Hearing Loss via Wavelet Entropy and Particle Swarm Optimized Trained Support Vector Machine

FANG-ZHOU BAO, KOJI NAKAMURA

Abstract


This paper proposed a method that combines wavelet entropy with particle swarm optimized support vector machine to detect hearing loss (HL). The dataset for this task contains 75 images: 25 healthy controls, 25 left-sided hearing loss patients, and 25 right-sided hearing loss patients. The results shows that our method has higher accuracy than some traditional method. The sensitivities over healthy control, left-sided HL patients, and right-sided HL patients are 85.20± 3.79, 85.20± 4.64, and, 86.40± 5.06, respectively. The overall accuracy is 85.60± 0.84.

Keywords


Hearing loss, Wavelet entropy, Particle swarm optimization, Support vector machine.Text


DOI
10.12783/dtetr/ecae2018/27724

Full Text:

PDF

Refbacks

  • There are currently no refbacks.