Computer diagnosis or numerical diagnosis begins from the criticism on the feature of the medical system at present. Medical information at present is given as a system of deductive description and mostly qualitative ones. As a fact, however, disease profiles in one disease have numerous faces and one disease profile corresponds to many diseases. Therefore the process of diagnostic thinking can not be deterministic, but probabilistic. In numerical diagnosis a disease profile is dealt as a multi-variate and evaluated by the statistical measure called likelihood. In the field of the estimation of the name of the diseases the following will be not worthy : researches on cerebellar tumors by Takahashi et al. (Univ. of Tokyo), brain tumors in cortex by Domae et al. (ibid), apoplexia by Fukui et al. (ibid), traumatic injury of the head by Sano et al. (ibid), heart diseases by Dohi et al. (ibid), acute abdomen dy Shikata (Sumida Hospital), liver diseases by Kosaka et al. (Okayama Univ.) and so on. In the field of pattern recognition of laboratory data electrocardiogram by Okajima et al. (Nagoya Univ.), by Kimura (Nippon Med. College), by Matsuo (Tohoku Univ.) and by others, phonocardiogram by Yoshimura (Jikei Med. College) and by Machii (Univ. of Tokyo), findings of gastrocamera by Hirokado et al. (Kyushu Univ.) and so on. In most cases numerical diagnosis gives more excellent result than doctors'.
In most of these researches multivariate statistics are applied. Boolean algebra and Bayesian probability, introduced by American researchers Ledley and Lusted, revealed that these are not adequate as a measure of diagnostic inference, either theoretically or experimentally.
Though there is no doubt on the effectiveness of numerical diagnosis, probability itself has no final meaning. It is accepted as one of the elements of medical decision theory. Another field to be developed in future is the estimation of the prognosis.

View full abstract

There are not a few cases in which the existence of the fetal waves can hardly be confirmed because of mixture of noises interfering minute amplitudes of them, but still in such a case the confirmation is seriously demanded clinically. The clinicians face a great difficulty, when the fetal waves are not surely identified, e. g., when they are not apparent either due to the fetal death or to the poor recording system. Therefore, the authors applied the mathematical management to such a hardly understandable recording and obtained better results in the detection of the fetal waves. The first method which the authors have adopted is as follows the soike waves are all marked and 10 sheets of the copies are made. Each of them is set in parallel to each other, with the horizontal deviation of each head of the copies by the time interval of ordinary fetal heart beats. One can see if there appears the dense, vertical lineup of the marks, that is, the autocorrelation is firstly sought and then the existence of the fetal waves is determined. This method is simply carried out only with pencil and sheets of paper. The second method is to find out the fetal waves by electronic computer automatically ; the fetal E. C. G. is put into the data recording, AD converted and moving averages diagrams for each 8 or 32 are made. From the difference of the diagrams the fetal, spike waves are sought. The autocorrelation is obtained and then as in the first method the existence of the fetal waves is determined. These three processings can be performed with the medium hybrid electronic computer, at relatively low expense.

View full abstract

Computer diagnosis of the roentgenogram is regarded as a typical two-dimensional pattern recognition problem. As the first step to approach the problem, we investigated the characteristics of the density distribution of the chest roentgenograms. Then we studied the method of automatic recognition of rib boundaries.
As the next step, we studied in this paper the method of automatic recognition of abnormal shadows in chest roentgenogram.
First, simple models of abnormal shadows in chest roentgenogram consisting of (1) some irregular figures whose uniform densities are known and whose shapes, sizes and positions are unknown, and (2) additive Gaussian noise with 0 mean and known variance, are assumed and recognition method is derived. Only outline of the method is stated here.
Next, this method is applied to the recognition of abnormal shadows in the real chest roentgenogram. In this experiment, input patterns (roentgenograms) are processed according to the following procedure : (i) measurement and sampling of film density, (ii) recognition of boundaries of dorsal portion of ribs, (iii) subtraction of density corresponding to dorsal portion of ribs, (iv) equalization of average density level in the horizontal direction, (v) recognition of boundaries of ventral portion of ribs, (vi) subtraction of density corresponding to ventral portion of ribs, (vii) recognition of abnormal shadows (tuberculous lesions).
Results of the experiment are stated here in detail. They show that the method used in this paper was proved useful for detection of some kinds of abnormal shadows in chest roentgenograms.

