In this study, we propose a new analytical method using cepstrum to estimate the propagation time of pulse waves. To test this method, we used a simulation model of the systemic circulatory system. In our model, the arterial tree was divided into 128 segments. Reflections of pulse waves were analyzed using cepstrum with and without cuff occlusion of the femoral artery. The simulation results show that forward and backward waves are clearly separated using cepstrum and that the propagation time of pulse waves can be estimated from the difference in the cepstrums with and without the cuff occlusion. The difference manifests itself principally as an impulsive wave. Therefore, the position can be recognized with high accuracy on quefurency. Commonly measurement of the propagation time of pulse wave between 2 sites of the systemic circulatory systemneeds two measuring devices located at each site. In our method, only a single measuring device is required to measurethe pulse propagation time. Applying the method to actual human blood pressure, we obtained results similar to those of the simulation.

View full abstract

The relation between a stimulus and a P300 component was investigated using a single stimulus paradigm and a compound stimulus paradigm. The single stimulus paradigm was designed to assess the relation between the intertarget interval (ITI) and the P300 component for an ITI of 3-10s. When the ITI was set long, P300 amplitude tended to increase. The change in P300 amplitude of an auditory evoked potential (AEP) was greater than that of a visual evoked potential (VEP), suggesting that the P300 of AEPs is more sensitive to the ITI. The compound stimulus paradigm was designed to examine the P300 component when two kinds of sensory stimuli (visual and auditory) were presented. Interstimulus intervals (ISIs) between the two different stimuli and between the same stimuli were set at five and ten seconds, respectively. The cumulative averaged waveforms showed a decreasing amplitude of event-related potentials in the following order: a single target was set and consistent with the corresponding sensory stimulus; both targets were set and each target was consistent with a sensory stimulus; a single target was set and each target was inconsistent with a sensory stimulus; and no target was set. In the compound stimulus paradigm, the VEP amplitude varied markedly with the target stimulus. The results suggest that VEPs are sensitive to the stimulus modality.

View full abstract

We have been developing a computerized system (“BULLsNET”) that can help the radiologist's diagnosis in the detection and classification of coronary artery diseases from myocardial SPECT bull's-eye stress images. We newly construct a system by adding a rest (redistribution) image analysis to the stress image analysis. The system consists of 2 major neural networks (NN) for recognition of 2 types of bull's-eye images. “Extent” and “Severity”, acquired in both stress and redistribution examinations. When the confidence level of the “Extent NN” is lower than 0.9, the “Severity NN”, in which each part of the coronary vessel regions in the bull's-eye plot is examined by three sub-neural networks, is applied. Recovery score is determined based on the area and intensity differences between SEVERITY images. Our new BULLsNET with the redistribution image analysis can classify three interpretations, ischemia and partial or total infarction. The performance of the redistribution image analysis was comparable to that of experienced radiologists. Our studies suggest that the neural network approach is useful for the computer-assisted interpretation of bull's-eye images. For further investigation to improve the BULLsNET, the stress image analysis will have to be upgraded.

View full abstract

We have developed an automated method for classification of masses in mammogram. The probability of malignancy was determined based on eight features extracted from mass shapes and analysis of mass spicules. The scheme consists of the following five steps: (1) Extraction of breast region, (2) Mass edge extraction, (3) Calculation of malignancy based on spicule patterns, (4) Extracting eight features from mass shapes and density, and (5) Classification by using artificial neural network (ANN). 599 cases including 17 masses with biopsy-proven were employed to estimate the performance of classification by the ANN. The scheme achieved a sensitivity of 82% at a specificity of 80%, and the area under the ROC curve (Az value) was 0.87. We conclude that our method is effective for discriminating mammographic masses.

