A prototype of automated blood cell analyzer has been developed. The process is performed by the pattern recognition method.
Four types of examinations are automatically operated :
1) differential count of normal and abnormal white blood cells,
2) morphological analysis of red blood cells,
3) numerical platelet estimates and
4) reticulocyte count (a slide preparation differing from other three).
The WBC differential count is the essential of the process and applies an adaptive and hierarchical classifier.fw Here, the word “adaptive” means self compensation using standard cells in each sample. The standard cells are automatically picked up from input cells.
The WBC differential count is processed at the rate of one sample per minute. Net discriminating rate of normal WBC differential count is better than 98%.

View full abstract

This paper, presents a method of quantitatively estimating the Achilles tendon reflex.
The foot motion induced by an electrical impulse was picked up by an accelerometer attached to the sole and then written on a recorder. The optimum parameters at the measurements were determined experimentally.
The acceleration waveform was a damped oscillation. The degree of the reflex was estimated with the ratio of two successive peak to peak values multiplied by its period in the waveform. The values in normal human beings were in the range of 5 to 9 and those of affected side of hemiplegic patients were above 8. These values showed fairly close correlation with the results obtained by physical examinations.
The validity of the method in this paper was ascertained qualitatively by a computer simulation. Only spindles on the same sides as the aiming muscles were considered as receptors. All the organs in the foot motion system were represented by simple models.

View full abstract

In this study, a neural network model is proposed to obtain -the numerical description of pain mechanisms. The model simulates peripheral receptors, afferent L- and S-fibers and the receptive cells of the spinal cord area Adaptation effect and conduction velocity of each fiber are considered in the model. The activities of peripheral and spinal cord neural cells are represented by the Wilson-Cowan's differential equation, considering the ongoing activity of neurons. Pain and touch sensibilities are estimated by the firing activities of lamina V (Transmission, T) cells and lamina IV cells, respectively.
The results of model simulation have been obtained for single square-wave pulse and periodic pulse sequences applied on peripheral receptors. The duration of initial bursts of T and IV cells, which occur soon after the beginning of stimulation, depends on the stimulus intensity and approaches a plateau with increasing intensity. In case the stimulus intensity is increased above the threshold of S-fibers, secondary burst of T cells appears about 50 msec after the stimulation, which is equivalent to the conduction time of S-fibers from the periphery towards the spinal cord cells. High-frequency and high-intensity stimulation gives the typical firing patterns of pain modality, represented by the high secondary burst of T cells which continues during the stimulation. From the results of periodic pulse stimulation, the so-called pain and absolute threshold characteristics have been obtained and have been compared with the results of cutaneous electrical stimulation. Both results are shown by quite a similar relationship between stimulus pulse frequency and stimulus intensity and their characteristics are given by the power function's law of frequency and intensity.
The proposed neural network model mimics the pain modality very well, and the results of simulation are in good agreement with some of physiological and experimental results.

View full abstract