畑 豊

Measurement of cerebral volume and surface area using magnetic resonance (MR) image is effective for quantitative diagnosis of cerebral diseases. The measurement should require a brain segmentation process. Although many approaches for adult brain have been studied, there are few studies for neonatal brain. This study proposes a brain segmentation method for neonatal brain. Based on system of systems engineering technology, the proposed approach is composed from two systems; automated fuzzy logic based skull striping (AFSS) system and contour shape based modeling (CSM) system. AFSS segments the cerebral region based on Bayesian classification with Gaussian mixture model. CSM evaluates the skull stripping result of AFSS, and updates AFSS system parameters. Experimental results in 34 neonates (revised age between 2 weeks 1 day and 2 years 5 months) showed that the proposed approach segmented the brain region with sensitivity of 98.1% and false-positive rate of 27.9%. © 2011 IEEE.

This paper proposes a YURAGI-Analysis for brain imaging under the skull. In it, we employ 1.0MHz and 0.5MHz ultrasonic waves. We consider the weighted sum of these waves and attempt to extract the skull depth and image the sulcus under it. We add 1.0MHz and 0.5MHz, and we add the waves of 1.0MHz and Gaussian noise as the YURAGI analysis. We visualize the sulcus and skull. First, we calculate the thickness of the skull from the each of two synthesized waves. The thickness is determined from the surface and bottom points determined from the wave based on fuzzy inference. The sulcus surface was extracted from B-mode images for the each of two synthesized waves. As the result using a cow scapula as the skull and steel ditch as the human sulcus, we successfully calculated skull thickness. We extracted the sulcus width within the error of 5.86 mm and depth within the error of 1.94 mm. As for imaging the sulcus under the skull, the highest effectiveness of the synthesized wave is 96.30% when the weight of 0.5MHz waves is 0.60, and the one of YURAGI-Analysis wave is 97.15% when the weight is 0.003. Thus, YURAGI-Analysis is useful to this study. © 2011 Copyright Society of Photo-Optical Instrumentation Engineers (SPIE).

This paper describes an effective system of selecting most suitable rasp for the patients. We suggest the ultrasonic system with a single ultrasonic probe. Firstly, we make a knock to the upper point of the rasp inserted with a hammer which has the trigger signals. We detected the acoustic signals with the knocking signals. An ultrasonic probe is fixed in the upper side of the rasp with a built-in magnet. A personal computer got the acoustic data. In the surgery, the surgeon tries to adapt for the patient from the small size rasp to the larger size rasp in turn. There is a correlation between the degree of tightening and the attenuation time of acoustic signal for the knocked sound by a hammer when inserting the rasp. The higher tightened degree implies shorter attenuation period. At the present time, the surgeon's experience chooses the suitable rasp. Therefore, we suggest the rasp determination system which selects the suitable rasp by fuzzy inference which dynamically corresponds for each patient. As the result, we successfully determined the suitable rasp in comparison to the results obtained from the practical surgery. We could indicate the indexes in order to judge how degree the rasp fits in the clinical treatment. © 2011 IEEE.

This paper proposes a body weight prediction method using autoregressive (AR) model and Fuzzy-AR model. First, we employ 6 persons body weight change data of 365 days. AR model predicts body weight of a day from these time-series data. We calculate an order of AR model for each person by Akaike's Information Criterion. In the experiment, we predicted body weight change of next day for those subjects. The AR model obtained 0.798 in correlation coefficient between predicted and truth values. Second, we propose a Fuzzy-AR model that predicts body weight of next p days from last p days, where p is the order of AR model. In this method, we propose a Fuzzy-AR model with the fuzzy membership function using last p days data. In the experiment, the Fuzzy-AR model obtained 0.558 in correlation coefficient on 2 subjects. © 2011 IEEE.

Both burden of human and performance of sensing technology should be carefully considered in bioinstrumentation. The article proposes the ideas of invasion, intrusion, and consciousness in biomedical sensing to improve its comfort and performance. Consideration of the three concepts is very important to realize burden-reduction of human. They also save sensing accuracy when sensing targets and sensor heads are influenced by invasion, intrusion, or consciousness during sensing. Realizing noninvasive, nonintrusive, and unconscious sensing requires solutions against uncertainty in sensory signals and estimation models. Sensory signals from these sensing methods might superimpose various types of signals besides target one. Target signals would be appropriately extracted by the causal analysis-based model. Case studies were investigated by considering the method and experimental results were reported in the article. © 2011 IEEE.

