomical region of interest can be determined based on a respective frame of the angiogram, wherein when the at least one stenosis is determined based on at least one frame of the one or more frames of the angiogram, the angiogram is determined to comprise at least one frame with the diagnostic value; and
when the angiogram comprises at least one frame with the diagnostic value, determining, based on the angiogram, a score using a trained machine-learning algorithm, wherein the score quantifies the diagnostic value of the angiogram.
2 . The computer-implemented method of claim 1 , further comprising:
when the angiogram comprises no frame with the diagnostic value or if the score is below a pre-defined score threshold, stopping acquisition of the angiogram and adjusting one or more imaging parameters associated with the angiography exam, and acquiring one or more frames of a further angiogram based on the adjusted one or more imaging parameters.
3 . The computer-implemented method of claim 2 , further comprising:
discarding the angiogram.
4 . The computer-implemented method of claim 1 , further comprising:
when the score equals to or is above a pre-defined score threshold, stopping the angiography exam.
5 . The computer-implemented method of claim 1 ,
wherein a further trained ML algorithm is used to determine whether the at least one stenosis can be determined.
6 . The computer-implemented method of claim 1 ,
wherein the at least one pre-defined criterion further comprises whether a specific segment of a blood vessel within the anatomical region of interest can be determined based on a respective frame of the angiogram; and
when the specific segment of the blood vessel within the anatomical region of interest can be determined based on at least one frame of the one or more frames of the angiogram, the angiogram is determined to comprise at least one frame with the diagnostic value.
7 . The computer-implemented method of claim 6 ,
wherein the at least one pre-defined criterion further comprises whether the specific segment of the blood vessel comprises a contrast above a contrast threshold in at least one frame of the one or more frames of the angiogram;
when the angiogram comprises at least one frame of the one or more frames in which the contrast of the specific segment of the blood vessel is above the contrast threshold, the angiogram is determined to comprise at least one frame with the diagnostic value.
8 . The computer-implemented method of claim 1 ,
wherein the score comprises a probability that at least one abnormality can be diagnosed based on the angiogram.
9 . The computer-implemented method of claim 1 ,
wherein determining of the score is further based on at least one query associated with a diagnosis purpose.
10 . A computer-implemented method comprising:
obtaining multiple angiograms associated with an anatomical region of interest of a patient;
determining, based on at least one pre-defined criterion comprising whether at least one stenosis in a segment of a blood vessel within the anatomical region of interest can be determined based on a respective frame of the angiogram, whether the respective angiogram of the multiple angiograms comprises at least one frame with a diagnostic value, wherein when the at least one stenosis is determined based on at least one frame of the one or more frames of the angiogram, the respective angiogram is determined to comprise at least one frame with the diagnostic value;
determining, for each of the multiple angiograms that comprise at least one frame with diagnostic value, a score using a trained machine-learning algorithm, wherein the score quantifies the diagnostic value of the respective angiogram; and
sorting the multiple angiograms by score descending.
11 . The computer-implemented method of claim 10 , further comprising:
displaying and/or processing the multiple angiograms in an order associated with the sorting.
12 . The computer-implemented method of claim 10 , further comprising:
when one of the scores equals to or is above a pre-defined score threshold, stopping the angiography exam.
13 . The computer-implemented method of claim 10 ,
wherein the score comprises a probability that at least one abnormality can be diagnosed based on the angiogram.
14 . An angiography system comprising:
a memory configured to store program code; and
a processor configured to execute the program code, the processor being configured by the program code to:
obtain one or more frames of an angiogram, which angiogram is acquired during an angiography exam of an anatomical region of interest;
based on at least one pre-defined criterion, determine, among the one or more frames, whether the angiogram at least comprises one frame with a diagnostic value, wherein the at least one pre-defined criterion comprises whether a specific segment of a blood vessel within the anatomical region of interest can be determined based on a respective frame of the angiogram, wherein when the specific segment of the blood vessel within the anatomical region of interest can be determined based on at least one frame of the one or more frames of the angiogram, the angiogram is determined to comprise at least one frame with the diagnostic value; and
when the angiogram comprises at least one frame with the diagnostic value, determine, based on the angiogram, a score using a trained machine-learning algorithm, wherein the score quantifies the diagnostic value of the angiogram.
15 . The angiography system of claim 14 , wherein the processor is configured to, when one of the scores equals to or is above a pre-defined score threshold, stop the angiography exam.
16 . The angiography system of claim 14 , wherein the score comprises a probability that at least one abnormality can be diagnosed based on the angiogram.
17 . The angiography system of claim 14 , wherein the processor is configured to, when the angiogram comprises no frame with the diagnostic value or if the score is below a pre-defined score threshold, stop acquisition of the angiogram and adjust one or more imaging parameters associated with the angiography exam, and acquire one or more frames of a further angiogram based on the adjusted one or more imaging parameters.