259 294

Cited 2 times in

Feasibility of Deep Learning-Based Analysis of Auscultation for Screening Significant Stenosis of Native Arteriovenous Fistula for Hemodialysis Requiring Angioplasty

Authors
 Jae Hyon Park  ;  Insun Park  ;  Kichang Han  ;  Jongjin Yoon  ;  Yongsik Sim  ;  Soo Jin Kim  ;  Jong Yun Won  ;  Shina Lee  ;  Joon Ho Kwon  ;  Sungmo Moon  ;  Gyoung Min Kim  ;  Man-Deuk Kim 
Citation
 KOREAN JOURNAL OF RADIOLOGY, Vol.23(10) : 949-958, 2022-10 
Journal Title
KOREAN JOURNAL OF RADIOLOGY
ISSN
 1229-6929 
Issue Date
2022-10
MeSH
Angioplasty ; Arteriovenous Fistula* ; Auscultation ; Constriction, Pathologic ; Deep Learning* ; Feasibility Studies ; Female ; Humans ; Male ; Middle Aged ; Renal Dialysis
Keywords
Angioplasty ; Arteriovenous fistula ; Auscultation ; Deep learning ; Renal dialysis
Abstract
Objective: To investigate the feasibility of using a deep learning-based analysis of auscultation data to predict significant stenosis of arteriovenous fistulas (AVF) in patients undergoing hemodialysis requiring percutaneous transluminal angioplasty (PTA).

Materials and methods: Forty patients (24 male and 16 female; median age, 62.5 years) with dysfunctional native AVF were prospectively recruited. Digital sounds from the AVF shunt were recorded using a wireless electronic stethoscope before (pre-PTA) and after PTA (post-PTA), and the audio files were subsequently converted to mel spectrograms, which were used to construct various deep convolutional neural network (DCNN) models (DenseNet201, EfficientNetB5, and ResNet50). The performance of these models for diagnosing ≥ 50% AVF stenosis was assessed and compared. The ground truth for the presence of ≥ 50% AVF stenosis was obtained using digital subtraction angiography. Gradient-weighted class activation mapping (Grad-CAM) was used to produce visual explanations for DCNN model decisions.

Results: Eighty audio files were obtained from the 40 recruited patients and pooled for the study. Mel spectrograms of "pre-PTA" shunt sounds showed patterns corresponding to abnormal high-pitched bruits with systolic accentuation observed in patients with stenotic AVF. The ResNet50 and EfficientNetB5 models yielded an area under the receiver operating characteristic curve of 0.99 and 0.98, respectively, at optimized epochs for predicting ≥ 50% AVF stenosis. However, Grad-CAM heatmaps revealed that only ResNet50 highlighted areas relevant to AVF stenosis in the mel spectrogram.

Conclusion: Mel spectrogram-based DCNN models, particularly ResNet50, successfully predicted the presence of significant AVF stenosis requiring PTA in this feasibility study and may potentially be used in AVF surveillance.
Files in This Item:
T202204070.pdf Download
DOI
10.3348/kjr.2022.0364
Appears in Collections:
1. College of Medicine (의과대학) > Dept. of Radiology (영상의학교실) > 1. Journal Papers
1. College of Medicine (의과대학) > Dept. of Surgery (외과학교실) > 1. Journal Papers
Yonsei Authors
Kwon, Joon Ho(권준호) ORCID logo https://orcid.org/0000-0002-6178-7252
Kim, Gyoung Min(김경민) ORCID logo https://orcid.org/0000-0001-6768-4396
Kim, Man Deuk(김만득) ORCID logo https://orcid.org/0000-0002-3575-5847
Kim, Soo Jin(김수진)
Moon, Sungmo(문성모)
Sim, Yongsik(심용식) ORCID logo https://orcid.org/0000-0003-2711-2793
Won, Jong Yun(원종윤) ORCID logo https://orcid.org/0000-0002-8237-5628
Han, Ki Chang(한기창) ORCID logo https://orcid.org/0000-0002-9701-9757
URI
https://ir.ymlib.yonsei.ac.kr/handle/22282913/192240
사서에게 알리기
  feedback

qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse

Links