This paper describes a testicular tubules evaluation using 1.0MHz ultrasonic array probe. In this system, we evaluate a diameter of testicular tubules. We employ an ultrasonic array probe with the center frequency of 1.0MHz. We employ evaluation index that cumulative relative frequency of amplitude values. In the experiment, we employ 24 nylon lines as the testicular tubules. Amplitude of large nylon line echo is larger than that of small nylon echo. For the evaluation, we calculate cumulative relative frequency amplitude of acquisition data. Fuzzy if-then rules are made by the cumulative relative frequency of large and small lines. We evaluate a rate of large lines among all lines by using the fuzzy MIN-MAX center-of-gravity method. In this experiment, the proposed method successfully evaluated the rate of the large lines. We changed the rate of large lines in 24 nylon lines, and tested our method 20 times for each rate. We evaluated the rate with 5.77 % in mean absolute error. © 2011 IEEE.

This paper describes a personal verification method based on fuzzy logic using dynamic sole pressure distribution while walking. The method employs a pair id sole pressure distribution change, and it is acquired by a mat type load distribution sensor. As a preliminary experiment for shoes, we take sole pressure data by bare foot and two kind of slippered foot. We extract thirty nine gait features from each sole pressure distribution change. We calculate a fuzzy degree of a feature from two fuzzy if-then rules and them fuzzy membership functions for a feature. These fuzzy membership functions are statistically determined by learning data. The fuzzy degree of acquisition sole pressure data is calculated by total of fuzzy degree of all features. The method verifies the walking person by using the fuzzy degree of the acquisition sole pressure data. When the fuzzy degree of acquisition data higher than a threshold, we verify the walking person as the target person. In our experiments, we employed 11 volunteers and took sole pressure data six times for each volunteer and foot situation. When the learning data included same kind of test data, we obtained low equal error rate. We obtained low false acceptance rate. © 2011 IEEE.

Cerebral contour extraction from magnetic resonance (MR) images is a fundamental work to analyze brain MR images. The methods can be roughly classified into three approaches, voxel-based, mesh-based and particle-based. Each method has advantages and disadvantages. Especially, particle based method can extract the complicated sulci with sub-voxel accuracy. The remained work is to develop a method for estimating probability of particle transition among gray matter, white matter and cerebrospinal fluid. This paper proposes a new method for calculating the particle transition probability based on fuzzy inference technique. The proposed method was applied to computer synthesized MR images and neonatal brain MR images of volunteers.

About 80% of paediatric intractable epilepsy patients have accompanying cortical dysplasia. However, there are no established methods for noninvasive detection of cortical dysplasia. This paper proposes a novel method for automatically detecting cortical dysplasia using paediatric MR images. In order to evaluate cortical dysplasia in MR images, texture features and fractal dimension were extracted with an automated method and support vectors were used to evaluate the degree of cortical dysplasia. The proposed method was applied to three paediatric epilepsy patients. The automated method identified the cortical dysplasia lesion with a sensitivity of 94% a mean specificity of 85%, and a mean efficiency of 87%.

This paper describes an ultrasonic array system by analyzing data for visualizing brain image under skull. We perform the experiment with a cow scapula as a skull and a steel sulcus as a lateral cerebral sulcus. We employ 1.0MHz and 0.5MHz ultrasonic array probes. The waves of them have some superiority, respectively. The 0.5MHz waves have higher penetration and the 1.0MHz waves have higher resolution. With taking each advantage of them, we synthesize with the waves of 1.0MHz and 0.5MHz. After that, we visualize the sulcus and skull. We calculate the thickness of the bone from the synthesized waves. The thickness is determined from the surface and the bottom points. The surface point is easily determined from the wave. The bottom point is determined by using fuzzy inference. The sulcus surface was extracted from a B-mode image made from the synthesized waves. As the result, we have calculated the best result of the weight of 0.4. Moreover, we extracted the steel sulcus images of the highest efficiency with the weight of 0.5. © 2011 IEEE.

