Using AI to measure Parkinson's disease severity at home.

We present an artificial intelligence (AI) system to remotely assess the motor performance of individuals with Parkinson's disease (PD). In our study, 250 global participants performed a standardized motor task involving finger-tapping in front of a webcam. To establish the severity of Parkinsonian symptoms based on the finger-tapping task, three expert neurologists independently rated the recorded videos on a scale of 0-4, following the Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS). The inter-rater reliability was excellent, with an intra-class correlation coefficient (ICC) of 0.88. We developed computer algorithms to obtain objective measurements that align with the MDS-UPDRS guideline and are strongly correlated with the neurologists' ratings. Our machine learning model trained on these measures outperformed two MDS-UPDRS certified raters, with a mean absolute error (MAE) of 0.58 points compared to the raters' average MAE of 0.83 points. However, the model performed slightly worse than the expert neurologists (0.53 MAE). The methodology can be replicated for similar motor tasks, providing the possibility of evaluating individuals with PD and other movement disorders remotely, objectively, and in areas with limited access to neurological care.

A Blueprint for the Conduct of Large, Multisite Trials in Telemedicine.

Recent literature supports the efficacy and efficiency of telemedicine in improving various health outcomes despite the wide variability in results. Understanding site-specific issues in the implementation of telemedicine trials for broader replication and generalizability of results is needed. Lessons can be learned from existing trials, and a blueprint can guide researchers to conduct these challenging studies using telemedicine more efficiently and effectively. This viewpoint presents relevant challenges and solutions for conducting multisite telemedicine trials using 7 ongoing and completed studies funded by the Patient-Centered Outcomes Research Institute portfolio of large multisite trials to highlight the challenges in implementing telemedicine trials. Critical issues of ensuring leadership and buy-in, appropriate funding, and diverse and representative trials are identified and described, as well as challenges related to clinical, informatics, regulatory, legal, quality, and billing. The lessons learned from these studies were used to create a blueprint of key aspects to consider for the design and implementation of multisite telemedicine trials.

New directions in clinical trials for frontotemporal lobar degeneration: Methods and outcome measures.

INTRODUCTION: Frontotemporal lobar degeneration (FTLD) is the most common form of dementia for those under 60 years of age. Increasing numbers of therapeutics targeting FTLD syndromes are being developed. METHODS: In March 2018, the Association for Frontotemporal Degeneration convened the Frontotemporal Degeneration Study Group meeting in Washington, DC, to discuss advances in the clinical science of FTLD. RESULTS: Challenges exist for conducting clinical trials in FTLD. Two of the greatest challenges are (1) the heterogeneity of FTLD syndromes leading to difficulties in efficiently measuring treatment effects and (2) the rarity of FTLD disorders leading to recruitment challenges. DISCUSSION: New personalized endpoints that are clinically meaningful to individuals and their families should be developed. Personalized approaches to analyzing MRI data, development of new fluid biomarkers and wearable technologies will help to improve the power to detect treatment effects in FTLD clinical trials and enable new, clinical trial designs, possibly leveraged from the experience of oncology trials. A computational visualization and analysis platform that can support novel analyses of combined clinical, genetic, imaging, biomarker data with other novel modalities will be critical to the success of these endeavors.