Method to Quantify Parkinsonian Motor Signs for Interventional Drug Trials
Improved Biomarkers and Clinical Outcome Measures, 2015
Current methods to evaluate motor impairments in Parkinson’s disease rely on subjective examinations. Our team seeks to develop an objective assessment of motor deficits by monitoring natural interactions with a keyboard (on a computer or smart device). This approach provides a window to how the brain behaves during typical daily use of these devices (e.g. writing a report or sending an email) and thus has the potential to be used easily and regularly.
The motor response to treatments used for Parkinson’s can be detected measuring finger movements while using an electronic device.
We will evaluate newly diagnosed Parkinson's disease patients who are prescribed dopaminergic medication by their neurologist. Each participant will use their digital device for the duration of the study and will have a thorough motor examination at three time points. With our technology we plan to detect a motor change induced by these medications as accurately as the standard tests used currently. Importantly, the data gathered are non-sensitive and based only on timing information.
Impact on Diagnosis/Treatment of Parkinson’s Disease:
Our project has the potential to facilitate the development of therapies for Parkinson's disease by providing objective measurements that reflect more accurately the degree of motor impairment. This technology can be widely deployed, is non-intrusive and suitable for routine monitoring. Thus, this enabling technology will allow the drug response to be objectively quantified and routinely monitored with a minimal effort and maximized compliance from the patient and physician side.
Next Steps for Development:
After this study, we plan to test this same technology in a drug trial to see if the expected benefits are confirmed.
In this project, we evaluated whether keyboard typing patterns can be used to detect the response to Parkinson's disease (PD) medications. We used several artificial intelligence approaches to evaluate how long each key is pressed during conventional typing on a laptop. We found moderate agreement between typing metrics and commonly used scales to evaluate Parkinson's, such as United Parkinson's Disease Rating Scale Section III (UPDRS-III). Furthermore, these early results suggest that typing metrics collected at week 8 of treatment may be able to predict which people respond to the drug by week 24. Finally, this approach was easily adopted and well tolerated by study participants. Collectively, these results advance our progress in establishing technologies that facilitate and accelerate the development and testing of medications to treat PD.
- Giancardo L, Sanchez-Ferro A, Arroyo-Gallego T, et al. Computer keyboard interaction as an indicator of early Parkinson ' s disease. Nat Publ Gr. 2016:1-10.
N.B. More submissions are planned
- Sánchez Ferro Á, Sánchez Mendoza C, Butterworth I, et al. "Caracterización motora de la enfermedad de Parkinson a través de la interacción natural con dispositivos electrónicos (neuroQWERTY)".
Presented at the LXVII meeting of the Spanish Neurological Association. November, 2015. Valencia, Spain.
- Sánchez Ferro Á, Sánchez Mendoza C, Butterworth I, et al. Evaluating Parkinsonian motor features via the routine use of consumer electronics -- neuroQWERTY.
Presented at 20th International Congress of Parkinson's Disease and Movement. June 2016. Berlin, Germany.
J.W. Kieckhefer Professor of Medical and Electrical Engineering at Massachusetts Institute of Technology
Location: Cambridge, Massachusetts, United States