Many neurological diseases cause motor impairments that limit autonomy and reduce health-related quality of life. Upper-limb motor impairments, in particular, significantly hamper the performance of essential activities of daily living, such as eating, bathing, and changing clothing. Assessment of impairment is necessary for tracking disease progression, measuring the efficacy of interventions, and informing clinical decision making. Impairment is currently assessed by trained clinicians using semi-quantitative rating scales that are limited by their reliance on subjective, visual assessments. Furthermore, existing scales are often burdensome to administer and do not capture patients' motor performance in home and community settings, resulting in a severely under-sampled view of patients' conditions. Quantitative, objective assessment of upper-limb impairment outside clinical settings could address these limitations, but existing technological solutions generally impose a variety of practical burdens on patients, such as a need to wear many sensors or regularly perform a tightly controlled set of motor tasks. This dissertation first presents data analytic methods that exploit how the central nervous system plans voluntary movements and demonstrates, in controlled settings, that analysis of upper-limb movements can yield information relevant to upper-limb impairment in stroke survivors and patients with ataxia. Fully leveraging these promising findings, this work then further refines and validates these data analytic methods towards the goal of seamless monitoring and assessment of upper-limb function in stroke survivors using only inertial data obtained from patients' natural activities of daily living and a single wrist-worn sensor. This work ultimately aims to support a paradigm shift in how motor impairment is assessed, in which fine-grained and longitudinal tracking of disease progression will enable personalized rehabilitation regimens to optimize therapeutic interventions and promote patient-centric care.