Because almost all nations have come into aging society, it has been much more important to provide systems approach on health care than ever. There are many studies on human body from the view points of diseases however health has not been clearly defined. It would be difficult to clarify the notion of health because of various concepts of values and some cultural issues. With regarding this situation, Systems Healthcare would be new study area considering human health and wellness based on sensing and information technology. There have been developed various smart devices and services to prevent diseases and to improve health based on sensory data to realize the condition of our health. Causality between lifestyle and vital signals plays a important role for realizing them. In this article, the notion of Systems Healthcare is proposed and discussed. © 2011 IEEE.

Japanese government has limited CO 2 emissions and energy consumption for enterprises for these years. Especially, Japanese manufacturing industry have to fight to reduce energy consumption because they consumes a large amount of energy. Meanwhile, productivity must be kept when reducing energy consumption in manufacturing. The paper proposes the method of health monitoring with considering transparent productivity for QCDE optimization in manufacturing process. QCDE stand for Quality, Cost, Delivery, and Energy. he proposed method aims to reduce energy consumption without any QCD reduction by monitoring precise energy consumption of manufacturing equipment. Our proposed method quantifies operational condition of equipments by the energy consumption of equipments. The experimental results yielded that wasted energy consumption by the target equipment was extracted via monitoring its condition and energy consumption. © 2011 IEEE.

Among lots of vital signals, heart-rate (HR) is an important index for diagnose human's health condition. For instance, HR provides an early stage of cardiac disease, autonomic nerve behavior, and so forth. However, currently, HR is measured only in medical checkups and clinical diagnosis during the rested state by using electrocardiograph (ECG). Thus, some serious cardiac events in daily life could be lost. Therefore, a continuous HR monitoring during 24 hours is desired. Considering the use in daily life, the monitoring should be noninvasive and low intrusive. Thus, in this paper, an HR monitoring in sleep by using air pressure sensors is proposed. The HR monitoring is realized by employing the causal analysis among air pressure and HR. The causality is described by employing fuzzy logic. According to the experiment on 7 males at age 22-25 (23 on average), the correlation coefficient against ECG is 0.73-0.97 (0.85 on average). In addition, the cause-effect structure for HR monitoring is arranged by employing causal decomposition, and the arranged causality is applied to HR monitoring in a setting posture. According to the additional experiment on 6 males, the correlation coefficient is 0.66-0.86 (0.76 on average). Therefore, the proposed method is suggested to have enough accuracy and robustness for some daily use cases. © 2011 Copyright Society of Photo-Optical Instrumentation Engineers (SPIE).

Home security in night is very important, and the system that watches a person's movements is useful in the security. This paper describes a classification system of adult, child and the other object from distance distribution measured by an infrared laser camera. This camera radiates near infrared waves and receives reflected ones. Then, it converts the time of flight into distance distribution. Our method consists of 4 steps. First, we do background subtraction and noise rejection in the distance distribution. Second, we do fuzzy clustering in the distance distribution, and form several clusters. Third, we extract features such as the height, thickness, aspect ratio, area ratio of the cluster. Then, we make fuzzy if-then rules from knowledge of adult, child and the other object so as to classify the cluster to one of adult, child and the other object. Here, we made the fuzzy membership function with respect to each features. Finally, we classify the clusters to one with the highest fuzzy degree among adult, child and the other object. In our experiment, we set up the camera in room and tested three cases. The method successfully classified them in real time processing. © 2011 Copyright Society of Photo-Optical Instrumentation Engineers (SPIE).

This paper describes a biometric personal authentication method based on fuzzy logic using dynamics of sole pressure distribution while walking. The method employs a pair of right and left sole pressure data. These data are acquired by a mat type load distribution sensor. The proposed method has two processes. First, we calculate a fuzzy degree of each sole pressure data. In this process, we extract several gait features based on weight shift and shape of footprint. Fuzzy ifthen rules for each registered person are introduced. In it, their parameters are statistically optimized in learning process. Second, we combine fuzzy degrees of right and left sole. In this process, we employ five operators. The method authenticates walking person with the combined fuzzy degree. We calculate the fuzzy degree of an interest person for all registered persons, and identify the interest person as the registered person with the highest fuzzy degree. While, we verify the interest person as the target person if the fuzzy degree of the interest person calculated for a target person is higher than a threshold. In an experiment on 50 volunteers, we obtained low false rejection and false acceptance rates. © 2011 Copyright Society of Photo-Optical Instrumentation Engineers (SPIE).

