On the Selection of Non-Invasive Methods Based on Speech Analysis Oriented to Automatic Alzheimer Disease Diagnosis

Abstract

The work presented here is part of a larger study to identify novel technologies and biomarkers for early Alzheimer disease (AD) detection and it focuses on evaluating the suitability of a new approach for early AD diagnosis by non-invasive methods. The purpose is to examine in a pilot study the potential of applying intelligent algorithms to speech features obtained from suspected patients in order to contribute to the improvement of diagnosis of AD and its degree of severity. In this sense, Artificial Neural Networks (ANN) have been used for the automatic classification of the two classes (AD and control subjects). Two human issues have been analyzed for feature selection: Spontaneous Speech and Emotional Response. Not only linear features but also non-linear ones, such as Fractal Dimension, have been explored. The approach is non invasive, low cost and without any side effects. Obtained experimental results were very satisfactory and promising for early diagnosis and classification of AD patients.

On the early diagnosis of Alzheimer's Disease from multimodal signals: A survey

“…The speaking and conversational skills of the AD patients deteriorate from the early stages of the disease [40]. They are likely to lose vocabulary, make big pauses while they are speaking or just stop abruptly because they are not able to continue the conversation.…”

“…The clinical hallmark and earliest manifestation of AD is episodic memory impairment [40]. Not remembering recently learnt information is the most common symptom, which is discernible from the early stages of the disease.…”

“…These problems are present from the very early stages of the disease, and progress and worsen at the same time that the cognitive decline [276,283]. Due to the ease with which voice recordings can be obtained, in the recent years, speech features have been used in works related to automatic MCI and AD diagnosis systems [27,40,[284][285][286][287][288]. Nonetheless, few studies report speech based classification results that suggest that they have potential to become part of a multimodal diagnosis system.…”

Selection of Entropy Based Features for Automatic Analysis of Essential Tremor

“…The results were evaluated using Accuracy (Acc, in %) and Classification Error Rate (CER, in %) [7,[42][43][44]. For training and validation steps, we used k-fold cross-validation with k = 10, where accuracy will be the average of the k iterations.…”

A Preliminary Review of Behavioural Biometrics for Health Monitoring in the Elderly

“…By analysing continuous speech, the system calculates an index whose values show whether the subject can be classified as being affected by AD. Emotional Temperature (ET) is another parameter that, combined with other traditional speech parameters, can improve and facilitate the early diagnosis of AD (López-de-Ipiña et al, 2013). In López-de-Ipiña et al (2014), the Fractal Dimension (FD) of the speech signals is combined with linear parameters in the feature vector in order to enhance the performance of the original system while controlling the computational cost.…”