The Feasibility of Kinarm-Based Assessment of Neurological Impairment in Primary Brain Tumour Patients

Abstract

This thesis investigated the feasibility of using Kinarm-based neurological assessment in primary brain tumour patients, a clinical population with a high incidence of neurological impairment. Current clinical assessment methods are limited by the low resolution and subjective nature of the clinical examination. Moreover, these tools are time- consuming and prone to floor and ceiling effects due to their narrow scale. Interactive robots such as the Kinarm offer multi-dimensional neurological assessment and utilize quantitative and continuous outcome measures. The Kinarm therefore provides a more robust, objective, and quantifiable approach to assessing a broad range of sensory, motor, and cognitive functions in individuals with primary brain tumours. Brain tumour patients were recruited from the neurosurgery clinics at Kingston General Hospital and Hotel Dieu. Feasibility was defined as successful study recruitment, high data capture rates, tolerable levels of fatigue, and identification of impairment by Kinarm assessment and correlation with an existing assessment tool (RBANS). Our assessment protocol included a traditional assessment (RBANS), a fatigue inventory (BFI), and a robot-based assessment (Kinarm). In a cohort of 16 brain tumour participants, we found that Kinarm-based assessment was capable of assessing neurological functioning. The robot identified impairment in at least one neurological domain in the majority of brain tumour participants (81.5%). Using Pearson correlation coefficients, concurrent validity was established between several Kinarm tasks and RBANS domains. Kinarm assessment also demonstrated that brain tumour participants often exhibited multiple deficits across a broad range of cognitive, sensory, and motor tasks, thus revealing a pattern of global neurological impairment. This finding was supported by strong and significant Kinarm inter-task correlations. Fatigue assessments showed mild levels of fatigue and data capture rates were high suggesting Kinarm-based assessment was tolerable. Participant recruitment rate however was low and resulted in our small sample size. In conclusion, we provide evidence that the Kinarm may be a feasible neurological assessment tool in brain tumour patients and may be used in future studies investigating neurological impairment in this population and the impact of novel therapeutic interventions.