Deep Learning for Classification of Normal Swallows in Adults.

Cervical auscultation is a method for assessing swallowing performance. However, its ability to serve as a classification tool for a practical clinical assessment method is not fully understood. In this study, we utilized neural network classification methods in the form of Deep Belief networks in order to classify swallows. We specifically utilized swallows that did not result in clinically significant aspiration and classified them on whether they originated from healthy subjects or unhealthy patients. Dual-axis swallowing vibrations from 1946 discrete swallows were recorded from 55 healthy and 53 unhealthy subjects. The Fourier transforms of both signals were used as inputs to the networks of various sizes. We found that single and multi-layer Deep Belief networks perform nearly identically when analyzing only a single vibration signal. However, multi-layered Deep Belief networks demonstrated approximately a 5% to 10% greater accuracy and sensitivity when both signals were analyzed concurrently, indicating that higher-order relationships between these vibrations are important for classification and assessment.

A Matched Dual-Tree Wavelet Denoising for Tri-Axial Swallowing Vibrations.

Swallowing disorders affect thousands of patients every year. Currently utilized techniques to screen for this condition are questionably reliable and are often deployed in non-standard manners, so efforts have been put forth to generate an instrumental alternative based on cervical auscultation. These physiological signals with low signal-to-noise ratios are traditionally denoised by well-known wavelets in a discrete, single tree wavelet decomposition. We attempt to improve this widely accepted method by designing a matched wavelet for cervical auscultation signals to provide better denoising capabilities and by implementing a dual-tree complex wavelet transform to maintain time invariant properties of this filtering. We found that our matched wavelet did offer better denoising capabilities for cervical auscultation signals compared to several popular wavelets and that the dual tree complex wavelet transform did offer better time invariance when compared to the single tree structure. We conclude that this new method of denoising cervical auscultation signals could benefit applications that can spare the required computation time and complexity.

Speech signal modification to increase intelligibility in noisy environments.

The role of transient speech components on speech intelligibility was investigated. Speech was decomposed into two components--quasi-steady-state (QSS) and transient--using a set of time-varying filters whose center frequencies and bandwidths were controlled to identify the strongest formant components in speech. The relative energy and intelligibility of the QSS and transient components were compared to original speech. Most of the speech energy was in the QSS component, but this component had low intelligibility. The transient component had much lower energy but was almost as intelligible as the original speech, suggesting that the transient component included speech elements important to speech perception. A modified version of speech was produced by amplifying the transient component and recombining it with the original speech. The intelligibility of the modified speech in background noise was compared to that of the original speech, using a psychoacoustic procedure based on the modified rhyme protocol. Word recognition rates for the modified speech were significantly higher at low signal-to-noise ratios (SNRs), with minimal effect on intelligibility at higher SNRs. These results suggest that amplification of transient information may improve the intelligibility of speech in noise and that this improvement is more effective in severe noise conditions.