View full abstract

In the previous study, the multi-element tonometry sensor-based noninvasive PWV measurement system was evaluated for its frequency response characteristics, and for its accuracy in pulse wave propagation time measurement in simulated circulatory system. Excellent results supported the use of this system in noninvasive measurement of PWV possibly from the carotid and femoral arteries. However, its accuracy has not been validated in an actual physiological system. Since the noninvasive pulse wave measurement is affected by the tissue between the sensor and the vessel, hold-down force applied to position the sensor, and physiological variables such as heart rate and blood pressure levels, further study was undertaken to quantify their effects upon pulse wave, hence PWV measurement. For the animal model, we used the common carotid artery of female goats whose body weight was close to average human size (60kg). The study was divided into two groups; in Group I the tonometric sensor was directly applied to the exposed left common carotid artery, while in Group II the sensor was applied over the skin and subcutaneous tissues covering the artery. As a control measurement, a 1.4F Millar catheter was inserted inside the carotid artery in the vicinity of the tonometric measurement site. Also, a 4F Millar catheter was inserted through a contra-lateral carotid artery and its tip was advanced to the root of the aorta for measurement of aortic pressure waveform. Both the Millar and tonometric measurements were referenced to the aortic waveform for computation of PWV. First, we changed the hold-down forces of the sensor applied to the artery. Second, systolic blood pressure levels were changed from 85 to 170mmHg by drugs. Third, heart rate was changed from 80 to 145 beats/minute by drugs and a pacemaker. The results showed an excellent correlation between the tonometric and Millar measurements with and without skin and subcutaneous tissues. The correlation coefficients between the Millar and tonometric methods for PWV measurements were 0.99 with and without skin and subcutaneous tissues. The bias ±2SD for the pulse transmission time between the two methods (the Millar minus tonometric) were -1.14±0.76ms for the exposed artery, while for the covered artery -0.87±0.56ms. These results confirmed that the arterial wall, subcutaneous tissue and skin do not affect the accuracy of pulse wave measurements under varying physiological conditions. It was thus concluded that the multi-element sensor could be used for effective and accurate noninvasive PWV measurement in vivo.

View full abstract

A miniature Clark-type oxygen sensor for long-time continuous monitoring of oxygen concentration was fabricated using micromachining technology. The cathode electrodes are formed on a silicon chip, which is mounted on the top of a catheter of 1mm diameter using an oxygen permeable tube. In order that the oxygen sensors can operate for a long time, the cathode electrodes are insulated from the electrolyte with an organic and an inorganic film, and the cathode and the anode electrodes are sufficiently spatially separated in the sensor. Since the anode electrode is set at the end of the catheter filled with electrolyte, it is connected to the cathode electrically through holes formed on the cathode chip. The combination of silicone rubber and a thin film of amorphous Teflon as an oxygen gas permeable membrane not only shows a low drift in more than 120 hours continuous operation, but also maintains a stable response and a constant sensitivity after a long-time use.

View full abstract

Optical imaging based on intrinsic signals was widely used in visualizing the functional organization of the cortex. However, in general, the evoked optical change is small compared to the background optical intensity. In this study, based on the well-known statistical method of t-test, we proposed a novel method to extract the neuronal activity-driven activation region. The feasibility of the proposed method is finally evaluated with signal cell electrode recording.

View full abstract

Diabetic neuropathy induces disorder in blood flow, hence it is possible to estimate the autonomic function by analyzing it. We divided subjects into two groups, normal N and diabetic neuropathy DN (+) groups. The blood flow was measured at the fingertips by a non-invasive laser Doppler measuring technique. We propose two methods to estimate autonomic function. The first method is to examine the blood flow fluctuation with component analysis. Thirty minutes after taking glucose, many normal subjects showed distinct amplitude increase in the frequency range between 0.05-0.15Hz which reflects sympathetic nerve activity, while average blood flow decreases. On the other hand, in DN (+) group, the increase was hardly observed. The second method is to investigate the relation between 1st and 2nd time derivatives of blood flow from the viewpoint of circulating control of blood flow. For the normal group, we observed the clear feature that there were many points which lay near zero for the 1st derivative value. The existence of these points means there exists a period for the normal group in which the proper blood flow is attained and no changes in flow are necessary. In the DN (+) group, we could not observe these characteristic points. From these two results obtained by these methods, we can distinguish DN (+) patients from normal subjects.

View full abstract

Gender identity disorder (GID) is a condition in which sex differs from gender identity. The basic methodology used to diagnose such a disorder involves examination by more than one psychiatrist. It is necessary to develop objective, quantitative criteria for diagnostics in order to improve the credibility of diagnosis, and to aid in decisions as to whether a patient is eligible for sex reassignment surgery. It has long been believed that there are sex-related differences in the shape and size of the corpus callosum (which interconnects the two cerebral hemispheres), and that these differences could form the basis of an objective, quantitative diagnostic technique. Unfortunately, there are, as yet, no established methods. In this study, we used MRI to investigate sex-related differences in the corpus callosum. We described the callosal figures at the midsagittal plane, using Fourier descriptors and central moments to represent the shape. We found that females have stooping corpus callosum relative to males, at a significance level of 10^-14. This suggests that sex-related differences in the corpus callosum may consist of differences in the callosal configurations in the cranium, rather than the shape and size of the corpus callosum. Furthermore, using statistical tests (a likelihood ratio test, Kruskal-Wallis test and Mann-Whitney test), we found that most of the GID subjects in our study (MTF, 4; FTM, 11) have a principal orientation of the corpus callosum that is closer to female than male in sex. This means that the callosal principal orientation may qualify as an objective, quantitative criterion for MTF diagnosis.

View full abstract