Mild Cognitive Impairment (MCI) is a prevalent condition among older adults and can progress to dementia if not detected early. Neuropsychological tests, such as the Rey Complex Figure Test (RCFT), are widely used for early detection by evaluating visuospatial construction, memory, and executive functions. However, traditional manual scoring is time-consuming and subject to inter-rater variability. Recent advances in artificial intelligence (AI) and computer vision provide new opportunities for automated and objective evaluation. This study proposes and evaluates a fully automated system for scoring RCFT drawings, aiming to enhance diagnostic precision, reduce subjectivity, and facilitate large-scale cognitive screening. A dataset of RCFT drawings was digitized and analyzed using a combination of computer vision techniques and machine learning models. The system was designed to identify core structural elements, quantify accuracy and organizational strategies, and compute standardized scores aligned with test norms. Preliminary analyses suggest that the automated approach can approximate human ratings and highlight organizational strategies in the drawings. While performance varies across specific scoring domains, the system shows potential to reduce rater subjectivity and evaluation time. Early results also indicate that novel digital features, such as drawing complexity or error patterns, may provide complementary insights, although further validation with larger datasets is needed. Fully automated RCFT scoring offers a reliable, efficient, and scalable alternative to manual evaluation. By combining standardized assessment with novel digital biomarkers, this approach has the potential to support earlier and more accurate detection of MCI and related neurocognitive disorders, while streamlining both clinical and research applications.