小橋 昌司

There are various cerebral diseases that deform the cerebral shape with region specificity. So it is effective to quantify the deformation change of cerebral gyri. This study introduces new index called gyral deformation index (GDI) that is defined as a ratio of area of gyrus of interest to area of cerebrum in the defined projection plane. To calculate the gyral areas, this paper proposes a gyral labeling method in the projection plane using magnetic resonance images. The new method finds the boundaries between the gyri by optimizing deformable boundary models aided by fuzzy logic. The proposed method was applied to quantify, the cerebral deformation of infants on a plane which is perpendicular to the longitudinal fissure. The comparison results with the manual delineation showed that the method labels gyri with a mean sensitivity of 92.8% and a mean false positive rate Of 0.1% for 14 infantile subjects (3 weeks - 4 years 3 months old).

In this paper, we propose an approach to extract features of center of foot pressure (COP) obtained by a load distribution sensor and apply this method to develop a biometrics personal identification system. Biometrics technology, as a method of personal identification, plays an important role in our daily lives. In our experiment, we have a user stand on load distribution sensor with slipper, and acquire pressure data during a simple motion, as touching a bell nearby by one hand but without movements of feet. We propose a biometrics personal identification system with less information, time and low space. First, we calculate the site of COP from the obtained pressure data. Features for identification are extracted from the position and the movement of COP. Second, we built a k-out-of-n system and a neural network (NN) model with the feature parameter. Third, we input test data to the two systems. Finally, we give a comparison of these two methods. We employ 11 volunteers. The experimental result reveals that the proposed identification method can achieve an accuracy of 12.0% in FRR (False Rejection Rate) and 1.0% in FAR (False Acceptance Rate).

Recently, the increase of core burden due to the increase of number of the elderly dementia patients is a matter Of concern in Japan. However, dementia of the elderly tends to be wrongly recognized as the effect of aging, and there are many, cases in which early detection are difficult. In this paper, we focus on the cognitive impairment as One of the core symptoms of dementia, and propose the fuzzy estimation system to detect the level of dementia through monitoring the participants' sleep using air pressure and ultrasonic sensor systems which were developed by our laboratory As a result of applying this method to twenty-three women in a nursing home, we could confirm the high correlation between the degree of dementia and the truth value, the score of Revised Hasegawa's dementia scale.

Wearable joint kinematics monitoring is effective for analyzing daily human behavior, for evaluating the knee functionality in the daily life, etc. This study proposes a new wearable joint kinematics monitoring system using a composite sensor which is composed of inertial and magnetic sensors. For each segment of the joint, the composite sensor is attached. By indentifying the two sets of acceleration and magnetic vector at the center of knee joint, difference of pose between the sensors are estimated. And, all of three joint angles; flexion/extension (f/e) angle, internal/external (i/e) rotation angle and varus/valgus (v/v) angle, are calculated. The proposed system was applied to knee joint kinematic analysis. The measurement accuracy was evaluated by comparing with optical sensor system. The mean error of measuring f/e angle was -1.61 +/- 3.09 deg; of measuring i/e angle was 0.93 +/- 1.75 deg; and of measuring v/v angle was 1.83 +/- 1.79 deg. To demonstrate the applicability of wearable monitoring, the proposed method was applied to measure knee joint angles during walking and during stair-climbing.

This paper describes a human health monitoring system by an ultrasonic sensor and an mat sensor systems. The system is realized with constrain-free, low cost and bed-side usage applicable. In it the ultrasonic sensor system obtains the state of a patient in bed by placing it under a bed frame. The mat sensor system detects heart beats and respiration signals by placing it to the mattress on the bed. This means that we can measure autonomic nerve system by using the heart rate and contribute the diagnosis of sleep apnea. This system employs fuzzy logic techniques to detect them. Thus, the system of systems with fuzzy logic can noninvasively and unconsciously provide human health information with high accuracy.

