Labelled and unlabelled hand acceleration data captured unobtrusively from PD patients and Healthy Controls

<p>The dataset contains acceleration signals captured in-the-wild&nbsp;via the IMU sensor embedded in modern smartphones, for&nbsp;the purpose of detecting tremorous episodes, related to Parkinson&#39;s Disease (PD). It contains two different groups of subjects:</p> <ul> <li>tremor_sdata.pickle --&gt; A group of 45 subjects that have been subjected to neurological examination (the same dataset as https://zenodo.org/record/3519213)</li> <li>tremor_gdata.pickle --&gt; A group of 454 subjects who just self-reported their PD status</li> </ul> <p>All subjects contributed&nbsp;accelerometer data using their personal smartphones,&nbsp;for a period spanning many months. Tri-axial acceleration values were recorded automatically whenever a phone call was realized. The recording lasted for 75 seconds at&nbsp;the most. Each phone call thus resulted in one&nbsp;recorded accelerometer signal, also referred to as session. Each subject&nbsp;contributed a different amount of sessions depending on the number of phone&nbsp;calls they realized during the data collection period as well as their participation time (they were free to drop-out at any time).&nbsp;A detailed description of the capturing process&nbsp;as well as analysis results, can be&nbsp;found in the related research article.</p> <p>The data is presented as a python dictionary, indexed by the subject ids. Each element of the dictionary is a list with the following significance:</p> <table> <thead> <tr> <th scope="col">Index</th> <th scope="col">Meaning</th> </tr> </thead> <tbody> <tr> <td>0</td> <td>List of np.arrays, Each array contains the power spectral density for an acceleration segment of 5s duration</td> </tr> <tr> <td>1</td> <td>Dictionary, Denotes subject updrs</td> </tr> <tr> <td>2</td> <td>List of str containing a unique identifier of the acceleration session that each segment in the other lists belongs to</td> </tr> <tr> <td>3</td> <td>List of np.arrays, Each array contains the pre-processed acceleration values for a 5s segment</td> </tr> </tbody> </table> <p>&nbsp; </p><p>The subject updrs is represented as dictionary containing the following tremor-related annotation values (FOR THE FIRST GROUP ONLY):<br> * updrs16: scalar int<br> The value related to tremor as described in item 16&nbsp;of the part II of the MDS-UPDRS scale, as reported by the subject.</p> <p></p> <p>* updrs20_right: scalar int in range [0, 4]<br> The value related to rest tremor in the right hand&nbsp;as described in item 20 of the part III of the MDS-UPDRS&nbsp;scale, as reported by the attending neurologist.</p> <p>* updrs20_left: scalar int in range [0, 4]<br> Same as above but for left hand.</p> <p>* updrs21_right: scalar int in range [0, 4]<br> The value related to action/postural tremor in the right hand&nbsp;as described in item 21 of the part III of the MDS-UPDRS scale, as reported&nbsp;by the attending neurologist.</p> <p>* updrs21_left: scalar int in range [0, 4]<br> Same as above but for left hand.</p> <p>* sp_expert: scalar int in range [0, 1]<br> A binary tremor annotation created by a group of signal processing experts,&nbsp;upon visually examining the contributed signals in both time and frequency domain&nbsp;and taking into consideration the UDPRS scores of each subject. This was necessary&nbsp;due to the intermittent nature of tremor, as well as a number of considerations&nbsp;related to the in-the-wild nature of the data capturing process. For more details,&nbsp;we refer the reader to the dataset description in the related research article.<br> A &#39;1&#39; value indicates that the subject has tremor.<br> A &#39;0&#39; value indicates that the subject doesn&#39;t have tremor.</p> <p>* pd_status: scalar int in range [0, 1]<br> A &#39;1&#39; value indicates that the subject is a PD patient.<br> A &#39;0&#39; value indicates that the subject is a Healthy Control</p> <p>Note: Each annotation value refers to the subject as a whole, and not in any one&nbsp;session.</p>