translate

gulpfile.js

Displaying 361 - 380 of 1717

Addressing Age-Related Pen-Based Target AcquisitionDifficulties

K. Moffatt

Pen and mouse based interactions were collected for investigating differences in target acquisition across the a range of ages. Dataset is collected from young pre-old and older adults. They undergo a variety of pen and mouse based target acquisition tasks in clinical settings.
Tags: 
Older adultsAgeing

Detecting Motor Impairment in Early Parkinson’s Disease via Natural Typing Interaction With Keyboards: Validation of the neuroQWERTY Approach in an Uncontrolled At-Home Setting

M. Gray, L. Giancardo, Á. Sánchez-Ferro

A dataset of keypress timings for detecting Parkinson's disease is collected. Key press times were recorded while participants conducted a controlled typing task at their home.
Tags: 
Parkinson's diseaseMotor impairment

WebinSitu: a comparative analysis of blind and sighted browsing behavior

Jeffrey P. Bigham, Anna C. Cavender, Jeremy T. Brudvik, Jacob O. Wobbrock, Richard E. Ladner

This work analyzes and compares browsing behaviours between blind and sighted participants. An HTTP proxy connection monitored and collected all the user interactions with webpages for a week.
Tags: 
blindness

Evaluation of Supervised Learning Algorithms Based on Speech Features as Predictors to the Diagnosis of Mild to Moderate Intellectual Disability

G. Aggarwal, L. Singh

Speech features are collected as an early indicator of the disorder which can be used to train machine learning algorithms for differentiating between speech of normally developing children and children with intellectual disability. 2 tasks related to sustained phonation and repetition were given. Speech recordings were done in a quiet room using head fitted recorder (Sony UX533F)
Tags: 
Speech features in intellectual disabilityIntellectual and Developmental Disability

Accessible Texts for Autism: An Eye-Tracking Study

V. Yaneva, I. Temnikova, R. Mitkov

This work uses eye tracking technology with ASD participants in order to evaluate text documents. Particpants were asked to read texts with photographs (20 photographs in total), texts with symbols (19 symbols in total) and 9 plain texts (with no images) while their eye sight was tracked and were asked a multiple choice question after each folowed by questions on test diffuculty and reading preference.
Tags: 
accessibilityReading

MCMC Bayesian inference for heart sounds screening in assistive environments

M. Maragoudakis, E. Loukis

A method for automatic screening of heart sound signals, for diagnostic purposes in both the traditional medicine and the emerging ICT-based assistive environments. Signals were acquired through diffrent types of stethoscopes with varying sensors, use mode and sound filters.
Tags: 
Cardiac diseases

Detecting readers with dyslexia using machine learning with eye tracking measures

L. Rello, M. Ballesteros

A dataset of eyetracking movements of people with dyslexia and normal vision while reading different combinations of texts and typefaces are collected for comparison. Each participant read 12 different texts with 12 different typefaces. Participants were were between 11 and 54 years of age.
Tags: 
dyslexia

jquery.herotabs

by simonsmith

A tiny, fully accessible tab switcher for jQuery.

Multimodal assessment of Parkinson’s disease: a deep learning approach

J. C. Vasquez-Correa, T. Arias-Vergara, J. R. Orozco-Arroyave, B. Eskofier, J. Klucken, E. Noth

This work aims to model difficulties in starting or to stoping movements in speech, handwriting, and gait and use those transitions to train convolutional neural networks to classify patients and healthy subjects. The subjects performed a total of 14 tasks divided into writing and drawing tasks on a tablet that captured six different signals: x-position, yposition, in-air movement, azimuth, altitude, and pressure.
Tags: 
Parkinson's disease

DEVISIGN-G

Y. Li, C. Juan

A dataset of 3D depth video and skeleton data is collected for Chinese sign language. 4 signers performed 26 alphabets and 10 number signs twice each and 4 signers performed them once, this was recorded using a 3D Kinect sensor.
Tags: 
Sign languageChinese Sign Language

Spanish PerLA

J. Mantero, B. Gallardo-Pauls

People with Alzeimer's disease speakout narrative phrases to build database of speech data for understanding patterns in their articulation. Transcripts and audio are collected as speakers read out text in Spanish.
Tags: 
DementiaAlzheimer's disease

The mPower study, Parkinson disease mobile data collected using ResearchKit

Stephen H. Friend, Andrew D. Trister

A mobile app to collect various test like, tapping acitvity, voice samples, memory activity and more is collected from people with and without self-identified Parkinson's disease. Participants used the smartphone app to record and register 4 activities, tapping, memory test, walking and voice samples were recorded
Tags: 
Parkinson's disease

Chinese Sign Language Recognition Dataset

J. Huang, W. Zhou, Q. Zhang, H. Li, W. Li

This dataset is introduced for continuous sign language recognition with sentence-level annotations. A Microsoft Kinect camera is used for all recording, providing RGB, depth and body joints modalities in all videos.
Tags: 
Sign languageChinese Sign Language

Ulm Corpus

A. Haege

This work is part of a longitudinal study that aims to undertand the relations with different abilities and the development of stuttering in childhood. Preschool and primary school children were involved in an experimental playing situation where the children play with their mothers in the phoniatric clinic laboratory.
Tags: 
StutteringEarly Childhood StutteringFluency disordersStuttering-like disfluencies

Vanryckeghem Corpus

M. Vanryckeghem

This work aims to differentially diagnose stuttering and cluttering symptoms of Dutch speaking children and adolescents with Down syndrome. The data is used to analyze the dysfluencies.
Tags: 
Down's SyndromeStutteringCluttering

Population-scale hand tremor analysis via anonymized mouse cursor signals

R. White, E. Horvitz

Cursor logs from an anonymized sample of millions of searchers on the Microsoft Bing search engine are collected and analyzed. They also conduct a 1000 participant study to confirm the validity of the population scale results. Microsoft bing search queries are used as proxies and the cursor movements are recorded.
Tags: 
Parkinson's diseaseEssential tremor

Characterisation of mental health conditions in social media using Informed Deep Learning

G. Gkotsis, A. Oellrich, S. Velupillai, M. Liakata, Tim J. P. Hubbard, Richard J. B. Dobson , R. Dutta

This work aims to classify reddits posts for a probable mental illness that was being discussed in the post. Subreddits were scrapped.
Tags: 
Mental illnessMental health

English Pitt

F. Boller, J. Becker

These transcripts and audio files were collected as part of a larger protocol of the Alzheimer and Related Dementias Study. Patients were given 4 tasks and there were then recorded semi-structured interviews examining the continued use of preferred External Memory Aids (EMAs) and experiences of individuals with mild neurocognitive disorder (mNCD) who completed a group intervention 1.5 years prior to the interviews.
Tags: 
DementiaAlzheimer's disease

Activity detection and recognition of daily living events

K. Avgerinakis, A. Briassouli, I. Kompatsiaris

Video activity data from people with dementia, people with mild cognitive impairment, and healthy ones are collected to develop activity detection and recognition systems. Participants are asked to perform a set of activities like serving beverage in a home-like environment.
Tags: 
DementiaActivities of Daily Living

The SIGNOR Corpus of Slovene Sign Language

Š. Vintar, B. Jerko, M. Kulovec

The work aims to compile a representative corpus of SSL and then transcribe and annotate the video data to get the information needed for describing the lexicon and grammar of the language. Data from 72 informants have been collected from different Slovenian regions and thorugh the support of deaf clubs. Recording sessions include daily life conversations and video elicited conversations.
Tags: 
Sign languageSlovene Sign Language