TY - GEN
T1 - Fractional-order dynamic errors for analyzing residual arteriovenous access stenosis
AU - Chen, Wei Ling
AU - Mai, Yi Chen
AU - Lin, Chia Hung
AU - Kan, Chung Dann
PY - 2016/7/2
Y1 - 2016/7/2
N2 - In clinical examinations, acoustic methods are non-invasive and inexpensive diagnostic tools for the assessment of aortic and vascular stenosis or occlusion. To ensure the early detection of a dysfunctional arteriovenous access (AVA), we developed a monitoring interface by using phonoangiography (PCG) techniques to examine stenosis among hemodialysis patients. AVAs are vital lines for hemodialysis patients. Clinically, when the AVA lumen is reduced to less than 50% of normal lumen, percutaneous transluminal angioplasty or surgical intervention must be performed to restore normal AVA function. A method based on the Burg AR model combined with the fractional-order dynamic error (FODE) was proposed for evaluating the relationship between the power spectral density and vascular access DOS. However, in a clinical trial of 42 patients, the coefficients of determination was 0.3842 at the V-site anastomosis. For this project, the biophysical model describes an appropriate AVG model for describing the pumping action of the heart, and presents the experimental results. In addition, an arteriovenous graft biophysical model was created for comparing the spectral differences among the different stenotic ratio situations and elucidating the effects of other interference factors.
AB - In clinical examinations, acoustic methods are non-invasive and inexpensive diagnostic tools for the assessment of aortic and vascular stenosis or occlusion. To ensure the early detection of a dysfunctional arteriovenous access (AVA), we developed a monitoring interface by using phonoangiography (PCG) techniques to examine stenosis among hemodialysis patients. AVAs are vital lines for hemodialysis patients. Clinically, when the AVA lumen is reduced to less than 50% of normal lumen, percutaneous transluminal angioplasty or surgical intervention must be performed to restore normal AVA function. A method based on the Burg AR model combined with the fractional-order dynamic error (FODE) was proposed for evaluating the relationship between the power spectral density and vascular access DOS. However, in a clinical trial of 42 patients, the coefficients of determination was 0.3842 at the V-site anastomosis. For this project, the biophysical model describes an appropriate AVG model for describing the pumping action of the heart, and presents the experimental results. In addition, an arteriovenous graft biophysical model was created for comparing the spectral differences among the different stenotic ratio situations and elucidating the effects of other interference factors.
UR - http://www.scopus.com/inward/record.url?scp=85021060354&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85021060354&partnerID=8YFLogxK
U2 - 10.1109/ICMLC.2016.7860914
DO - 10.1109/ICMLC.2016.7860914
M3 - Conference contribution
T3 - Proceedings - International Conference on Machine Learning and Cybernetics
SP - 280
EP - 284
BT - Proceedings of 2016 International Conference on Machine Learning and Cybernetics, ICMLC 2016
PB - IEEE Computer Society
T2 - 2016 International Conference on Machine Learning and Cybernetics, ICMLC 2016
Y2 - 10 July 2016 through 13 July 2016
ER -