No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Forecasting Parkinson's Patient's Wearing-off Periods by Employing Stacked Super Learner
Abstract
Parkinson's disease (PD) patients experience a range of symptoms, necessitating personalized treatment programs. Anti-PD medications are commonplace, yet a scenario can develop in which the Parkinson's medication being taken is no longer as effective as it once was. It may result in the re-emergence of symptoms prior to the next medicine intake. This is referred to as "wearing off". Over time, the duration of "wearing-off" shortens, requiring effective symptom management in collaboration between patients and doctors. This study aims to develop a prediction model to determine the occurrence of "wearing-off" in anti-PD medicine. To create the predictive model, real-world data, including fitness tracker records and self-reported symptoms from a smartphone application , were used. However, conducting such a study in real-world settings presents challenges, including high-class imbalance and intra-class variability when collecting self-reported symptoms from patients. Traditional imbalance learning approaches have drawbacks, such as information loss with under-sampling and overfitting or over-optimism with oversampling methods. To address these challenges, we propose a cost-sensitive hybrid optimum ensemble classifier framework. This technique adjusts class weights to give higher importance to minority classes, effectively addressing the class imbalance issue. The approach combines outputs from diverse weighted base classifiers using a stacked generalization method, harnessing the strengths of multiple high-performing base learners while mitigating their individual limitations.
ResearchGate has not been able to resolve any citations for this publication.
Patients with Parkinson's disease (PD) are very vulnerable to the negative effects of psychological distress: neuropsychiatric symptoms, such as anxiety and depression, are highly prevalent in PD; motor symptoms (such as tremor) typically worsen in stressful situations; and dopaminergic medication is less effective. Furthermore, animal studies of PD suggest that chronic stress may accelerate disease progression. Adequate self‐management strategies are therefore essential to reduce the detrimental effects of chronic stress on PD. Mindfulness‐based interventions encourage individuals to independently self‐manage and adapt to the challenges created by their condition. In PD, emerging clinical evidence suggests that mindfulness‐based interventions may reduce psychological distress and improve clinical symptoms, but insight into the underlying mechanisms is lacking. In this viewpoint, we provide a systematic overview of existing mindfulness trials in PD. Furthermore, we discuss the cerebral mechanisms involved in acute and chronic stress, and the impact of mindfulness‐based interventions on these networks. In addition, we delineate a hypothetical mechanistic framework of how chronic stress may increase the susceptibility for neuropsychiatric symptoms in PD and may potentially even influence disease progression. We end with offering recommendations for future research.
Background:
Development of "Wearing Off" (WO) of motor and non-motor function in Parkinson's disease (PD) adversely affects quality of life. This suggest that identifying and treating WO is important. However, identification of WO depends on people with PD (PwP) recognising and reporting WO and there is a perception that WO may be significantly underestimated.
Objective:
We investigate the feasibility of identifying "Wearing Off" using objective measurement and assess the clinical benefit in rectifying it.
Method:
In this study, 200 PwP were studied for evidence of WO using a continuously worn wearable system. Eighty-five patients (43%) were found to have WO and treatment was changed to mitigate the effects of WO.
Results:
Factors, such as duration of disease, high baseline MDS-UPDRS (motor component), high Percent Time in Bradykinesia (PTB), high Levodopa Equivalent Daily Dose (LEDD), frequent Levodopa doses and younger age of onset, are associated with severity of motor complications. Patients with more severe WO experienced worse motor and non-motor symptoms and lower quality of life. Quality of life significantly improved in PwP when WO was treated.
Conclusion:
The findings reported in this study provide evidence that identifying and treating WO improves outcomes of PwP and that objective measurements may help clinicians to identify and treat WO.
