Correlation of subthalamic nuclei T2 relaxation times with neuropsychological symptoms in patients with Parkinson's disease

PassFlow: a multimodal workflow for predicting deep brain stimulation outcomes

PURPOSE

Deep Brain Stimulation (DBS) is a proven therapy for Parkinson's Disease (PD), frequently resulting in an enhancement of motor function. Nonetheless, several undesirable side effects can occur after DBS, which can worsen the quality of life of the patient. Thus, the clinical team has to carefully select patients on whom to perform DBS. Over the past decade, there have been some attempts to relate pre-operative data and DBS clinical outcomes, with most focused on the motor symptomatology. In this paper, we propose a machine learning-based method able to predict a large number of DBS clinical outcomes for PD.

METHODS

We propose a multimodal pipeline, referred to as PassFlow, which predicts 84 clinical post-operative clinical scores. PassFlow is composed of an artificial neural network to compress clinical information, an image processing method from the state-of-the-art to extract morphological biomarkers our of T1 imaging, and an SVM to perform the regressions. We validated PassFlow on 196 PD patients who undergone a DBS.

RESULTS

PassFlow showed correlation coefficients as high as 0.71 and were able to significantly predict 63 out of the 84 scores, outperforming a comparative linear method. The number of metrics that are predicted with this pre-operative information was also found to be correlated with the number of patients with this information available, indicating that the PassFlow method is still actively learning.

CONCLUSION

We presented a novel, machine learning-based pipeline to predict a variety of post-operative clinical outcomes of DBS for PD patients. PassFlow took into account various bio-markers, arising from different data modalities, showing high correlation coefficients for some scores from pre-operative data only. It indicates that many clinical outcomes of DBS can be predicted agnostic to the specific simulation parameters, as PassFlow has been validated without such stimulation-related information.

T2-relaxometry predicts outcome of DBS in idiopathic Parkinson's disease

Introduction

Deep brain stimulation (DBS) nowadays is a well-established treatment of motor symptoms in Parkinson's disease. The subthalamic nucleus (STN) is a common target for DBS, because motor improvements have been shown to be superior to best medical therapy, if DBS electrodes have been appropriately positioned. DBS target identification can be assisted by MRI beyond structural imaging by spatially resolved measurement of T2-relaxation times (T2r).

Aim

We pose the question, whether T2r of the STN is linked to the severity of the disease and whether outcome of DBS may be correlated to an asymmetric manifestation of the disease. Further, we investigated if abnormal T2r in the STN may be predictive for outcome of DBS.

Methods

Twelve patients underwent preoperative MR imaging including a multi echo relaxometry sequence (3 Tesla, Siemens Medical Systems, Erlangen, Germany) ahead of DBS. T2r were determined for STN, substantia nigra (SN), red nucleus (RN) and centrum semiovale (CSO). Unified Parkinson's disease Rating Scale (UPDRS) scores were tested before and after DBS. Patients' T2r and deduced values representing left-right asymmetry of measurements were correlated with UPDRS scores and measures for outcome of DBS. Furthermore, patients' T2r were compared with T2r measurements in 12 healthy controls (HC).

Results

Patients' T2r for SN (mean 45.4 ms ± 4.4 ms) and STN (mean 56.4 ms ± 3.8 ms) were significantly shorter than T2r in HCs for SN (mean 60.7 ± 4.6) and STN (mean 66.1 ms ± 4.0 ms). While no mean T2r asymmetry was found in the SN, patients' mean T2r for STN showed a weakened left-right correlation (Pearson correlation coefficient 0.19 versus 0.72 in HC) indicating asymmetric degeneration. T2r asymmetry was not linked to the more severely affected hemisphere.

The respective lower T2r within the left or right target region was significantly correlated to the outcome in terms of UPDRS III improvement in “off” state (Pearson correlation 0.82 corresponding to p ≪ 0.01). Patients with T2r of STN lower than 50 ms showed no response to DBS in the UPDRS.

The maximum T2r for SN correlated to the improvement between UPDRS “off” minus and “on” (Dopamine response) but failed to predict DBS outcome.

Conclusions

The lower boundaries of T2r in the STN predict motor outcome in DBS. T2r asymmetry in the STN is not associated with increased clinical symptoms, but with response to therapy. Thus, patients with very low T2r may be inappropriate candidates for DBS.

•

Parkinson's disease features are reflected in spatially resolved measurement of MR T2-relaxation times.

•

Deep brain stimulation therapy target identification can be assisted by measurement of MR T2-relaxation times.

•

T2r asymmetry in the subthalamic nucleus is not associated with increased clinical symptoms, but with response to therapy.

•

Lower boundaries of T2r in the STN predict motor outcome in Deep brain stimulation therapy

•

Patient selection criteria may be improved by including parameters based on MR relaxometry.