This paper discusses a data analysis by YURAGI for a heart rate non-constraining monitoring system Three signals are employed: primary signal is obtained by a mat-type sensor, which is placed between a bed and subject, the second one is obtained by an ultrasonic vibration senor attached to bed frame, and third one is Gaussian noise. We compare the results from the synthesized data of the first and second signals with those of first signal and the noise. We employ weighted sum as the synthesized method. We consider Gaussian noise as YURAGI. The extraction algorithm was developed based on fuzzy logic. The comparison was done on 10 healthy volunteers and we evaluated the accuracy for various weight ratio. Here, we must concern the accuracy because the tiny accuracy difference causes large difference in the autonomic nerve system assessment. As the result, the results obtained from both synthesized signals were superior to that from mat-type sensor signal only. Thus, YURAGI analysis is useful to for detecting heart rate by mat-type sensor. © 2011 Copyright Society of Photo-Optical Instrumentation Engineers (SPIE).

This paper describes a foot age estimation system using fuzzy logic. The method employs sole pressure distribution change data. The sole pressure data is acquired by a mat type load distribution sensor. The proposed method extracts step length, step center of sole pressure width, distance of single support period and time of double support period as gait features. The fuzzy degrees for young age, middle age and elderly groups are calculated from these gait features. The foot age of the walking person on the sensor is estimated by fuzzy MIN-MAX center of gravity method. In the experiment, the proposed method estimated subject ages with good correlation coefficient. © 2011 IEEE.

This paper describes a trans-skull ultrasonic system that measures the blood flow velocity in brain under skull. In this system, we use an ultrasonic array probe with the center frequency of 1.0 MHz. The system determines the blood flow and locate blood vessel by Doppler effect. This Doppler effect is examined by the center of gravity shift in the frequency domain. We use three silicon tubes of different thickness imitated to blood vessel. We confirmed the change of frequency quantity of Doppler effect according to the water current velocity. The experimental result shows that the system detects the flow velocity by Doppler effect under skull and do automatic extracting method of water current. © 2011 IEEE.

Implanted knee kinematic analysis plays one of important role in clinical and research fields of total knee arthroplasty. Although there are some studies to analyze X-ray images for estimating 3-D knee kinematics, most of them cannot analyze dynamic video because they strongly depend on manual interaction of giving initial pose/position. This paper utilizes particle filter for analyzing dynamic radiograph video of implanted knee. By using particle filter, the proposed method does not require not only user interaction but also computational iteration of parameter optimization. As the result, we shorten the computation time and improved the estimation accuracy. The estimation error was lower than 0.7 mm for rotation, and 0.5 mm for translation including out-plane, and the computation time was 1.27 sec per frame using a cluster computer. © 2011 IEEE.

Personal classification using sole pressure change is essential for intelligent control, security on home because the users do not need to have anything, such as ID-Card, PIN code and so on. In our study, we propose a personal classification system by sole pressure change obtained by mat type pressure sensor. Mat type pressure sensor is placed on the floor in the entrance of home. We employ four features for classifying each family member and do personal classification based on Euclidean distance based method. As the experimental result on healthy 60 volunteers ranged from 20 to 80 years old, we have evaluated the performance. The results showed that the proposed system successfully classified them and it is especially useful in home intelligent system.

Composites of bone marrow stromal cells (BMSCs)/β-tricalcium phosphate (β-TCP) have been increasingly used as bone substitutes and studied as a bone graft model for bone tissue engineering. The number of seeded cells in the composites is a crucial factor for achieving successful bone tissue regeneration. In this study, we showed that the actual number of cells in BMSC/β-TCP composites 24 h after seeding at densities of 1.0×10 6, 1.5×106, 2.0×106, and 1.0×107 cells/mL was 2.8±1.5×105, 3.4±2.3×105, 3.7±1.0×105, and 3.7±1.8×105, respectively, indicating that even when one regular cell-seeding concentration was applied to the β-TCP, the actual number of cells in the individual BMSC/β-TCP composites varied considerably. In clinical setting, it is important to choose composites containing an appropriate number of cells before implanting them to patients. In an attempt to searching for the practical tools that can nondestructively evaluate the actual number of cells in β-TCP after cell seeding, we looked into ultrasound system and developed a nondestructive and quantitative ultrasound device. We successfully demonstrated for the first time that ultrasound amplitude effectively responded to the quantity of BMSC/β-TCP composites after 24-h cell seeding, and was well correlated to the actual number of cells contained (r=0.903). Using this ultrasound device, orthopedic surgeons can choose composites that contain favorable number of cells before implantation.Our device could be a valuable, convenient, and nondestructive tool for future bone tissue engineering. © 2010 Mary Ann Liebert, Inc.