Freezing of gait (FoG) is a common motor symptom in patients with Parkinson’s disease (PD). FoG impairs gait initiation and walking and increases fall risk. Intelligent external cueing systems implementing FoG detection algorithms have been developed to help patients recover gait after freezing. However, predicting FoG before its occurrence enables preemptive cueing and may prevent FoG. Such prediction remains challenging given the relative infrequency of freezing compared to non-freezing events. In this study, we investigated the ability of individual and ensemble classifiers to predict FoG. We also studied the effect of the ADAptive SYNthetic (ADASYN) sampling algorithm and classification cost on classifier performance. Eighteen PD patients performed a series of daily walking tasks wearing accelerometers on their ankles, with nine experiencing FoG. The ensemble classifier formed by Support Vector Machines, K-Nearest Neighbors, and Multi-Layer Perceptron using bagging techniques demonstrated highest performance (F1 = 90.7) when synthetic FoG samples were added to the training set and class cost was set as twice that of normal gait. The model identified 97.4% of the events, with 66.7% being predicted. This study demonstrates our algorithm’s potential for accurate prediction of gait events and the provision of preventive cueing in spite of limited event frequency.
Wearable devices can capture objective day-to-day data about Parkinson’s Disease (PD). This study aims to assess the feasibility of implementing wearable technology to collect data from multiple sensors during the daily lives of PD patients. The Parkinson@home study is an observational, two-cohort (North America, NAM; The Netherlands, NL) study. To recruit participants, different strategies were used between sites. Main enrolment criteria were self-reported diagnosis of PD, possession of a smartphone and age≥18 years. Participants used the Fox Wearable Companion app on a smartwatch and smartphone for a minimum of 6 weeks (NAM) or 13 weeks (NL). Sensor-derived measures estimated information about movement. Additionally, medication intake and symptoms were collected via self-reports in the app. A total of 953 participants were included (NL: 304, NAM: 649). Enrolment rate was 88% in the NL (n = 304) and 51% (n = 649) in NAM. Overall, 84% (n = 805) of participants contributed sensor data. Participants were compliant for 68% (16.3 hours/participant/day) of the study period in NL and for 62% (14.8 hours/participant/day) in NAM. Daily accelerometer data collection decreased 23% in the NL after 13 weeks, and 27% in NAM after 6 weeks. Data contribution was not affected by demographics, clinical characteristics or attitude towards technology, but was by the platform usability score in the NL (χ² (2) = 32.014, p<0.001), and self-reported depression in NAM (χ²(2) = 6.397, p = .04). The Parkinson@home study shows that it is feasible to collect objective data using multiple wearable sensors in PD during daily life in a large cohort.
Although there is clinical demand for new technology that can accurately measure Parkinsonian tremors, automatic scoring of Parkinsonian tremors using machine-learning approaches has not yet been employed. This study aims to fill this gap by proposing machine-learning algorithms as a way to predict the Unified Parkinson’s Disease Rating Scale (UPDRS), which are similar to how neurologists rate scores in actual clinical practice. In this study, the tremor signals of 85 patients with Parkinson’s disease (PD) were measured using a wrist-watch-type wearable device consisting of an accelerometer and a gyroscope. The displacement and angle signals were calculated from the measured acceleration and angular velocity, and the acceleration, angular velocity, displacement, and angle signals were used for analysis. Nineteen features were extracted from each signal, and the pairwise correlation strategy was used to reduce the number of feature dimensions. With the selected features, a decision tree (DT), support vector machine (SVM), discriminant analysis (DA), random forest (RF), and k-nearest-neighbor (kNN) algorithm were explored for automatic scoring of the Parkinsonian tremor severity. The performance of the employed classifiers was analyzed using accuracy, recall, and precision, and compared to other findings in similar studies. Finally, the limitations and plans for further study are discussed.
Background:
The early detection of wearing-off in Parkinson disease (DEEP) observational study demonstrated that women with Parkinson's disease (PD) carry an increased risk (80.1%) for wearing-off (WO). This post hoc analysis of DEEP study evaluates gender differences on WO and associated phenomena.
Methods:
Patients on dopaminergic treatment for ≥ 1 year were included in this multicenter observational cross-sectional study. In a single visit, WO was diagnosed based on neurologist assessment as well as the use of the 19-item wearing-off questionnaire (WOQ-19); WO was defined for scores ≥ 2. Post hoc analyses were conducted to investigate gender difference for demographic and clinical features with respect to WO.
Results:
Of 617 patients enrolled, 236 were women and 381 were men. Prevalence of WO was higher among women, according to both neurologists' judgment (61.9% versus 53.8%, P = 0.045) and the WOQ-19 analysis (72.5% versus 64.0%, P = 0.034). In patients with WO (WOQ-19), women experienced ≥ 1 motor symptom in 72.5% versus 64.0% in men and ≥ 1 nonmotor symptom in 44.5% versus 36.7%, in men.
