A double scanning microphotometer for image analysis: hardware, software and biomedical applications.

TLDR

The authors present a new image processing system for densitometry and pattern analysis of microscopic specimens, integrating hardware, software, and biomedical applications. The system acquires data by motorized stage scanning and photomultiplier digitization, then processes images in a single pass using automata theory to avoid memory storage. This approach allows continuous, high‑magnification imaging of entire preparations and enables automatic identification and counting of cells in different mitotic phases.

Abstract

A new image processing system designed for densitometry and pattern analysis of microscopic specimens is described with special regard to the hardware, the software and the biologic applications. The data acquisition procedure involves the combination between the scanning of the preparation by means of a motorized stage and the scanning of successive fields by a mechanical device. The signal provided by the photomultiplier is converted into digital values which are directed to an on-line computer. The data processing is based on a one-pass computation involving automata theory and therefore it avoids the storage of the image in the computer memory. In so doing, an entire and continuous image of the whole preparation can be processed at the highest magnification of the microscope whatever the size of the analyzed specimen may be. A biologic application of the system is reported and concerns the automatic identification and counting of cells in the various phases of the mitotic cycle.