شماره ركورد كنفرانس :
3297
عنوان مقاله :
Classification of Normal and Dysphagia in Patients with GERD Using Swallowing Sound Analysis
عنوان به زبان ديگر :
Classification of Normal and Dysphagia in Patients with GERD Using Swallowing Sound Analysis
پديدآورندگان :
Basiri Babak Department of Electrical Engineering of Faculty of Electrical and Computer Engineering - K. N. Toosi University of Technology - Tehran , Vali Mansour Department of Electrical Engineering of Faculty of Electrical and Computer Engineering - K. N. Toosi University of Technology - Tehran , Agah Shahram Department of Internal Medicine - School of Medicine - Iran University of Medical Sciences - Tehran
كليدواژه :
signal analysis , dysphagia , classification , swallowing sound
عنوان كنفرانس :
نوزدهمين سمپوزيوم بين المللي هوش مصنوعي و پردازش سيگنال
چكيده لاتين :
In recent years, acoustical analysis of the swallowing
mechanism has received considerable attention and because of
many damages of invasive methods, it is preferred. This paper
proposes acoustic-based method to separate dysphagia patients
with reflux disorder from normal persons. In this work, we have
used swallowing sound of 22 individuals (11 normal and 11
abnormal). Swallowing sound signals were recorded with sound
recorder over the trachea and ambient noise was removed and
spectral features were extracted from the sounds. Classification is
done by non-linear support vector machines, using leave-one-out.
According to the experimental results, the system can classify
66.1% of total swallow signals correctly (signal accuracy) and
95.7% of the total subject in a group of healthy and dysphagia
patients (subject accuracy). The experimental results show that
the proposed system can provide concrete features for clinicians
to diagnose dysphagia in reflux patients.