Conclusions:
Our results suggest WO as more common among women, for both motor and nonmotor symptoms. Prospective studies are warranted to investigate this potential gender-effect.
Sleep disturbances are among the most common nonmotor complaints of patients with Parkinson's disease (PD), and can have a great impact on quality of life. These disturbances manifest in a variety of ways; for instance, insomnia, sleep fragmentation, and excessive daytime sleepiness. Sleep-related movement disorders such as restless legs syndrome and periodic leg movements may share a common pathophysiology, and occurrence of rapid eye movement behavior disorder may predate the onset of PD or other synucleinopathies by several years. Medications for PD can have a significant impact on sleep, representing a great challenge to the treating physician. Awareness of the complex relationship between PD and sleep disorders, as well as the varied way in which sleep disturbances appear, is imperative for successful long-term management.
The prevalence of self-rated sleep disturbance in patients with Parkinson's disease (PD) and their spouses was compared with healthy controls, and the association of sleep disturbance with demographic, psychological, and disease variables was assessed.
The sleep ratings from three groups, PD patients, their spouses, and healthy controls, were compared using analyses of variance. Stepwise regressions were used to predict sleep disturbance for each group and gender.
Participants completed questionnaires as part of a nationwide survey in Germany.
Participants included 153 PD-spouse pairs and a group of 103 healthy controls.
Zung Self-Rating Depression Scale and self-ratings of sleep disturbance, stress level, and disease symptoms (for PD patients).
Sleep disturbances were significantly higher in women than in men in all groups. For PD patients, sleep disturbance occurred frequently in 25% of male and 41% of female participants and was best predicted by the patient's depression rating. For spouses, frequent sleep disturbance was reported by 27% of male and 48% of females and was likewise predicted by the spouse's own rating of depression. A second, relatively less common type of sleep disturbance was also reported by spouses. This disturbance was associated with waking during the night to help the patient and was best predicted by patient factors.
Improvement of sleep quality of caregivers may be an important component of treatment to reduce distress caused by PD.
Heart rate variability (HRV) decrease in Parkinson's disease (PD) could only be a consequence of reduce motor activity besides of being a marker of cardiovascular dysautonomia. Under continuously recorded and standardised motor activity, we studied thirty patients compared to controls in 3 PD stages: group I: less than 2 year-evolution, slight impaired without L-dopa; group II: mildly impaired with L-dopa; group III: advanced PD with motor complications. No difference was observed between group I and controls. The diurnal low frequency power (LF) and the ratio of LF/high frequency (HF) power decreased in groups II and III. The nocturnal vagal indicators: HF power and pNN50 were decreased in group III. Those parameters were correlated with Off-drug-motor handicap, suggesting an evolutive HRV decrease with disease severity but not with On-drug-motor activity. The low LF despite the higher motor activity in group III, due to dyskinesias, suggested a defective cardiovascular up-regulation.
Bradykinesia is a cardinal symptom of Parkinson's disease (PD) and describes the slowness of movement revealed in patients. Current PD therapies are based on dopamine replacement, and given that bradykinesia is the symptom that best correlates with the dopaminergic deficiency, the knowledge of its fluctuations may be useful in the diagnosis, treatment and better understanding of the disease progression. This paper evaluates a machine learning method that analyses the signals provided by a triaxial accelerometer placed on the waist of PD patients in order to automatically assess bradykinetic gait unobtrusively. This method employs Support Vector Machines to determine those parts of the signals corresponding to gait. The frequency content of strides is then used to determine bradykinetic walking bouts and to estimate bradykinesia severity based on an epsilon-Support Vector Regression model. The method is validated in 12 PD patients, which leads to two main conclusions. Firstly, the frequency content of the strides allows for the dichotomic detection of bradykinesia with an accuracy higher than 90%. This process requires the use of a patient-dependant threshold that is estimated based on a leave-one-patient-out regression model. Secondly, bradykinesia severity measured through UPDRS scores is approximated by means of a regression model with errors below 10%. Although the method has to be further validated in more patients, results obtained suggest that the presented approach can be successfully used to rate bradykinesia in the daily life of PD patients unobtrusively.
