Search

Objective Although a number of clinical factors have been linked to falls in Parkinson’s disease (PD), the diagnostic value of gait parameters remains subject to debate. The objective of this retrospective study was to determine to what extent the combination of gait parameters with clinical characteristics can distinguish between fallers and non-fallers. Methods Using a video motion system, we recorded gait in 174 patients with PD. The patients’ clinical characteristics (including motor status, cognitive status, disease duration, dopaminergic treatment and any history of falls or freezing of gait) were noted. The considered kinematic gait parameters included indices of gait bradykinesia and hypokinesia, asymmetry, variability, and foot clearance. After a parameters selection using an ANCOVA analysis, support vector machine algorithm was used to build classification models for distinguishing between fallers and non-fallers. Two models were built, the first included clinical data only while the second incorporated the selected gait parameters. Results The “clinical-only” model had an accuracy of 94% for distinguishing between fallers and non-fallers. The model incorporating additional gait parameters including stride time and foot clearance performed even better, with an accuracy of up to 97%. Conclusion Although fallers differed significantly from non-fallers with regard to disease duration, motor impairment or dopaminergic treatment, the addition of gait parameters such as foot clearance or stride time to clinical variables increased the model’s discriminant power. Significance: This predictive model now needs to be validated in prospective cohorts.

Introduction Many clinical factors have been associated with falling in PD but the utility of gait analysis remains uncertain. To determine if additional use of clinical characteristics of the disease combined with gait parameters distinguish PD fallers from non-fallers Material and methods One hundred and seventy-four subjects were recorded during stabilized with a videomotion system. Clinical characteristics included UPDRS 3, disease duration, MMSE, occurrence of falls and freezing of gait. Kinematic parameters of gait first include gait speed, stride length and time, gait asymmetry, gait variability, foot clearance. A classification approach was investigated to build models allowing distinguishing fallers from non-fallers. Two different models were built, one by considering only the clinical data: age, disease duration, UPDRS III and MMSE scores and the second by adding to these data, gait analysis data. Support vectors machines (SVM) algorithm was used to build the models. The models were learned to tune their parameters using 4/5 of the patients’ data, randomly selected and validated blindly of the remaining 1/5 data. Results Clinical factors have an accuracy of 0.91 to distinguish fallers from non-fallers. Adding some relevant parameters such as stride length, foot clearance, variability improves this accuracy of 0.95. This was confirmed with a test sample with an accuracy of 0.87. Discussion/conclusion Fallers distinguished from non-fallers by disease duration and motor impairment but adding gait measurement improves the accuracy of the clinical data. Further research should concentrate to validate these models in prospective cohorts.

Over the last years, different efforts were made for the early prediction of cognitive worsening in different neurodegenerative diseases. EEG was widely highlighted as a potentially clinically suitable exploration modality for this purpose. Specific EEG abnormalities have been reported in different pathologies and mainly for the discrimination of mild cognitive patients from demented patients, as a general reduction of power in the alpha and beta bands and high power of widespread delta and theta rhythms in the latter group. However, findings were mainly limited for groups and population description without reaching the patient individual diagnosis. This study aims to investigate EEG rhythms alterations in a large population formed by patients complaining of cognitive distress followed in the Lille Memory Center and diagnosed according to the newest diagnostic criteria of Alzheimer’s Disease, Lewy Body disease, cerebrovascular dementia and mild cognitive impairments. Data including standard 19 electrodes resting-state closed eyes EEG, acquired in the clinical neurophysiological department and clinical evaluations including cognitive status by means of MMSE score and neuropsychological assessment, were collected from 320 patients. EEG from 300 other patients, matched in term of sex and age with the first patients’ group were also considered. These patients were explored in EEG for non-cognitive indications and thus considered as a control group. Different EEG analysis are considered as: visual assessment, quantitative analysis in the common frequency bands and connectivity. For the two last approaches, machine learning algorithms are investigated to build patients individual classification models.