Leon Glass is an American scientist who has studied various aspects of the application of mathematical and physical methods to biology, with special interest in vision, cardiac arrhythmia, and genetic networks.
Glass' early work and eponymous patterns were fostered by mentor Christopher Longuet-Higgins, who guided him in the application of statistical methods to visual perception. Glass patterns are formed from superimposed random dot patterns: an original image with a second image which has been generated through a linear or nonlinear transformation of the original. A variety of different spatial patterns such as circles, spirals, hyperbolae, can be perceived in the superimposed image set, depending on the nature of the transformation between the two sets of dots. This discovery provided insight into mathematical nature of human perception by suggesting that the visual cortex is capable of computing a large number of autocorrelations in parallel. David Marr first coined the term "Glass Patterns" in his 1982 work on visual perception, resulting in an increased interest in the phenomenon. Because of their mathematical simplicity and physiological underpinnings, Glass patterns have subsequently been used in dozens of electrophysiology and visual psychophysics experiments, resulting in additional understanding of the physiology of visual perception. Glass may be best known for his work with colleagues at McGill University, suggesting that certain physiological disorders may be considered dynamical diseases. These are characterized by sudden changes in the qualitative dynamics of a physiological control mechanism, which leads to disease. These features are illustrated in the Mackey-Glass equation. According to James Gleick, who recounted conversations with Glass in his book, foundational work inchaos by the McGill group was performed using animal models. He quotes Glass saying: "Many different rhythms can be established between a stimulus and a little piece of chicken heart". Since the initial description of dynamical diseases, a large number of researchers have analyzed mathematical models of physiological systems. Examples of dynamical diseases have been described in medical fields as diverse as hematology, cardiology, neurology, and psychiatry. Dynamical disease modeling has been used to understand cardiac arrhythmia, and specific model detection algorithms are now being programmed into pacemakers so that pathological patterns can be detected and corrected.
