Yasunari Yokota

Callosal Shapes at the Midsagittal Plane: MRI Differences of Normal Males, Normal Females, and GID

We investigated differences in corpus callosum shape at the midsagittal plane using MRI for different subjects: normal males, normal females, and subjects with gender identity disorder (GID). We first represented callosal shapes with Fourier descriptors of callosal contours. Using linear support vector machines with soft-margin, we next determined a hyperplane that separates normal males and females most optimally in the vector space spanned by Fourier descriptors. We then proposed a measure that has prominent sex difference: it is defined as the coordinate of a given callosal shape on the subspace orthogonal to the obtained hyperplane. Use of the measure provides discrimination of someones sex with 74.17% accuracy. We further showed that the value of the measure for GID more strongly reflects their mental sex, i.e. gender, than their physical sex

Analysis of measured values of ejaculation time in healthy males

Although it has been deemed irrelevant to address ejaculation time in terms of mean values, our study was designed as a preliminary step in determining the normal range of ejaculation time (from the start of stimulation of the erect penis to ejaculation) as a criterion for assessing the degree of ejaculation disorder. Ejaculation experiments were performed with informed and consenting healthy volunteers about 20 years of age, using identical manual stimulation by the same woman, and ejaculation time was measured. The mean +/- standard deviation for the ejaculation time was 156.5 +/- 80.7 seconds, which was shown to be erection time dependent.

An approximate method for Bayesian entropy estimation for a discrete random variable

This article proposes an approximated Bayesian entropy estimator for a discrete random variable. An entropy estimator that achieves least square error is obtained through Bayesian estimation of the occurrence probabilities of each value taken by the discrete random variable. This Bayesian entropy estimator requires large amount of calculation cost if the random variable takes numerous sorts of values. Therefore, the present article proposes a practical method for calculating an Bayesian entropy estimate; the proposed method utilizes approximation of the entropy function by a truncated Taylor series. Numerical experiments demonstrate that the proposed entropy estimation method improves estimation precision of entropy remarkably in comparison to the conventional entropy estimation method.

Eye movement inhibits the facilitation of perceptual filling-in

When a small figure is presented in human peripheral vision, it becomes invisible and invaded by surrounding texture, within a few seconds. This visual illusion is called perceptual filling-in. Time to filling-in (filling-in time) is varied by the properties of small figure, surround texture and some experimental conditions. In our preliminary study (Yokota, IEEE/IC-EMBS 2005), we found that incomplete fixation distributes filling-in time. Furthermore, that we can see nothing by restraining eye movement artificially is well known. Therefore, we can consider that filling-in time is influenced by eye movement. Although it has been recently reported that eye movement influences the filling-in occurrence (Martinez-Conde, Neuron 2006), the relation between eye movement and the filling-in time has rarely been reported. For this study, we measured the filling-in time for three subjects, for four surrounding textures, with simultaneous recording of eye movement. The results show that the filling-in time correlates to the standard deviation of the power of the eye distance from the fixation point. Furthermore, we found relatively strong correlation between the filling-in time and the power of high frequency component 50–200 (Hz) in the eye movement, though the correlation of the power of low frequency component 10–50 (Hz) is not so high. Thus we suppose that filling-in is inhibited by small involuntary eye movement.

Estimation of carotid diameter with heartbeat on longitudinal B-mode ultrasonic images

Arteriosclerosis, i.e. degrading arterial elasticity, causes cerebral infarction and myocardial infarction. Premonition of arteriosclerosis is important because arteriosclerosis is never cured completely. We therefore propose a method to track the carotid diameter to evaluate the initial carotid elasticity accompanied with a persons heartbeat using the longitudinal B-mode ultrasonic images. If the diameter of carotid artery can be accurately estimated, it is possible to implies for initial inference of the presence of arteriosclerosis. In future studies, it has to be assessed with comparisons to anaother established method.