Objective
Assessing the frequency of Wearing-Off (WO) in Parkinson's disease (PD) patients, and its impact on Quality of Life (QoL).
Methods
Consecutive ambulatory patients, who were on dopaminergic treatment for ≥1 year, were included in this multicentre, observational cross-sectional study. In a single visit, WO was diagnosed based on neurologist assessment as well as using the validated Italian version of a patient self-rated 19-question Wearing-Off Questionnaire (WOQ-19); WO was defined for scores ≥ 2. QoL was evaluated by the 8-item Parkinson's Disease Questionnaire (PDQ-8).
Results
617 subjects were included, with a mean anti-Parkinson treatment duration of 6.6 ± 4.6 years, 87.2% were on levodopa treatment. Neurologists identified presence of WO in 351 subjects (56.9%), whereas 415 subjects (67.3%) were identified by the self-administered WOQ-19. In patients with a <2.5 years disease duration, WO was diagnosed in 12 subjects (21.8%) by neurologists and in 23 subjects (41.8%) by the WOQ-19. The most frequent WO symptoms, as identified by WOQ-19, were “slowness of movements” (55.8%) and “reduced dexterity” (48.8%). Younger age, female gender, Unified Parkinson's Disease Rating Scale (UPDRS) part II score and duration of anti-Parkinson treatment were found significantly associated with WO. The number of motor (p < 0.0001) and non-motor (p < 0.0001) WO symptoms correlated with PDQ-8 total score.
Conclusions
WO is common already at the early stages of PD and is underestimated by routine neurological clinical evaluation. The number of WO symptoms, both motor and non motor, increases along with disease duration and has a negative impact on patients QoL.
Freezing of gait (FOG) is one of the most disabling symptoms that affect patients with Parkinson's disease (PD). Although the pathophysiology underlying FOG largely remains an enigma, several lines of evidence suggest that the autonomic nervous system might be involved. To this end, we tested the hypothesis that heart rate (HR) increases during FOG and, further, that HR increases just before FOG. To evaluate these hypotheses, 15 healthy older adults, 10 patients with PD who experienced FOG, and 10 patients who did not were studied. Patients with PD were tested during their “off” medication state. HR and HR variability were measured as subjects carried out tasks that frequently provoke FOG; 120 FOG episodes were evaluated. During FOG, HR increased (P = 0.001) by an average of 1.8 bpm, compared with HR measured before the beginning of FOG. HR also increased just before FOG, by 1 bpm (P < 0.0001). In contrast, during sudden stops and 180° turns, HR decreased by almost 2 bpm (P < 0.0001). HR variability was not associated with FOG. To our knowledge, these findings are the first to document the association of FOG to autonomic system activation, as manifested by HR dynamics. One explanation is that the changes in HR before and during FOG may be a sympathetic response that, secondary to limbic activation, contributes to the development of freezing. Although further studies are needed to evaluate these associations, the current results provide experimental evidence linking impaired motor blockades to autonomic nervous system function among patients with PD. © 2010 Movement Disorder Society
We have previously reported that the use of a 32-symptom Wearing-off Questionnaire (WOQ-32) identified wearing off more frequently than a clinician's evaluation or the complications subscale of the Unified Parkinson Disease Rating Scale (UPDRS). However, this prototype tool was not designed for clinical practice and required simplification for daily use. Although wearing off is a commonly understood concept among neurologists caring for Parkinson disease patients, there are a number of definitions in the literature. For the purpose of this study and to include both motor and nonmotor parkinsonian symptoms, wearing off was defined as a generally predictable recurrence of motor and nonmotor symptoms that precedes scheduled doses of anti-parkinsonian medication and usually improves after those doses. Using this definition, retrospective analysis and expert opinion were used to identify the 9 most predictive and relevant of the symptoms previously identified as part of the WOQ-32. The resulting 9-symptom questionnaire (WOQ-9) identified 158 (95.8%) of the 165 subjects captured by the 32-Symptom Wearing-off Questionnaire as having wearing off, excluding 7 subjects reporting only balance difficulty (n = 3), numbness (n = 2), difficulty standing (n = 1), and abdominal discomfort (n = 1). Subjects reporting wearing off with the WOQ-9 were significantly younger, had been longer diagnosed with Parkinson disease, experienced a longer duration of levodopa therapy, exhibited a higher UPDRS total score, had higher levodopa equivalent dosages, and increased dyskinesia compared with patients not identified as wearing off with the WOQ-9. No statistical differences were noted with respect to sex, UPDRS subsection scores, Schwab & England Scale, or Hoehn & Yahr Scale.
