Effect of levodopa on handwriting tasks of different complexity in Parkinson's disease: a kinematic study

Comparison of CNN-Learned vs. Handcrafted Features for Detection of Parkinson's Disease Dysgraphia in a Multilingual Dataset

Parkinson's disease dysgraphia (PDYS), one of the earliest signs of Parkinson's disease (PD), has been researched as a promising biomarker of PD and as the target of a noninvasive and inexpensive approach to monitoring the progress of the disease. However, although several approaches to supportive PDYS diagnosis have been proposed (mainly based on handcrafted features (HF) extracted from online handwriting or the utilization of deep neural networks), it remains unclear which approach provides the highest discrimination power and how these approaches can be transferred between different datasets and languages. This study aims to compare classification performance based on two types of features: features automatically extracted by a pretrained convolutional neural network (CNN) and HF designed by human experts. Both approaches are evaluated on a multilingual dataset collected from 143 PD patients and 151 healthy controls in the Czech Republic, United States, Colombia, and Hungary. The subjects performed the spiral drawing task (SDT; a language-independent task) and the sentence writing task (SWT; a language-dependent task). Models based on logistic regression and gradient boosting were trained in several scenarios, specifically single language (SL), leave one language out (LOLO), and all languages combined (ALC). We found that the HF slightly outperformed the CNN-extracted features in all considered evaluation scenarios for the SWT. In detail, the following balanced accuracy (BACC) scores were achieved: SL—0.65 (HF), 0.58 (CNN); LOLO—0.65 (HF), 0.57 (CNN); and ALC—0.69 (HF), 0.66 (CNN). However, in the case of the SDT, features extracted by a CNN provided competitive results: SL—0.66 (HF), 0.62 (CNN); LOLO—0.56 (HF), 0.54 (CNN); and ALC—0.60 (HF), 0.60 (CNN). In summary, regarding the SWT, the HF outperformed the CNN-extracted features over 6% (mean BACC of 0.66 for HF, and 0.60 for CNN). In the case of the SDT, both feature sets provided almost identical classification performance (mean BACC of 0.60 for HF, and 0.58 for CNN).

Novel insights into the effects of levodopa on the up- and downstrokes of writing sequences

Motor control of automatized and overlearned sequences, such as writing, is affected in Parkinson's disease (PD), impacting patients' daily life. Medication effects on motor performance are not only task-specific, but also variable within tasks. The nature of this variance is still unclear. This study aimed to investigate whether medication affects writing sequences differently when producing up- or downstrokes. Writing was assessed in healthy controls (HC) (N = 31) and PD (N = 32), when ON and OFF medication in a randomized order (interspersed by two months). Subjects wrote a sequential pattern with an increasing size on a digital tablet. Writing outcomes were movement vigor (amplitude and velocity), error and end-point variability, and sequence continuation, calculated separately for up- and downstrokes. Results showed that PD patients OFF-medication reduced movement vigor (amplitude) for up- and downstrokes compared to HC. Clear deficits were found for up- but not for downstroke error in PD patients in OFF, suggesting a directional bias. Dopaminergic medication improved motor vigor by increasing writing amplitude and upstroke continuation, but this occurred at the cost of the downstroke trajectory. Other writing outcomes did not improve with medication intake. In conclusion, we interpret these findings as that the impact of dopamine is complex, highly task-specific, supporting the most highly energy demanding components of a writing sequence. As medication did not regulate downstroke writing, we recommend supplementary training to address task demands that were less modulated by dopamine (registration: https://osf.io/gk5q8/ , 17 July 2018).

Convolutional neural network ensemble for Parkinson's disease detection from voice recordings

The computerized detection of Parkinson's disease (PD) will facilitate population screening and frequent monitoring and provide a more objective measure of symptoms, benefiting both patients and healthcare providers. Dysarthria is an early symptom of the disease and examining it for computerized diagnosis and monitoring has been proposed. Deep learning-based approaches have advantages for such applications because they do not require manual feature extraction, and while this approach has achieved excellent results in speech recognition, its utilization in the detection of pathological voices is limited. In this work, we present an ensemble of convolutional neural networks (CNNs) for the detection of PD from the voice recordings of 50 healthy people and 50 people with PD obtained from PC-GITA, a publicly available database. We propose a multiple-fine-tuning method to train the base CNN. This approach reduces the semantical gap between the source task that has been used for network pretraining and the target task by expanding the training process by including training on another dataset. Training and testing were performed for each vowel separately, and a 10-fold validation was performed to test the models. The performance was measured by using accuracy, sensitivity, specificity and area under the ROC curve (AUC). The results show that this approach was able to distinguish between the voices of people with PD and those of healthy people for all vowels. While there were small differences between the different vowels, the best performance was when/a/was considered; we achieved 99% accuracy, 86.2% sensitivity, 93.3% specificity and 89.6% AUC. This shows that the method has potential for use in clinical practice for the screening, diagnosis and monitoring of PD, with the advantage that vowel-based voice recordings can be performed online without requiring additional hardware.

