Predicting Parkinson’s Disease and Its Pathology via Simple Clinical Variables

Abstract

Background: Parkinson’s disease (PD) is a chronic, disabling neurodegenerative disorder. Objective: To predict a future diagnosis of PD using questionnaires and simple non-invasive clinical tests. Methods: Participants in the prospective Kuakini Honolulu-Asia Aging Study (HAAS) were evaluated biannually between 1995–2017 by PD experts using standard diagnostic criteria. Autopsies were sought on all deaths. We input simple clinical and risk factor variables into an ensemble-tree based machine learning algorithm and derived models to predict the probability of developing PD. We also investigated relationships of predictive models and neuropathologic features such as nigral neuron density. Results: The study sample included 292 subjects, 25 of whom developed PD within 3 years and 41 by 5 years. 116 (46%) of 251 subjects not diagnosed with PD underwent autopsy. Light Gradient Boosting Machine modeling of 12 predictors correctly classified a high proportion of individuals who developed PD within 3 years (area under the curve (AUC) 0.82, 95%CI 0.76–0.89) or 5 years (AUC 0.77, 95%CI 0.71–0.84). A large proportion of controls who were misclassified as PD had Lewy pathology at autopsy, including 79%of those who died within 3 years. PD probability estimates correlated inversely with nigral neuron density and were strongest in autopsies conducted within 3 years of index date (r = –0.57, p < 0.01). Conclusion: Machine learning can identify persons likely to develop PD during the prodromal period using questionnaires and simple non-invasive tests. Correlation with neuropathology suggests that true model accuracy may be considerably higher than estimates based solely on clinical diagnosis.

Externally Validated Deep Learning Model to Identify Prodromal Parkinson’s Disease from Electrocardiogram

“…Furthermore, there is the potential to include demographic and clinical variables as additional inputs with the predicted outcomes from the ECG CNN model, to increase the accuracy of predicting prodromal PD within an extended time window. Karabayir et al 32 Another limitation of this study was that our chart reviews at LUC showed that out of 47 cases identi ed in LUC, only 29 of them were real PD cases. Therefore, our study also highlights the challenges around nding PD patients via her-based queries.…”

“…Studies found that heart rate variability (HRV) determined from 5-minute ECGs is reduced in prevalent PD, and results from a single prospective study showed that lower HRV was associated with an increased risk of incident PD. 34 Prior work by our group utilized machine learning approaches to predict prodromal PD using clinical variables in one study 32 and a proof-of-concept study using standard 10second printed ECGs in another study 30 . Both studies resulted in moderate accuracy, however the data and model developed in the latter study was biased towards an elderly population and also lacked an external cohort for model validation.…”

Pretreatment identification of 90-day readmission among heart failure patients receiving aquapheresis treatment

“…We implemented a previously developed Feature Importance and Direction Analysis (FIDA) 21 to identify the most important predictors of 90-day readmission and identify the direction of the interaction between input and output. In FIDA, the importance of a predictor is assessed by artificially increasing its value by one standard deviation (or swapping with a reference category for nominal variables or increasing one unit for ordinal variables) and assessing the change in predicted risk compared.…”

Identifying prodromal symptoms at high specificity for Parkinson’s disease

“…Several predictive models for PD have been suggested in the last decade, ten of which were critically appraised in a systematic review by Chen et al (2023) . Three models: Mahlknecht et al (2016) , Faust et al (2020) , and Karabayir et al (2022) were recommended by Chen et al (2023) , which consisted of 12, 17, and 541 predictors, respectively, including age and smoking status. The following non-motor symptoms: daytime sleepiness and cognitive impairment were included in Karabayir et al (2022) , urinary dysfunction, constipation and depression were included in Mahlknecht et al (2016) and Faust et al (2020) included 536 diagnosis or procedure codes.…”

“…Three models: Mahlknecht et al (2016) , Faust et al (2020) , and Karabayir et al (2022) were recommended by Chen et al (2023) , which consisted of 12, 17, and 541 predictors, respectively, including age and smoking status. The following non-motor symptoms: daytime sleepiness and cognitive impairment were included in Karabayir et al (2022) , urinary dysfunction, constipation and depression were included in Mahlknecht et al (2016) and Faust et al (2020) included 536 diagnosis or procedure codes. However, there is still much to learn and uncover on this journey towards a specific diagnostic test for PD.…”