Publisher Summary Over the past few decades, cross-fertilization between the fields of neurophysiology, physics, mathematics, and computer science has helped in the complex efforts to understand the mechanisms used by the central nervous system and has provided a wealth of challenging theoretical problems. On the engineering side, neuronally inspired models and devices have proved themselves very useful in applied pattern recognition and robotics. Artificial neural nets are used commercially for complex technological processes and for providing rapid solutions to optimization problems. This chapter explains how very simple neuronal networks, composed of only a few hundred neurons, can be complicated and dynamically rich. The main challenge is to better control and understand this dynamics at the cellular and ensemble level with emphasis on the emerging network properties. It is also essential to supplement pharmacological manipulations with electrical stimulation of increasing pattern complexity at single electrodes and on a spatial pattern of electrodes. The observed richness of the dynamical repertoire of small neuronal networks serves a definite purpose in living systems, and such studies may become a possible source of inspiration for yet another generation of hardware devices and computing algorithms.

Paraplegic status in patient with spina bifida (myeloschisis) remains for the entire life. It is now becoming more apparent that there are two contradictory therapeutic goals to be selected; termination of the pregnancy under the prenatal diagnosis, or radical surgical intervention to reconstruct the neural structures. In early 1980s, clinical application of nerve anastomosis in patients with spina bifida was reported by two authors (Epstein F. et al and Oi, S. et al), namely “Intercostal nerve lumbar nerve root anastomosis”.

Publisher Summary This chapter discusses neural prosthetic interfaces between neural and electronic computing. Many diseases and injuries of the nervous system produce focal disabilities by interrupting sensory input or motor output while leaving intellectual function intact. Researchers in artificial intelligence and robotics have developed satisfactory electronic devices and signal processing algorithms to replace these peripheral functions. Neuroscientists have achieved a solid understanding of the nature of signal encoding and transmission in the afferent and efferent pathways of most systems, and biomedical engineers have succeeded in developing implantable electronic devices that are sophisticated, miniature, and biocompatible. For all of these reasons, many researchers are actively engaged in developing neural prosthetic devices to treat a wide range of disabilities including deafness, blindness, and paralysis. Central to all of these clinical applications is the need for highly parallel and often bidirectional communications between electronic circuitry and neural pathways. This presents fundamental problems in biophysics and biomaterials because of the differences in the way that electrical currents propagate as electron motion in solid-state conductors and semiconductors as compared to ionic flow through membranes and fluids.

Fundamental developments in feedforward artificial neural networks from the past thirty years are reviewed. The central theme of this paper is a description of the history, origination, operating characteristics, and basic theory of several supervised neural network training algorithms including the Perceptron rule, the LMS algorithm, three Madaline rules, and the backpropagation technique. These methods were developed independently, but with the perspective of history they can all be related to each other. The concept which underlies these algorithms is the “minimal disturbance principle”, which suggests that during training it is advisable to inject new information into a network in a manner which disturbs stored information to the smallest extent possible.

This chapter discusses electrical neurostimulation and neuromodulation of voiding dysfunctions and pelvic pain. Interest in the electrical control of bladder dysfunction began in the 1950s. Most reports concerned the stimulation of the detrusor directly or the spinal cord. Only a few discussed pelvic or sacral nerve stimulation. Direct pelvic nerve stimulation initially demonstrated encouraging results in animal experiments; however, in humans, parasympathetic innervation is unsuitable for direct electrode application. Direct detrusor stimulation, which has the advantage of easy electrode placement and high specificity of target organ, was unsuccessful because of fibrosis, and electrode malfunction from bladder movement and bladder erosion. External surface electrodes and implantable electrodes to stimulate the spinal cord failed to achieve bladder emptying. Although the precise mechanism of normal bladder control remains unclear, it is generally accepted that normal micturition depends on a balance between excitatory and inhibitory descending pathways and feedback from afferents in the effector organs bladder and urethra. Electrostimulation has been applied through various approaches: externally by perineal needles, surface electrodes, electric pessaries and anal plugs, and internally by electrode implants in the pelvic floor muscles.

Conventional methods for control of dynamical systems require knowledge of the system in analytical form. In practical applications this knowledge is mostly available only in empirical form by taking measurements. Neural networks are the appropriate tools for dealing with such empirical information. This paper demonstrates the capabilities of neural networks in control. After giving an impression about neural techniques relevant for control problems we present simulation results of a robot arm path tracking system with a neural inverse model. The relevance of our investigations for neurobionics is pointed out.

The emerging field of neurotechnology deals with compensation of functional deficits of the human nervous system by means of novel computer and microsystems technologies. Neurotechnology requires the development of adaptive pulse-processing neural computers and novel multicontact neural interfaces to allow for functional substitution of and bidirectional communication with specific parts of the nervous system. Adaptive pulse-processing neural computers for neuroprosthetic applications have to serve a number of specific functions in different environments. Representative input or output indicates that only a small fraction of the nerve fibers of the contacted nerve or fiber tract will become available for bidirectional communication via the single-axon contacts of a multicontact neural interface. This chapter presents some theses that describe the reasons to start a joint effort in the field of neurotechnology.

Mind and brain were historically analyzed by different research programs. In the tradition of classical Al brain was understood as computer hardware of the most advanced machinery, while mind is the corresponding software program with deterministic algorithms. Even knowledge-based expert systems are conceived by algorithmical representations of highly developed Al programming languages. But theoretical results of mathematical logics (Church, Turing, Godel, etc.) and practical problems of programming limit the mechanization of thought! Computational neuroscience promises a way-out. Several concepts of research such as parallelism, non-linearity, selforganization, and complex systems are merged in the new paradigm of connectionism which gives new insight in the philosophy of human nature and mind. But research must be supplemented by the responsibility for maintaining the health of brain and mind. Thus, neurobionics become a fertile interdisciplinary research program with ethical and anthropological dimensions. Historically, it is shown how this program can be understood in the long tradition of philosophy of nature and epistemology.

Publisher Summary Structure formation in nature is often based on simple interaction between spatially distributed basic units. To deal with structure formation, it is necessary to understand the relation between its local and global aspects. According to John von Neumann's definition, a cellular automaton consists of a finite number of cells that may be depicted as, for instance, the various squares in a planar checkerboard pattern. To each of these cells, a finite number of states are ascribed among which the cell is allowed to vary. Given a configuration of states, one for each cell, the next configuration of states is determined according to local interaction rules. This chapter describes the approach of cellular automata, which can be used to find out which particular feature of the dynamics of a complex molecular system may be responsible for the observed formation of particular patterns or structures.

Neurosurgery as a medical discipline, responsible for maintaining the health of the biological carrier (=brain) of the human mind, thus hovers between ethics as a philosophical discipline, basic natural science disciplines, such as computational neuroscience, and the clinical application of this knowledge in the case of the patient suffering from illness.