A new feature extraction method for signal classification applied to cat spinal cord signals

In the spinal cord of the anesthetized cat, spontaneous cord dorsum potentials (CDPs) appear synchronously along the lumbo-sacral segments. These CDPs have different shapes and magnitudes. Previous work has indicated that some CDPs appear to be specially associated with the activation of spinal pathways that lead to primary afferent depolarization and presynaptic inhibition. Visual detection and classification of these CDPs provides relevant information on the functional organization of the neural networks involved in the control of sensory information and allows the characterization of the changes produced by acute nerve and spinal lesions. We now present a novel feature extraction approach for signal classification, applied to CDP detection. The method is based on an intuitive procedure. We first remove by convolution the noise from the CDPs recorded in each given spinal segment. Then, we assign a coefficient for each main local maximum of the signal using its amplitude and distance to the most important maximum of the signal. These coefficients will be the input for the subsequent classification algorithm. In particular, we employ gradient boosting classification trees. This combination of approaches allows a faster and more accurate discrimination of CDPs than is obtained by other methods.

The neocortical column

In the middle of the twentieth century, Rafael Lorente de Nó (1902?1990) introduced the fundamental concept of the ?elementary cortical unit of operation,? proposing that the cerebral cortex is formed of small cylinders containing vertical chains of neurons (Lorente de Nó, 1933, 1938). On the basis of this idea, the hypothesis was later developed of the columnar organization of the cerebral cortex, primarily following the physiological and anatomical studies of Vernon Mountcastle, David Hubel, Torsten Wiesel, János Szentágothai, Ted Jones, and Pasko Rakic (for a review of these early studies, see Mountcastle, 1998). The columnar organization hypothesis is currently the most widely adopted to explain the cortical processing of information, making its study of potential interest to any researcher interested in this tissue, both in a healthy and pathological state. However, it is frequently remarked that the nomenclature surrounding this hypothesis often generates problems, as the term ?Column? is used freely and promiscuously to refer to multiple, distinguishable entities, such as cellular or dendritic minicolumns or afferent macrocolumns, with respective diameters of menor que50 and 200?500 ?m. Another problem is the degree to which classical criteria may need to be modified (shared response properties, shared input, and common output) and if so, how. Moreover, similar problems arise when we consider the need to define area-specific and species-specific variations. Finally, and what is more an ultimate goal than a problem, it is still necessary to achieve a better fundamental understanding of what columns are and how they are used in cortical processes. Accordingly, it is now very important to translate recent technical advances and new findings in the neurosciences into practical applications for neuroscientists, clinicians, and for those interested in comparative anatomy and brain evolution.