Our study aimed to investigate the cardiovascular autonomic regulation related to the wearing-off phenomenon in Parkinson's disease (PD). We measured blood pressure (BP) and heart rate (HR) at rest and during orthostatic test in 16 patients with PD with wearing-off and in 15 patients with PD without wearing-off both before (baseline) and repetitively at 1-h intervals for up to 4 h after the morning PD medication dose.
The patients with wearing-off had fluctuation of BP during the observation period, BP increasing when the motor performance worsened and vice versa. The mean supine BP was at its highest at the baseline measurement (patients with wearing-off, 145 ± 18 mmHg; patients without wearing-off, 138 ± 17 mmHg), fell during the first hour (patients with wearing-off, 119 ± 17 mmHg; patients without wearing-off, 126 ± 18 mmHg), and then rose again toward the end of the observation period (patients with wearing-off, 136 ± 15 mmHg; patients without wearing-off, 138 ± 18 mmHg). This BP change was statistically significant only in PD patients with wearing-off (P < 0.001).
In conclusion, BP seems to fluctuate with motor impairment in PD patients with wearing-off. This fluctuation may represent autonomic dysfunction caused by the PD process itself, the effect of PD medication, or both.
Wearing-off scales in parkinson's disease: Critique and recommendations
- A Antonini
- P Martinez-Martin
- R K Chaudhuri
- M Merello
- R Hauser
- R Katzenschlager
- P Odin
- M Stacy
- F Stocchi
- W Poewe
- O Rascol
- C Sampaio
- A Schrag
- G T Stebbins
- C G Goetz
Antonini, A., Martinez-Martin, P., Chaudhuri, R.K., Merello,
M., Hauser, R., Katzenschlager, R., Odin, P., Stacy, M., Stocchi, F., Poewe, W., Rascol, O., Sampaio, C., Schrag, A., Stebbins, G.T., Goetz, C.G.: Wearing-off scales in parkinson's disease: Critique and recommendations. Movement Disorders 26(12),
2169-2175 (2011). DOI https://doi.org/10.1002/mds.23875. URL
https://movementdisorders.onlinelibrary.wiley.com/doi/abs/10.1002/mds.23875
5th abc challenge: Forecasting parkinson's disease patients' wearing-off phenomenon datasets
- J N V H K C Inoue
Inoue, J.N.V.H.K.C.G.D.H.M.F.N.N.T.H.T.S.S.: 5th abc challenge: Forecasting parkinson's disease patients' wearing-off phenomenon datasets
(2023). DOI 10.21227/f9cx-cx72. URL https://dx.doi.org/10.21227/f9cx-cx72
Selegiline and blood pressure in patients with parkinson's disease
- V Pursiainen
- T J Korpelainen
- H T Haapaniemi
- A K Sotaniemi
- V V Myllylä
Pursiainen, V., Korpelainen, T.J., Haapaniemi, H.T., Sotaniemi,
A.K., Myllylä, V.V.: Selegiline and blood pressure in patients with
parkinson's disease.
Acta Neurologica Scandinavica 115(2), 104-108 (2007).
DOI https://doi.org/10.1111/j.1600-0404.2006.00742.x.
URL
https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1600-0404.2006.00742.x
Understanding wearing-off symptoms in parkinson's disease patients using wrist-worn fitness tracker and a smartphone
- J N Victorino
- Y Shibata
- S Inoue
- T Shibata
Victorino, J.N., Shibata, Y., Inoue, S., Shibata, T.: Understanding
wearing-off symptoms in parkinson's disease patients using wrist-worn
fitness tracker and a smartphone. Procedia Computer Science 196,
684-691 (2022). DOI https://doi.org/10.1016/j.procs.2021.12.064. URL
https://www.sciencedirect.com/science/article/pii/S1877050921022870.
International Conference on ENTERprise Information Systems / Pro-jMAN -International Conference on Project MANagement / HCist -International Conference on Health and Social Care Information Systems
and Technologies 2021