Artificial Intelligence and Its Application to Minimal Hepatic Encephalopathy Diagnosis

Hepatic encephalopathy (HE) is a brain dysfunction caused by liver insufficiency and/or portosystemic shunting. HE manifests as a spectrum of neurological or psychiatric abnormalities. Diagnosis of overt HE (OHE) is based on the typical clinical manifestation, but covert HE (CHE) has only very subtle clinical signs and minimal HE (MHE) is detected only by specialized time-consuming psychometric tests, for which there is still no universally accepted gold standard. Significant progress has been made in artificial intelligence and its application to medicine. In this review, we introduce how artificial intelligence has been used to diagnose minimal hepatic encephalopathy thus far, and we discuss its further potential in analyzing speech and handwriting data, which are probably the most accessible data for evaluating the cognitive state of the patient.

Arm swing responsiveness to dopaminergic medication in Parkinson's disease depends on task complexity

The evidence of the responsiveness of dopaminergic medication on gait in patients with Parkinson’s disease is contradicting. This could be due to differences in complexity of the context gait was in performed. This study analysed the effect of dopaminergic medication on arm swing, an important movement during walking, in different contexts. Forty-five patients with Parkinson’s disease were measured when walking at preferred speed, fast speed, and dual-tasking conditions in both OFF and ON medication states. At preferred, and even more at fast speed, arm swing improved with medication. However, during dual-tasking, there were only small or even negative effects of medication on arm swing. Assuming that dual-task walking most closely reflects real-life situations, the results suggest that the effect of dopaminergic medication on mobility-relevant movements, such as arm swing, might be small in everyday conditions. This should motivate further studies to look at medication effects on mobility in Parkinson’s disease, as it could have highly relevant implications for Parkinson’s disease treatment and counselling.

Detecting Effect of Levodopa in Parkinson's Disease Patients Using Sustained Phonemes

BACKGROUND

Parkinson's disease (PD) is a multi-symptom neurodegenerative disease generally managed with medications, of which levodopa is the most effective. Determining the dosage of levodopa requires regular meetings where motor function can be observed. Speech impairment is an early symptom in PD and has been proposed for early detection and monitoring of the disease. However, findings from previous research on the effect of levodopa on speech have not shown a consistent picture.

METHOD

This study has investigated the effect of medication on PD patients for three sustained phonemes; /a/, /o/, and /m/, which were recorded from 24 PD patients during medication off and on stages, and from 22 healthy participants. The differences were statistically investigated, and the features were classified using Support Vector Machine (SVM).

RESULTS

The results show that medication has a significant effect on the change of time and amplitude perturbation (jitter and shimmer) and harmonics of /m/, which was the most sensitive individual phoneme to the levodopa response. /m/ and /o/ performed at a comparable level in discriminating PD-off from control recordings. However, SVM classifications based on the combined use of the three phonemes /a/, /o/, and /m/ showed the best classifications, both for medication effect and for separating PD from control voice. The SVM classification for PD-off versus PD-on achieved an AUC of 0.81.

CONCLUSION

Studies of phonation by computerized voice analysis in PD should employ recordings of multiple phonemes. Our findings are potentially relevant in research to identify early parkinsonian dysarthria, and to tele-monitoring of the levodopa response in patients with established PD.

Levodopa improves handwriting and instrumental tasks in previously treated patients with Parkinson's disease

Motor symptoms in patients with Parkinson's disease may be determined with instrumental tests and rating procedures. Their outcomes reflect the functioning and the impairment of the individual patient when patients are tested off and on dopamine substituting drugs. Objectives were to investigate whether the execution speed of a handwriting task, instrumentally assessed fine motor behavior, and rating scores improve after soluble levodopa application. 38 right-handed patients were taken off their regular drug therapy for at least 12 h before scoring, handwriting, and performance of instrumental devices before and 1 h after 100 mg levodopa intake. The outcomes of all performed procedures improved. The easy-to-perform handwriting task and the instrumental tests demand for fast and precise execution of movement sequences with considerable cognitive load in the domains' attention and concentration. These investigations may serve as additional tools for the testing of the dopaminergic response.

A Kinematic Study of Progressive Micrographia in Parkinson's Disease

Progressive micrographia is decrement in character size during writing and is commonly associated with Parkinson's disease (PD). This study has investigated the kinematic features of progressive micrographia during a repetitive writing task. Twenty-four PD patients with duration since diagnosis of <10 years and 24 age-matched controls wrote the letter “e” repeatedly. PD patients were studied in defined off states, with scoring of motor function on the Unified Parkinson's Disease Rating Scale Part III. A digital tablet captured x-y coordinates and ink-pen pressure. Customized software recorded the data and offline analysis derived the kinematic features of pen-tip movement. The average size of the first and the last five letters were compared, with progressive micrographia defined as >10% decrement in letter stroke length. The relationships between dimensional and kinematic features for the control subjects and for PD patients with and without progressive micrographia were studied. Differences between the initial and last letter repetitions within each group were assessed by Wilcoxon signed-rank test, and the Kruskal-Wallis test was applied to compare the three groups. There are five main conclusions from our findings: (i) 66% of PD patients who participated in this study exhibited progressive micrographia; (ii) handwriting kinematic features for all PD patients was significantly lower than controls (p < 0.05); (iii) patients with progressive micrographia lose the normal augmentation of writing speed and acceleration in the x axis with left-to-right writing and show decrement of pen-tip pressure (p = 0.034); (iv) kinematic and pen-tip pressure profiles suggest that progressive micrographia in PD reflects poorly sustained net force; and (v) although progressive micrographia resembles the sequence effect of general bradykinesia, we did not find a significant correlation with overall motor disability, nor with the aggregate UPDRS-III bradykinesia scores for the dominant arm.