Estimation of Carotid Stiffness Using Ultrasonic Dynamic Images for Evaluating the Degree of Arteriosclerosis

Some methods for evaluating the degree of arteriosclerosis utilizing the magnitude of systaltic movement of carotid artery have been proposed. The magnitude of carotid movement depends on the stiffness of surrounding tissues around carotid artery as well as the stiffness of carotid artery itself. Therefore, accurate criterion for the degree of arteriosclerosis cannot be obtained only by observing the magnitude of carotid movement. This study proposes an estimation method of the “self-”stiffness of carotid artery by excluding influence of the stiffness of surrounding tissues. In the proposed method, the systaltic movement of surrounding tissues as well as carotid artery is tracked by an optical flow method using ultrasonic B-mode images drawing the cross section of carotid artery.

Propagation Route Estimation of Heart Sound through Simultaneous Multi-site Recording on the Chest Wall

While vibration source of the heart sound has been well-studied, the propagation route of the heart sound on the chest wall has not been clarified. In this study, we first measured the heat sound simultaneously at 64 positions on the chest wall of healthy subjects using 64 small accelerometers. We next calculated the delay times in the vibration propagation between all possible pair of neighboring positions. We proposed the estimation method for the propagation route of heart sound vibration on the chest wall by solving the shortest path problem in the graph-network theory only using such neighboring delay times. Applying to the recorded heart sound signals, we can obtain the estimated propagation route originated at the cardiac apex for the first heart sound, and that originated at aortic valve for the second heart sound.

An optimal entropy estimator for discrete random variables

This paper presents analytical formulations of the most important estimation errors-averaged squared bias error and mean squared error - for the class of entropy estimator expressed as a sum of single variable functions. The class of entropy estimator includes almost all important entropy estimators that have been proposed heretofore. Furthermore, this paper presents an optimal entropy estimator that can minimize mean squared error of the estimate under the condition that averaged squared bias error of the estimate is restricted to below an arbitrary value. A numerical experiment demonstrates that the proposed entropy estimator provides a lower mean squared error than conventional entropy estimators when entropy is estimated as an ensemble mean over plural entropy estimates obtained for different independent data. Such estimation is often utilized for biological signals, e.g., neural signals, because of biological tiredness and adaptation property.

The effect of spatio-temporal frequency of dynamic texture on perceptual filling-in

Humans do not see scotomas shaded by the optical disk and blood vessels on retina. Texture of surrounding scotomas fills in to occupy the scotoma area. Similarly, when viewing a monochromatic figure surrounding dynamic textured background in monocular peripheral vision, background texture appears to fill into the figure. To investigate spatial and temporal frequency characteristics of the filling-in process, we measured time to filling-in when a gray figure surrounding dynamic random-dot texture with limited bandwidth of spatiotemporal frequency was presented in peripheral vision of subjects. Based on subject reports, a model of filling-in mechanism is proposed. The model is represented by the spatial and temporal perceptual power of dynamic texture. Model validity is verified by applying measured data to the model.

Premonitory Symptom of Septic Shock in Heart Rate Variability

Delay of administering therapy increases the mortality of septic patients. Diagnosing sepsis, however, depends on the decision of an experienced specialist of whether to perform a blood test. Therefore, we believe that a real-time, non-invasive monitoring system for sepsis is useful to promote early intervention. Heart rate variability (HRV), a measure of autonomic nervous system (ANS) activity, has been suggested as an indicator of septic shock. When analyzing electrocardiograms of patients admitted in the intensive care unit (ICU), an important challenge is to distinguish HRV originating in the ANS from other factors such as arrhythmia. A stochastic model is used in this study to extract HRV caused by ANS automatically. Applying the proposed process before HRV estimation, we identified a distinctive V-shaped temporal pattern in HRV as a signal of sepsis. Our investigation continues to explain how that temporal pattern is useful to develop a real-time sepsis monitoring system for sepsis occurrence.