Detecting Preclinical Cognitive Change

The Clock Drawing Test — a simple pencil and paper test — has been used for more than 50 years as a screening tool to differentiate healthy individuals from those with cognitive impairment, and has proven useful in helping to diagnose cognitive dysfunction associated with neurological disorders such as Alzheimer’s disease, Parkinson’s disease, and other dementias and conditions. For nearly a decade we have been administering the test using a digitizing ballpoint pen that reports its position with considerable spatial and temporal precision. We developed a methodology that analyzes the pen stroke data from these drawings and computed a large collection of features (statistics) of the drawings. We applied a variety of machine learning techniques to the data, producing interpretable predictive and prescriptive models. The resulting scoring systems were designed to be as easy to use as scoring systems currently used by clinicians, but more accurate. These new tests do not rely on doctors’ subjective judgment, unlike the current (non-digitized) clock drawing test. We evaluated our new models with a variety of techniques, including operationalizing a number of widely used manual scoring systems so that we could use them as benchmarks. We explored the tradeoffs in performance and interpretability in classifiers built using a number of different subsets of these features and a variety of different machine learning techniques. While more work is required for FDA approval, the work we have done offers the possibility of substantial improvement in detecting cognitive impairment earlier than currently possible. Given that the costs incurred for just a single form of dementia — Alzheimer’s Disease — are currently $200 billion per year, and are projected to reach one trillion dollars annually by 2050, early detection of impairment will have considerable medical, sociological, and economic impact.