AUTOMATED IMAGE-ANALYSIS TECHNOLOGIES FOR BIOLOGICAL 3D LIGHT-MICROSCOPY

Techniques for three-dimensional (3D) light microsocpy of a wide varie ty of biological specimens are rapidly maturing. The next advances are to improve image contrast, make image resolution as isotropic as poss ible, and perform quantitative analysis. This article discusses method s of automatic image analysis and blind deconvolution to compensate fo r the microscope's point-spread function (impulse response). Special e mphasis is given to quantitative analysis based on adaptive segmentati on, the results of which can be used to quantify and trace individual structures, and to montage high-resolution fields as well as the resul ts of image analysis on these fields into a wide-area view with an ass ociated computer database representation. We exploit the specific adva ntages accruing from the 3D nature of the data to achieve new capabili ties not possible with 2D imaging. Fundamental and practical challenge s in this area and our progress to date including automated segmentati on, cell counting, neuron tracing, mosaic synthesis, ''blind'' deconvo lution of large 3D images, and high-speed computation are presented. ( C) 1997 John Wiley & Sons, Inc.