Conference paper Open Access

Early Parkinson's Disease Detection via Touchscreen Typing Analysis using Convolutional Neural Networks

Dimitrios Iakovakis; Stelios Hadjidimitriou; Vasileios Charisis; Sevasti Bostanjopoulou; Zoe Katsarou; Lisa Klingelhoefer; Simone Mayer; Heinz Reichmann; Sofia B. Dias; José A. Diniz; Dhaval Trivedi; Ray K. Chaudhuri; Leontios J. Hadjileontiadis


Citation Style Language JSON Export

{
  "publisher": "Zenodo", 
  "DOI": "10.5281/zenodo.3675381", 
  "author": [
    {
      "family": "Dimitrios Iakovakis"
    }, 
    {
      "family": "Stelios Hadjidimitriou"
    }, 
    {
      "family": "Vasileios Charisis"
    }, 
    {
      "family": "Sevasti Bostanjopoulou"
    }, 
    {
      "family": "Zoe Katsarou"
    }, 
    {
      "family": "Lisa Klingelhoefer"
    }, 
    {
      "family": "Simone Mayer"
    }, 
    {
      "family": "Heinz Reichmann"
    }, 
    {
      "family": "Sofia B. Dias"
    }, 
    {
      "family": "Jos\u00e9 A. Diniz"
    }, 
    {
      "family": "Dhaval Trivedi"
    }, 
    {
      "family": "Ray K. Chaudhuri"
    }, 
    {
      "family": "Leontios J. Hadjileontiadis"
    }
  ], 
  "issued": {
    "date-parts": [
      [
        2019, 
        7, 
        28
      ]
    ]
  }, 
  "abstract": "<p>Parkinson&rsquo;s Disease (PD) is the second most&nbsp;common neurodegenerative disorder worldwide, causing both&nbsp;motor and non-motor&nbsp; symptoms. In the early stages, symptoms&nbsp;are mild and patients may ignore their existence. As a result,&nbsp;they do not undergo any related clinical examination; hence delaying their PD diagnosis. In an effort to remedy such delay,&nbsp;analysis of data passively&nbsp; captured from user&rsquo;s interaction with&nbsp;consumer technologies has been recently explored towards&nbsp;remote screening of early PD motor signs. In the current study,&nbsp;a smartphone-based method analyzing subjects&rsquo; finger&nbsp;interaction with the smartphone screen is developed for the&nbsp;quantification of fine-motor skills decline in early PD using&nbsp;Convolutional Neural Networks. Experimental results from the&nbsp;analysis of keystroke typing in-the-clinic data from 18 early PD&nbsp;patients and 15 healthy controls have shown a classification<br>\nperformance of 0.89 Area Under the Curve (AUC) with 0.79/0.79&nbsp;sensitivity/specificity, respectively. Evaluation of the&nbsp;generalization ability of the proposed approach was made by its&nbsp;application on typing data arising from a separate self-reported&nbsp;cohort of 27 PD patients&rsquo; and 84 healthy controls&rsquo; daily usage&nbsp;with their personal smartphones (data in-the-wild), achieving&nbsp;0.79 AUC with 0.74/0.78 sensitivity/specificity, respectively. The&nbsp;results show the potentiality of the proposed approach to process&nbsp;keystroke dynamics arising from users&rsquo; natural typing activity&nbsp;to detect PD, which contributes to the development of digital&nbsp;tools for remote pathological symptom screening.</p>", 
  "title": "Early Parkinson's Disease Detection via Touchscreen Typing Analysis using Convolutional Neural Networks", 
  "type": "paper-conference", 
  "id": "3675381"
}
69
7
views
downloads
All versions This version
Views 6969
Downloads 77
Data volume 34.9 MB34.9 MB
Unique views 6262
Unique downloads 77

Share

Cite as