新生児低酸素性虚血性脳症は脳形状の変化を伴うため，症状進行度の診断にはMR画像を用いた評価が有効である．しかし，手作業で脳表抽出を行うと医師に多大な負担がかかる．そのため，新生児頭部 MR 画像から脳表を自動的に抽出する手法が望まれている．成人脳を対象とする脳表抽出法は多く提案されているが，新生児脳を対象とする手法は少ない．本研究では，新生児頭部MR画像を対象とした粒子を用いる大脳表面抽出法を提案する．提案手法は，MR 信号値を基に粒子を脳領域に配置し，粒子を移動させ，粒子の分布を MR 画像と一致させることで脳表の抽出を行う．新生児頭部 MR 画像に対して提案手法を適用した結果，良好に脳表を抽出できた．It is effective to evaluate magnetic resonance (MR) images for diagnosis of the acuteness of symptom because hypoxic ischemic encephalopathy (HIE) accompanies the change in the brain shape. However, there are many problems for the doctor when brain surface is manually extracted. Therefore, it is required that brain surface is extracted from neonatal MR images automatically. There are many techniques to extract brain surface from adult MR images, but there are few technique to extract brain surface from neonatal MR images. In this paper, we propose a new extraction method using particle method from neonatal MR images. First, particles are assigned based on MR signal. Second, particles are moved based on neighboring particles. The proposed method applied to neonatal MR images. The results showed that proposed method extracted brain surface with high accuracy.

人工膝関節置換術後の生体内の人工膝関節の 3 次元位置姿勢の推定には，X 線透視画像を用いたイメージマッチングによる手法が提案されてきた．しかし，従来手法では大腿骨・脛骨コンポーネント間の位置関係を深く考慮していなかった．本研究ではパーティクルフィルタを用いた人工膝関節の 3 次元位置姿勢推定法を提案する．本手法は人工膝関節の位置関係に関する事前知識をファジィメンバーシップ関数として定義する．実験の結果，大腿骨，脛骨コンポーネント間の位置関係を考慮した位置姿勢の推定が可能となった．For estimating 3-D pose position of implanted knee joint in vivo after total knee arthroplasty, some studies proposed 2-D/3-D image registration using X-ray fluoroscopy images. However, these conventional method are not considered relationship between the femoral and the tibial component. This paper proposes a method for estimating 3-D pose position of implanted knee based on particle filter. A priori knowledge on the relationship of implanted knee joint are utilized by using fuzzy membership functions. The experimental results for a patient and computer synthesized DR images showed that the proposed method adequately estimate 3-D pose position of the implanted knee joint with a priori knowledge on the relational position between the components.

Three-Dimensional (3-D) shape reconstruction of total knee arthroplasty (TKA) implants in vivo plays a key role to investigate implanted knee kinematics. TKA implants typically consist of metal femoral and tibial components and a polyethylene tibial insert. X-ray computed tomography (CT) causes severe metal artifacts, making the 3-D shape in reconstructed images extremely difficult to understand. This article proposes a new method of 3-D reconstruction from X-ray cone-beam images. Called a fuzzy visual hull, it introduces fuzzy logic in recognizing X-ray images. X-ray cone-beam images are fuzzified and back-projected into a fuzzy voxel space. Defuzzifying the fuzzy voxel space enables the 3-D TKA implant shape to be reconstructed. The results of evaluation using TKA implants in vitro and computer-synthesized images demonstrated that the fuzzy visual hull provides high robustness against noise added to X-ray cone-beam images. The new approach also reconstructed the 3-D polyethylene insert despite the difficulty of recognizing the region in conventional X-ray CT.