Knee kinematics in the every daily life or during the sports activities attracts the considerable attentions. Recent advances in micro electro mechanical system technologies produce many types of mobile sensors. This article introduces a system-of-systems (SoS) which estimates the knee kinematics by integrating MARG sensor and pressure sensor. Each sensor is with miniature size, low power consumption and wireless data transmission. The sensors measure the different signals simultaneously, and the intelligent data analysis system derives 3-degree-of-freedom knee joint angles and the knee joint moment by integrating the different kinds of signals. And, the present SoS can be applied to almost all activities in the daily life.

In this paper, we propose a personal identification by sole pressure change. We obtain sole pressure change of multiple steps by using two pressure sensor sheets. Each pressure sensor sheet is inserted into each shoe as an inner sock. Then, we extract characteristics of sole pressure change from the obtained data. We make template data of both feet from the extracted characteristics. We propose a Euclidean distance based method for personal identification. As the experimental result, we have recognized one of ten volunteers with over 90% accuracy.

It is known that hypoxic ischemic encephalopathy of the neonate decreases the cerebral volume. And, the volume decrement occurs at the different rate on each gyrus. Therefore, it is helpful for diagnosing such symptoms to segment the cerebral region from neonatal MR images and measure the volume of each gyrus. Many conventional methods for segmenting brain region from adult brain MR images have been proposed. However, in case of neonatal subjects, these methods cannot extract the sulci correctly. This paper proposes a method for extracting the cerebral contour accurately using cerebral surface model, which is the novel segmentation algorithm proposed in this paper. We applied the proposed method to computer synthesized images and MR images of a neonatal subject. The experimental results showed that the proposed method extracted the cerebral contour well.

Knee kinematics has been investigated to diagnose subjects with injured anterior cruciate ligament (ACL). Dejnabadi et al. presented a kinematics analysis method using accelerometers and gyroscopes. Although it can analyze the knee kinematics without accumulation of errors, it estimates only knee flexion/extension angle. Thus, this paper proposed a new method that can estimate all knee joint angles, flexion/extension, internal/external rotation, and varus/valgus angles. It is enabled by introducing earth magnetisms into Dejnabadi's method. The proposed method was validated by applying the proposed method a subject. And, the estimated angles were numerically evaluated by comparing the results with optical motion system.

This paper proposes an ultrasonic nondestructive evaluation method for estimating cellular quantity in artificial culture bone by an ultrasonic system. In order to measure the cellular quantity, we transmit ultrasound over the culture bone including stem cell. The characteristics of the amplitude and frequency of the obtained ultrasound waves are employed. Fuzzy inference system estimates the cellular quantity by using these characteristics. As a result, we can estimate cellular quantity in culture bone with 83.33% accuracy.

Rupture of anterior cruciate ligament (ACL) is a serious problem which causes in functional stability of the knee joint. To restore the knee injury, surgical operations, which replace the damaged ACL, have been applied. After ACL reconstruction, it is important to numerically evaluate the knee kinematics to diagnose the ACL reconstruction patients and to evaluate ACL reconstruction operation techniques. There are a few methods for analyzing 3-D knee kinematics based on image registration of multidetector-row CT (MDCT) and digital radiography (DR). However, estimation accuracy depends on the given initial pose/position by user-interactions. This paper proposes a fully-automated method for analyzing 3-D knee kinematics using MDCT and DR. The method firstly performs rough 3-D pose/position recognition with respect to anatomical landmarks, and secondly performs fine-image registration using digitally reconstructed radiography (DRR) images generated from MDCT images. Thus, the method requires no user-interaction. Performance of estimating 3-D knee kinematics was evaluated by using computer-synthesized images and subject's DR image.

Estimation of visual axis during sleep has been attracting a considerable attention. A simultaneous measurement system composed of functional MRI and infrared-video has been developed to investigate a relationship between eye-movement and brain function during sleep. Although there are some methods for measuring visual axis of opening eyes from video images, they cannot be applied to estimate visual axis of closing eyes during sleep. This paper proposes a method based on artificial neural network (ANN) for estimating visual axes during sleep from infrared-video images. Also, this paper introduces a novel calibration method using MRI. The method takes structural MR images of the eyeball and detects the visual axes from the MR images. And, using the detected visual axes and the simultaneously taken infrared-video image, the ANN is trained. The experimental results showed that the proposed method detected visual axes of the right and the left eyes with errors of 1.32+/-4.24 (RMSE+/-SD) deg and 1.26+/-4.20 deg, respectively.

