Combining Optical Imaging and Computational Modeling to Analyze Structure and Function of Living Neurons

We are investigating the computational properties of principal neurons in the mammalian brain. To manage the small size and intricate structure of neuronal dendrites, we employ advanced optical imaging techniques in combination with automatic image reconstruction and computational modeling to study their complex spatio-temporal pattern of activity. I. INTRODUCTION HE mammalian brain contains some 10 10 individual neurons that are highly inter-connected, each receiving input from somewhere between 10 2 and 10 5 other neurons. It is currently well-accepted that the information processing capability of the brain is not only based on the large number of neurons and their high connectivity, but also on the performance of individual neurons. Particularly, our view of the neuronal dendrite has changed from that of electric cables, passively conducting postsynaptic signals to the cell body, to that of active neuronal structures, capable of complex spatio-temporal electrochemical dynamics. Therefore, there exists a keen interest in understanding the information processing capability of a single neuron. The structural and functional analysis of individual neurons is complicated by their intricate shapes, small sizes, and fast response times. Traditionally, structure was observed and documented with histological techniques that require fixing, staining and physical micro-sectioning of brain tissue. Function was monitored with micro-pipettes; for technical reasons, their access was mainly limited to the cell body or the proximal dendrite. Today, sophisticated optical imaging techniques are increasingly employed for studying neurons, replacing these invasive approaches. The strong light scattering properties of living brain tissue