In evaluating thoracic function, it is effective to segment the five lung lobes from multidetector-row computed tomography (MDCT) images. Almost all of the conventional methods are based on extracting the lobar fissures; however, some parts of the fissures may not be observed from MDCT images due to CT artifacts and/or adhesions between the lung lobes. This article proposes an alternative method for segmenting the lung lobes. It is based on tubular tissue density, and is not based on lobar fissure extraction. The tubular tissues are the peripheral blood vessels and peripheral bronchi. Because tubular tissues do not exist on the boundary between the lung lobes, our method determines the boundary by finding a continuous three-dimensional space in which tubular tissues are absent. The boundary determination process is automatically performed using fuzzy control. The proposed method was applied to five normal subjects, one patient with chronic obstructive pulmonary disease, and one patient with emphysema. The absolute mean error of detecting lobar boundaries was 3.4 mm and that the volumetric accuracy for the proposed method was an absolute ratio of 3.8 and 5.9% for inspiration and expiration, respectively. The proposed method is also applicable to MDCT images in which the lobar fissures cannot be distinguished. © 2010 ISSN 1349-4198.

脳疾患は脳形状に領域特異的な変形をもたらすため,診断には脳形状変形の定量化が有効である.従来の定量化法は,ボクセルの輝度値に基づく評価を行うため,脳溝などの脳構造を考慮していない.そのため変形が著しい場合では,異なる解剖学的部位で評価する恐れがある.そこで我々は,脳溝特徴に基づく脳形状相同モデル化を提案する.同手法では脳溝特徴の指標となるsulcal-distribution index(SDI)を定義し,構築した標準脳モデルをSDIに基づいて各評価脳に対して変形を行う.本文では,同手法より変形した標準脳モデルを用いて移動量に基づく変形評価を行う.本手法を成人10名に適用した結果,従来法であるVBMと同様の部位に有意差が認められた.

皮質形成異常は難治性てんかん患者に多くみられる先天性脳疾患である.難治性てんかんの治療には責任病巣である皮質形成異常領域を切除する外科的手術が行われているため,小児頭部MR画像からの皮質形成異常部位検出は非常に有効である.しかし,皮質形成異常のMR画像上での画像特徴は明らかでなく,またその自動検出法も未だ確立されていない.本文では,大脳皮質の層構造に直交した直線の輝度値推移から算出される特徴量を用い,サポートベクターマシン(support vector machine; SVM)で皮質形成異常度を推定する手法を提案する.また,我々の提案した従来法との精度比較を行う.本手法を適用した結果,小児頭部MR画像から,感度90.9%,特異度93.1%,有効度92.0%で皮質形成異常領域を検出できた.

脳疾患の進行度合いの診断に,脳形状の変化や脳容積の測定はとても有効である.本研究では,T1強調またはT2強調で撮像された新生児・幼児の頭部MR画像を対象とした頭蓋除去法を提案する.本提案法では幼児は成長に伴い白質の信号値が変化することを考慮したガウシアン識別を頭部MR画像に適用することで,白質,灰白質,脳脊髄液,脂質,その他領域の存在確率マップを自動生成する.識別後,白質と灰白質の和領域の表面に,脳形状や画像特徴に関する知識に基づくファジィ動的脳表モデルを適用し最適化することで,脳領域を自動抽出する.本提案法を,修正齢-4週間から4歳1ヶ月の26名の被験者に適用した結果,良好に脳領域抽出が行えた.

人工膝関節置換術(Total knee arthroplasty, TKA)後の膝関節の動態解析は整形外科における研究において非常に重要な問題である．これまでの研究において2次元のX線透視画像と人工膝関節の3次元形状モデルを用いた2-D/3-D イメージマッチング手法が多く提案されているが，これらは静止画像による解析に留まり，膝関節の連続的な動作を動画像により解析する研究は少数である．また，従来の手法は高次元のパラメータ空間による局所解の問題があり，膝関節の連続的な動作は考慮されていない．本論文では，パーティクルフィルタを用いた人工膝関節の3次元動態解析システムを提案する．提案手法はパーティクルフィルタにより複数の候補を作成しパラメータの探索を行い2-D/3-D イメージマッチングを用いて人工膝関節の3次元位置姿勢を推定する．実験の結果，提案手法により膝関節の連続的な動作の解析，膝関節角度の推定が可能となった．