We propose an imaging system of brain surface and skull by considering the ultrasonic refraction of the skull. We do an experiment by using a cow scapula to imitate the skull bone and a biological phantom to imitate cerebral sulcus. We first visualize the shape of skull. We second calculate the thickness of the skull aided by fuzzy logic. Finally, we calculate the refractive angle of ultrasonic wave and visualize the image referring to the refraction of ultrasonic wave. In the result of applying this method, we can successfully visualize the phantom surface image.

In this paper, we describe an ultrasonography system for locating screw holes of intramedullary nail by one-direction freehand scanning using an ultrasonic array probe. Although conventional X-ray method can visualize the nail in the femur, it has serious problem of X-ray exposure. We propose a locating method of the nail screw holes by an ultrasonic array probe. We extract screw hole regions by calculating two fuzzy degrees: average of the intensity and variance of the intensity using fuzzy inference. Next, we do a registration between the obtained image with the true image, where the true image is the nail image obtained by scanning an array probe to the nail exactly. As the result, we could calculate the center distance of two screw holes within an error of 1.0 mm.

In this paper, we propose a personal identification method using sole information. We employ pressure changes of sole in walking. First, we do a preliminary examination using a load distribution sensor. We employ a neural network for a personal identification method. As the result, we show a possibility of an identification system by sole information using the load distribution sensor. Based on the result, we propose a personal identification method by sole pressure changes using three air pressure sensors. This method could identify one of five volunteers at 85.4% recognition rate.

Measuring visual axis on the eye closure will play one of important roles to investigate the brain function during sleep. It has been investigated using a simultaneous measurement system composed of functional MRI and infrared-video which takes palpebra images with eye closure. Although there are some methods for measuring visual axis from video images, they cannot be applied to estimate visual axis with eye closure became their methods are based on tracing pupil reflection or Purkinje image. This paper proposes a novel method for fully- automatically estimating visual axis with eye closure using infrared-video. The method evaluates intensity profile on palpebra using artificial neural network (ANN). The ANN is preliminary trained using visual axes detected from MR image of eyeball. The experimental results showed that the proposed method detected visual axes of right and left eyes within the errors of 1.30 +/- 3.34 (RMSE +/- SD) deg and 1.12 +/- 3.70 deg, respectively.

Analyzing knee kinematics after total knee arthroplasty (TKA) has been attracting considerable attentions because the knee kinematics can be used to evaluate TKA patients and to evaluate TKA operations and design of knee implants. Knee kinematics can be estimated by 2-D/3-D image registration from 3-D computer-aided design (CAD) models of knee implants to 2-D X-ray image. Although there are many studies for estimating knee kinematics, they have common problems that are dependency on initial pose/position and falling into local maxima. This study proposes a robust 2-D/3-D image registration method based on evolutional computing. The evolutional computing has both characteristics of global search performance and of local search performance. The characteristics are suitable for solving the problems of 2-D/3-D image registration. The proposed system has been evaluated by applying it to computer-synthesized images, X-ray images of phantoms, and X-ray images of TKA patients.

Generating surface shaded display images and measuring the volumes of cerebral ventricles using 3-D SPGR MR images will help to diagnose many types of cerebral diseases with quantitatively and qualitatively. However, manual segmentation of cerebral ventricles is time-consuming and is subject to inter- and intra-operator variation. This article proposes a fully automated method for segmenting cerebrospinal fluid (CSF) and cerebral ventricles from MR images. Our method segments the cerebral ventricles by using a representative line (RL), which can represent the abstract of the shape and position of the cerebral ventricles. The RL is found by fuzzy If-Then rules that can implement physicians' knowledge on the cerebral ventricles. The proposed method was applied to MR volumes of 20 normal subjects, 20 Alzheimer disease (AD) and 20 normal pressure hydrocephalus (NPH) patients. The segmentation error ratio of the lateral ventricles was 1.98% in comparison with the volumes of manually delineated region by a physician. Using the proposed method, we found that patients of NPH significantly increased the ratio of volume of the lateral ventricles to the total CSF volume in comparison with that of AD (significance level < 0.001).

