Automatic assessment of levodopa‐induced dyskinesias in daily life by neural networks

Abstract

We developed an objective and automatic procedure to assess the severity of levodopa-induced dyskinesia (LID) in patients with Parkinson's disease during daily life activities. Thirteen patients were continuously monitored in a home-like situation for a period of approximately 2.5 hours. During this time period, the patients performed approximately 35 functional daily life activities. Behavior of the patients was measured using triaxial accelerometers, which were placed at six different positions on the body. A neural network was trained to assess the severity of LID using various variables of the accelerometer signals. Neural network scores were compared with the assessment by physicians, who evaluated the continuously videotaped behavior of the patients off-line. The neural network correctly classified dyskinesia or the absence of dyskinesia in 15-minute intervals in 93.7, 99.7, and 97.0% for the arm, trunk, and leg, respectively. In the few cases of misclassification, the rating by the neural network was in the class next to that indicated by the physicians using the AIMS score (scale 0-4). Analysis of the neural networks revealed several new variables, which are relevant for assessing the severity of LID. The results indicate that the neural network can accurately assess the severity of LID and could distinguish LID from voluntary movements in daily life situations.

Dopaminergic-induced dyskinesia assessment based on a single belt-worn accelerometer

“…through a single sensor located on the trunk provided 82% of accuracy [12]. In the approach presented in this paper, a 95% of accuracy on trunk dyskinesia and any severe dyskinesia from any part of the body is obtained, surpassing by more than 10 % the previous results.…”

“…In addition, it is unknown if these studies analysed any of the usability restrictions previously described in this article, since the great majority of them use some sensors distributed on the body [12][13][14]17,18,20,22]. As far as authors know, the only paper, apart from the previously submitted by authors [16], that uses a single sensor for dyskinesia assessment is operating from the shoulder [15], which is a position contrary to the restrictions that have been presented so far.…”

Mobile Systems as a Challenge for Neurological Diseases Management – The Case of Parkinson's Disease

“…An important number of these studies is based on the study of various parameters of motor behaviour, in particular features related to the gait (R. Greenlaw et al, 2009;Salarian et al, 2007;. Other studies focused on the identification of ON/OFF fluctuations through the assessment of tremor (Van Someren et al, 1993), dyskinesias (Keijsers et al, 2003; and bradykinesia (Papapetropoulos et al, 2010). Some groups are also committed to use electromyogram (EMG) or voice analysis (Kimura et al, 2007).…”

Selective stimulation of the subthalamic nucleus

“…This may have negatively influenced our results on bradykinesia evaluation. Dyskinesia, which has been studied before by Keijsers et al (Keijsers et al 2002) and , is not considered in this paper. Nonetheless, this is an important aspect that can be used to identify on and off-periods.…”