translate

Cognitive, Language, Learning Disabilities & Low Literacy

A Smartphone-Based Tool for AssessingParkinsonian Hand Tremor

N. Kostikis

Smartphone sensors are used to classify tremors in Parkinson's patients corellated with the UPDRS scale. Participants wore smartphone based gloves and data during tremors was collected and features like accelertation were computed after the data was collected.
Tags: 
Parkinson's disease

The TYPALOC Corpus: A Collection of Various Dysarthric Speech Recordings in Read and Spontaneous Styles

Meunier C., Fougeron C., Fredouille C., Bigi B., Crevier-Buchman L., Delais-Roussarie E., Georgeton L., Ghio A., Laaridh I., Legou T., Pillot-Loiseau C., Pouchoulin G.

Dysarthic speakers pronounced various sentences and a corpus was collected to understand the phonetic variations between them and healthy controls(without Dysarthria) Audio was collected, transcribed and automatically aligned with the phonemes pronounced.
Tags: 
Parkinson's diseaseDysarthriaAmyotrophic Lateral SclerosisCerebellar Alteration

Automatic Identification of Mild Cognitive Impairment through the Analysis of Italian Spontaneous Speech Productions

D. Beltrami, L. Calzà, G. Gagliardi, E. Ghidoni, N. Marcello, R. Favretti, F. Tamburin

Liguistic symptoms of cognitive decline in elderly patients are investigated through speech. Participants read and answered three tasks, these utterances were labeled for features and transcribed.
Tags: 
Early DementiaDementiaMild Cognitive Impairment

Monitoring Parkinson's Disease Progression Using Behavioural Inferences, Mobile Devices and Web Technologies

J. Vega

A dataset using smartphone usage is collected from patients with Parkinson's disease to monitor the progression of the disease. Smartphone usage data from 2 patients with Parkinson's disease is collected, along with sensor data such as location and weather.
Tags: 
Parkinson's disease

Automatic Readability Assessment

L. Feng

A text corpora annotated with indicators of reading difficulty is collected to train a Machine Learning algorithm for predicting readign difficulty for a piece of text. English documents on various documents were collected
Tags: 
Mild Intellectual Disability

Automatic childhood autism detection by vocalization decomposition with phone-like units

D. Xu, Jeffrey A. Richards, J. Gilkerson, U. Yapanel, S. Gray, J. Hansen

To create a mechanism for child autism detection using the LETM (Language ENvironment Analysis) System, utilizing speech signal processing technology Recorded through a small digital recorder (DLP – digital language processor) worn by the child in the pocket of specially designed clothing.
Tags: 
Autism detectionAutismAutism Spectrum DisordersSpeech Analysis

Automatic Emotion Recognition in Robot-Children Interaction for ASD Treatment

M. Leo, M. Coco, P. Carcagni, C. Distante, M. Bernava, G. Pioggia, G. Palestra

This work analyzes facial expression to evaluate robot-ASD children interactions. Children were asked to imitate expression of robots and pictures were taken, and children's behaviours were analyzed using machine-learning strategies.
Tags: 
EmotionsEmotion RecognitionAutism Spectrum Disorders

Imprecise vowel articulation as a potential early marker of Parkinson's disease: effect of speaking task.

J. Rusz, R. Cmejla, T. Tykalova

This work aims to analyze vowel articulation in individuals with Parkinson's disease. Data on speaking tasks such as phonation, sentence, passage and monologues are collected.
Tags: 
Parkinson's diseaseDysarthria

Intonation and speech rate in Parkinson's disease: general and dynamic aspects and responsiveness to levodopa admission.

S. Skodda, W. Grönheit, U. Schlegel

This work aims to analyze fundamental frequency (F(0)) variability (fundamental frequency standard deviation [F(0)SD]) and net speech rate (NSR) in the course of reading in Parkinsonian patients' speech. German speaking patients with Parkinson's disease went through a reading task, and they were recorded.
Tags: 
Parkinson's disease

Unbiased and Mobile Gait Analysis Detects Motor Impairment in Parkinson's Disease

J. Klucken, J. Barth

A machine learning classifier to automatically detect motor impariment in patients with Parkinson's disease is constructed. To train the classifier a dataset of gait movements measured by acclerometer is collected. A mobile biosensor attached to the foot, eGaIT, is used to collect accelerometer, gyroscopes data during standarized gait and leg functions. Two participant groups are used to collect independent training and validation datasets.
Tags: 
Parkinson's disease