Rupture of anterior cruciate ligament (ACL) is a serious problem on playing sports, and may cause knee pain or hemarthrosis To restore the damaged ACL, many operation techniques for ACL reconstruction have been proposed. Therefore, analyzing the knee kinematics after ACL reconstruction is important to evaluate the ACL reconstruction operations and to improve them. However, few methods for quantitatively analyzing the three-dimensional (3D) knee kinematics have been proposed. In this paper, we propose a fully automated measurement system of 3D knee kinematics after ACL resonctruction using multidetector-row CT and X-ray fluoroscopic images.

Measurement of volume and surface area of the frontal, parietal, temporal and occipital lobes from magnetic resonance (MR) images shows promise as a method for use in diagnosis of dementia. This article presents a novel computer-aided system for automatically segmenting the cerebral lobes from 3T human brain MR images. Until now, the anatomical definition of cerebral lobes on the cerebral cortex is somewhat vague for use in automatic delineation of boundary lines, and there is no definition of cerebral lobes in the interior of the cerebrum. Therefore, we have developed a new method for defining cerebral lobes on the cerebral cortex and in the interior of the cerebrum. The proposed method determines the boundaries between the lobes by deforming initial surfaces. The initial surfaces are automatically determined based on user-given landmarks. They are smoothed and deformed so that the deforming boundaries run along the hourglass portion of the three-dimensional shape of the cerebrum with fuzzy rule-based active contour and surface models. The cerebrum is divided into the cerebral lobes according to the boundaries determined using this method. The reproducibility of our system with a given subject was assessed by examining the variability of volume and surface area in three healthy subjects, with measurements performed by three beginners and one expert user. The experimental results show that our system segments the cerebral lobes with high reproducibility.

Finding intracranial aneurysms plays a key role in preventing serious cerebral diseases such as subarachnoid hemorrhage. For detection of aneurysms, magnetic resonance angiography (MRA) can provide detailed images of arteries non-invasively. However, because over 100 MRA images per subject are required to cover the entire cerebrum, image diagnosis using MRA is very time-consuming and labor-intensive. This article presents a computer-aided diagnosis (CAD) system for finding aneurysms with MRA images. The principal components are identification of aneurysm candidates (= ROIs; regions of interest) from MRA images and estimation of a fuzzy degree for each aneurysm candidate based on a case-based reasoning (CBR). The fuzzy degree indicates whether a candidate is true aneurysm. Our system presents users with a limited number of ROIs that have been sorted in order of fuzzy degree. Thus, this system can decrease the time and the labor required for detecting aneurysms. Experimental results using phantoms indicate that the system can detect all aneurysms at branches of arteries and all saccular aneurysms produced by dilation of a straight artery in 1 direction perpendicular to the principal axis. In a clinical evaluation, performance in finding aneurysms and estimating the fuzzy degree was examined by applying the system to 16 subjects with a total of 19 aneurysms. The experimental results indicate that this CAD system detected all aneurysms except a fusiform aneurysm, and gave high fuzzy degrees and high priorities for the detected aneurysms. Copyright © 2006 The Institute of Electronics, Information and Communication Engineers.

