A new computational measure for detection of extrapyramidal symptoms

Extrapyramidal symptoms (EPS) associated with antipsychotic medication can provoke severe discomfort and disability. Some instruments have been proposed to assess EPS, but they are not widely used in clinical settings. This paper aims to provide a translational tool for the detection of EPS: lacunarity of patients’ handwritings. Lacunarity is a measure, taken from fractals studies, that describes the spatial complexity of an image. 63 patients and 50 controls participated in the study. Patients were divided into: patients under typical antipsychotics, under atypical antipsychotics, and without antipsychotics. Participants were asked to write down a story. The texts were binarized and lacunarity was calculated. Results showed higher heterogeneity in handwritings from all patients groups, relative to the control group. Moreover, handwritings from the patients who were on typicals showed a significantly higher lacunarity than handwritings from patients on atypicals. The lacunarity of written texts appears to be a promising measure for the detection and quantification of EPS.