Longitudinal Prognosis of Parkinsons Outcomes using Causal Connectivity
Parkinsons disease (PD) is a movement disorder and the second most common neurodengerative disease but despite its relative abundance, there are no clinically accepted neuroimaging biomarkers to make prognostic predictions or differentiate between the similar atypical neurodegenerative diseases Multiple System Atrophy and Progressive Supranuclear Palsy. Abnormal connectivity in circuits including the motor circuit and basal ganglia have been previously shown as early markers of neurodegeneration. Therefore, we postulate the combination patterns of interregional dysconnectivity across the brain can be used to form a patient-specific predictive model of disease state and progression in PD. These models, which employ connectivity calculated from noninvasively measured functional MRI, differentially predict between PD and the atypical lookalikes, predict progression on a disease-specific scale, and predict cognitive decline. Further, we identify the connections most informative for progression and diagnosis. When predicting the one-year progression in the Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS) and Montreal Cognitive assessment (MoCA), mean absolute errors of 1.8 and 0.6 basis points in the prediction are achieved respectively. A balanced accuracy of 0.68 is attained when distinguishing idiopathic PD versus the lookalikes and healthy controls. We additionally find network components strongly associated with the prognostic and diagnostic tasks, particularly incorporating connections within deep nuclei, motor regions, and the Thalamus. These predictions, using an MRI modality readily available in most clinical settings, demonstrate the strong potential of fMRI connectivity as a prognostic biomarker in Parkinsons disease.
Code (0)
등록된 구현이 없습니다.
Tasks
DiagnosticPrognosisSimilar Papers 제목 키워드 기반
Causal Longitudinal Prior-Fitted Networks for Counterfactual Outcome Prediction
Longitudinal treatment decisions from multivariate time-series data require predicting potential outcomes under future treatment sequences in the presence of time-varying confounding, heterogeneous patient dynamics, and …
Causal InferenceScalable Counterfactual Risk Estimation for Rare Events in Longitudinal Data
Estimating the causal effect of time-varying treatments on survival outcomes in large observational studies is computationally demanding, particularly when outcomes are rare. While g-formula-based methods such as the ite…
Causal InferenceLearning Causally Predictable Outcomes from Psychiatric Longitudinal Data
Causal inference in longitudinal biomedical data remains a central challenge, especially in psychiatry, where symptom heterogeneity and latent confounding frequently undermine classical estimators. Most existing methods …
Causal InferenceDetecting causality in Plant electrical signal by a hybrid causal analysis approach
At present, multi-electrode array (MEA) approach and optical recording allow us to acquire plant electrical activity with higher spatio-temporal resolution. To understand the dynamic information flow of the electrical si…
Harnessing the power of longitudinal medical imaging for eye disease prognosis using Transformer-based sequence modeling
Deep learning has enabled breakthroughs in automated diagnosis from medical imaging, with many successful applications in ophthalmology. However, standard medical image classification approaches only assess disease prese…
image-classificationImage ClassificationMedical Image ClassificationPrognosis+1