Finding intracranial aneurysms plays a key role in preventing serious cerebral diseases such as subarachnoid hemorrhage. For detection of aneurysms, magnetic resonance angiography (MRA) can provide detailed images of arteries non-invasively. However, because over 100 MRA images per subject are required to cover the entire cerebrum, image diagnosis using MRA is very time-consuming and labor-intensive. This article presents a computer-aided diagnosis (CAD) system for finding aneurysms with MRA images. The principal components are identification of aneurysm candidates (= ROIs; regions of interest) from MRA images and estimation of a fuzzy degree for each aneurysm candidate based on a case-based reasoning (CBR). The fuzzy degree indicates whether a candidate is true aneurysm. Our system presents users with a limited number of ROIs that have been sorted in order of fuzzy degree. Thus, this system can decrease the time and the labor required for detecting aneurysms. Experimental results using phantoms indicate that the system can detect all aneurysms at branches of arteries and all saccular aneurysms produced by dilation of a straight artery in I direction perpendicular to the principal axis. In a clinical evaluation, performance in finding aneurysms and estimating the fuzzy degree was examined by applying the system to 16 subjects with a total of 19 aneurysms. The experimental results indicate that this CAD system detected all aneurysms except a fusiform aneurysm, and gave high fuzzy degrees and high priorities for the detected aneurysms.

A conventional ultrasonography system can noninvasively provide human tissue and blood flow velocity information with real-time processing. In general, since the human skull prevents the disclosure of brain anatomy, we usually placed the sensor at the anterior and superior attachment site of the upper ear (the posterior temporal window) in adults. Due to this limitation, the conventional system cannot obtain transcranial information from arbitrary places in the skull. This paper describes a transcranial sonography system that can visualize the shape of the skull and brain surface from any point to examine skull fracture and brain disease such as cerebral atrophy and epidural or subdural hematoma. In this system, we develop anatomical knowledge of the human head, and we employ fuzzy inference to determine the skull and brain surface. To evaluate our method, three models are applied: the phantom model, the animal model with soft tissue, and the animal model with brain tissue. In all models, the shapes of the skull and the brain tissue surface are successfully determined. Next, the method is applied to two adults. As a result, we have determined the skin surface, skull surface, skull bottom, and brain tissue surface for the subjects' foreheads. Consequently, our system can provide the skull and brain surface information using three-dimensional shapes.

Most of the methods for investigating brain function observe the localization of brain activity responded by the external stimuli which are executed with simply and repeatedly. On the contrary, when brain function is rest state such as falling asleep, a spontaneous brain signal measured by EEG or the other modalities is used as a trigger of the stimulus. This paper proposes an analysis method for detecting activation areas and activation times from fMRI time series data without giving tasks or stimuli for a subject. The results indicated that the proposed method detected activation areas with similar accuracy to those obtained by SPM99. Also, the proposed method was applied to a subject who did not conduct any tasks. The results showed that the method detected various brain activations, which would be spontaneous brain event, while we give no stimuli or tasks.

MR cholangiography (MRC) is a commonly used imaging method for diagnosing the pancreatic duct. This article proposes a novel method for enhancing medical images called target image enhancement using representative line (TIER), and its application to MRC images. Our method first finds the representative line (RL) of the pancreatic duct using a fuzzy if-then rule. TIER presumes the approximate region of the pancreatic duct using the obtained RL, and then emphasizes the intensity in the approximate region. Therefore, our method does not require any segmentation procedures, which often may be hard work. We evaluated the ability to find the RL and to estimate the approximate region of the pancreatic duct by applying the proposed method to computer-simulated data and MRI phantom data. We also show the image-enhanced images for normal and abnormal patients. (C) 2004 Wiley Periodicals, Inc.

Segmentation of lung lobes from MDCT images can provide effective information for functional assessments of each lobe and detection of pulmonary diseases such as the emphysema and lung cancers. Conventional studies have detected pulmonary fissures that located between lung lobes. However, some parts of fissures may disappear in MDCT images because of artifacts or the adhesion between lung lobes. This paper proposes a novel method for segmenting lung lobes based on tubular tissues, which are the peripheral blood vessels and bronchus. Our method estimates the boundary surface between lung lobes by fitting a curved surface. The fitting is performed with fuzzy control, and it searches the boundary where the density of tubular tissues is low. As a result of applying the proposed method to two normal subjects, we could estimate the boundary surface between lung lobes and segment lung lobes successfully.