View full abstract

This paper describes a basic approach to automatic measurements of ECG waves. In this field of study, some results have been reported by parameter several groups, but considering from the technological point of view, these are not necessarily obtained through. appropriate processing. In computer recognition of ECG waves the effect of line noise and myoelectric signals must be eliminated, and the sampling period and the degree of quantization should be suitably chosen.
We had processed ECG waves by using Fourier analysis, and on the basis of the result we designed several filters to eliminate the harmful noises. The quantizing error in the above mentioned processing was discussed in the present paper. As the result, we have obtained more exact values of the parameters such as PQ interval and R magnitude useful for the heart disease diagnosis. Power spectra of ECG waves were also found to be useful. They could be classified into six typical patterns and this classification will be helpful in the diagnosis for heart diseases.

View full abstract

In this paper A simple method for calculation of cardiac output and mean transit time from dye-dilution curve was investigated with the use of gamma distribution function
An approximation derived from the property of the function gave the following formulae :
cardiac output=_??_
mean transit time=T_p+1.1× (T2-T₁) (sec)
Where, I : amount of dye injected, C_p : peak concentration, T_p : time of peak concentration, T₁andT₂ : forward and backward parts of half width, respectively.
The values calculated with these formulae were compared with those calculated with Hamilton's method for 51 recordings of 16 patients without arterio-venous shunting or regurgitation in the heart. The discrepancies of calculated cardiac outputs and mean transit times with the two methods were less than 2.1% and 3.2%, respectively.

View full abstract

An ECG monitoring device with the use of audiofrequency modulation is proposed. In this device, ECG wave form is modulated in pulse frequency and converted into symmetric square wave through flip-flop circuit. The frequency corresponding to the base line of ECG is chosen between 400 to 800 Hz and maximum sensitivity is about 300 Hz/mV. Experimental results indicate that several features of ECG wave form such as arrhythmias, P or T wave inversion, S-T segment depression and extrasystoles can easily be differentiated.
This method is advantageously applicable not only to a long-term ECG monitoring but to portable ECG device or to ECG transmission through conventional telephone lines.

View full abstract

To avoid the artefact of electric stimulation, some devices of isolation have been used in most bioelectric work. These include a simple transformer isolation and a radio frequency isolation, and the latter is now most widely used. The isolation of the transformer method is not complete and can not transmit square waves. The radio frequency method has a complete isolation and can pass square waves, but the circuit is rather complex. We devised a photo-isolation method and applied it to an all-transistorized stimulator. A neon tube is lighted by a trigger circuit, the light is transmitted from the tube to a photo-diode, and the change in the diode resistance triggers the following monostable-multivibrator, in which the width of the pulse is set. The pulse is amplified by the final transistor, variably attenuated and led to output terminals.
The maximum output voltage is almost the same as that of the battery supplying the final transistor. All batteries of the circuit after the photo-diode are floating from the earth and separated from those of the trigger circuit. The isolation is accomplished between the neon tube and the photo-diode. The capacitance between the output and the earth is about 7 pF and is in the same order as an ordinary isolator. Only one defect is a very small leak DC voltage from the output, derived from the dark current of the final transistor. It is about 4 mV in our device. But it can be decreased if a silicon transistor is used in the final one, and it can be neglected if a condenser with enough capacitance to pass the used square pulses is inserted in series with the output. Therefore, the present device is very likely to be used instead of the ordinary radio frequency method.

View full abstract

This paper deals with the practical applications of the mutual induction principle for the simultaneous measurements of the static and dynamic displacements (l₀±Δl) of the soft body tissuein vivo, such as heart, vessel, intestine, uterus, respiratory system and etc. The advantages of this method are based on the qualitative detection of displacements by tiny and light weighed couple of coils pasted on the surface of the tissuein vivo. One of the paired coil behaves as a transmitting coil operated around crystal controlled 450 kc and the amplitude modulated signal by distance is received by another coil and reproduced into the DC distance signal through IFT circuit. The results demonstrated stable and linear characteristics (± 1% error, i. e. less than 10 micron in distance) with high conversion ratio (4 volts/mm). Practically the pressure-diameter relationships of the vessel and heart wall, peristaltic movements of the intestine or uterus, and respiratory excursion of chest wall are continuously recorded. Three dimensional measurements are done by using three different exciting frequencies, namely 400, 450 and 500 kc into three couples of coils without any interferences among them.

View full abstract