This paper proposes an automated procedure for discriminating tissues using ultrasonic waves. Generally, the property of the echoes varies at each tissue. It is useful to discriminate the tissues according to difference of the property for diagnosis of diseases such as cancer, But, it is difficult to discriminate the tissue from the property because of its inter- and intraobserver variability. Therefore, the proposal method is aided by fuzzy inference and consists of two stages. The first stage constructs the expert system. In this stage, the fuzzy membership functions are automatically constructed at each characteristic value from pre-experimental data. The second stage predicts the tissues from another waves using the constructed expert system. Finally, the method was applied to six tissues. The experimental result indicated that all the tissues were discriminated (96.7%) with high accuracy than "C4.5" (93.3%).

In this paper, we propose an automated method for segmenting the cruciate ligament and the meniscus from CT knee images. The method first finds the candidate region of interests (ROIs) and the bone region by using intensity thresholding. The obtained bone region is decomposed into the femur and the tibia by watershed segmentation. To eliminate the cartilage and the cortical bone from the candidate region we can express these tissues by using the fuzzy if-then rules. To segment the ROIs we employed physician's knowledge; 'the cruciate ligament and the meniscus are wedged between the femur and the tibia', 'the shape of the meniscus is half-moon' and 'the cruciate ligament are located near the center of the knee'. The knowledge is converted to fuzzy if-then rules, and then the rules can compute the fuzzy degree for ROIs. To evaluate our method, it was applied to 5 normal subjects. Quantitative evaluation of the resultant images by a physician shows that our method can give interesting 2D reconstructed and 3D surface rendering images. These results would help us to understand 3D shape and to evaluate the condition of the cruciate ligament and the meniscus.

This paper proposes a method for extracting the cholecyst and the bile duct from magnetic resonance cholangiography (MR-C) volumetric images. We propose weighted fuzzy c-means clustering to classify an MR-C image into some clusters in which voxels have similar intensity and similar position. Then, the method finds the clusters corresponding to the cholecyst and the bile duct by evaluating the center vectors. Our experimental result on six subjects showed that this method could extract both the cholesyst and the bile duct.

Image segmentation is one of the fundamental techniques to develop a computer-aided diagnosis (CAD) system in the medical field. This paper first introduces rough sets into image segmentation method. In this method, attribute values of each pixel of an image of interest are given by using K-means clustering, and the attribute values divide the image into many regions. By applying value reduct, which is one of the typical concepts of rough sets, to the attribute values, dissimilarities between regions are calculated. Final clustering result is obtained by merging similar regions. To evaluate the performance of the proposed image segmentation method, it was applied to an artificial generated image, and a human brain Magnetic Resonance (MR) image. The results were also compared with convention K-means clustering.

This paper proposes an image segmentation method based on fuzzy if-then rules. It is a derivative of the conventional region growing method. This method represents expert's knowledge using fuzzy if-then rules, and embeds them as the growing criteria. To examine the proposed method, it has been applied to artificially generated images involving white Gaussian noise. In comparison with the conventional region growing method, the proposed method can segment region of interests(ROIs)with high robustness against to white noise. Moreover, it has been applied to dynamic mognetic resonance(MR)images of the Liver. The growing Criteria that represent physician's knowledge of MR images were derivedfrom the illustrated time-density curve of the liver, hepatic arteries, and veins after intravenous bolus injection. The experiments were done on three different normal volunteer with promising results.

This paper describes useful fuzzy logic techniques for medical image segmentation. Specific methods to be reviewed include fuzzy information granulation, fuzzy inference and fuzzy cluster identification. Fuzzy information granulation is introduced as a powerful scheme to find the thresholds to obtain the whole brain region in MR data. Fuzzy inference technique succeeds to segment the brain region into the left cerebral hemisphere, right cerebral hemisphere, cerebellum and brain stem. The fuzzy inference aided segmentation procedure is also useful to human foot CT image. Fuzzy cluster identification is adapted to determine the obtained clusters into blood vessel or other tissues in MRA image.