新生児脳疾患は脳形状の変化を伴うことが多いため，症状進行度の診断にはMR画像を用いた評価が有効である．しかし，手作業で複数のMR画像から脳表抽出を行うと医師に多大な負担が生じる．さらに，診断者間変動や診断者内変動が起こる可能性もある．そのため，新生児頭部MR画像から脳表を自動的に抽出する手法が求められている．これまでに成人頭部MR画像を対象とした脳表抽出法は多く提案されているが，新生児頭部MR画像を対象とした手法は少ない．本研究では，新生児頭部MR画像を対象とした粒子を用いる脳表抽出法を提案する．提案手法は，脳組織に見立てた3種類の粒子をMR信号値を基にして脳領域に配置し，周囲の粒子分布を基にして粒子の遷移，移動を行い，粒子の分布をMR画像と一致させることで脳表の抽出を行う．新生児頭部MR画像に対して提案手法を適用した結果，良好に脳表を抽出できた．

大脳皮質形成異常は難治性てんかんの責任病巣となる先天性脳疾患である．また小児においては，難治性てんかん患者の約8割が大脳皮質形成異常を原因として発症するため，小児頭部MR画像からの大脳皮質形成異常部位の特定は非常に重要である．大脳皮質形成異常部位摘出の手術計画の立案には異常の程度を定量的に評価することが有効であるが，小児頭部MR画像における大脳皮質形成異常の画像特徴は未だ明らかにされていない．そこで本論文では小児頭部MR画像において大脳皮質形成異常を評価する新しい特徴量を提案する．

Personal classification using sole pressure change is essential for intelligent control, security on home because the users do not need to have anything, such as ID-Card, PIN code and so on. In our study, we propose a personal classification system by sole pressure change obtained by mat type pressure sensor. Mat type pressure sensor is placed on the floor in the entrance of home. We employ four features for classifying each family member and do personal classification based on Euclidean distance based method. As the experimental result on healthy 60 volunteers ranged from 20 to 80 years old, we have evaluated the performance. The results showed that the proposed system successfully classified them and it is especially useful in home intelligent system. © 2010 The Institute of Electrical Engineers of Japan.

In case of pediatric, about 80% intractable epilepsy patients are accompanied with cortical dysplasia. It is very effective to detect cortical dysplasia lesions in order to we plan treatment of intractable epilepsy. Although MR image analysis attracts the considerable attentions because it is non-invasive for the human body, automated methods for detecting cortical dysplasia with MR images of pediatric brain are not established yet. In addition, image features on the pediatric brain MR images of cortical dysplasia are not clarified. This paper proposes an estimation method for detecting cortical dysplasia in pediatric brain MR images. And, fractal dimension is evaluated as a new image feature of cortical dysplasia. The experimental results in four patients (6, 7, 2 and 3 years old) with cortical dysplasia showed that a mean sensitivity of 94.2 %, a mean specificity of 93.8%, and a mean efficiency of 93.8 %.

There are few studies on computer-aided diagnosis (CAD) system for neonatal magnetic resonance (MR) images. In addition, because it is difficult to apply multiple examinations for neonate, we should maximize data which are derived from a set of MR images. Under these requirements, this paper introduces system of systems (SoS) framework into analysis of neonatal MR images. The present system called SoS-CAD for neonatal brain is composed from skull stripping system, brain shape homologous modeling system, gyral labeling system and cerebral contour extraction system. By collaborating the systems with each other, the SoS-CAD produces multiple numerical and geometrical data from a set of MR images. © 2010 IEEE.

This paper proposes a biometric personal authentication method based on one step foot pressure distribution change. We acquire the foot pressure distribution change by mat type load distribution sensor and use it as a personal authentication. We employ twelve features based on shape of footprint, and twenty seven features based on movement of weight while walking. A classifier for each feature is developed on the basis of fuzzy inference. The classifier is trained by a clonal selection algorithm in artificial immune system. A personal authentication system for one step is made every classifier for all features. We employed 10 volunteers, and we took the step data five times. We evaluated our method by five-fold cross validation method. We obtained low false rejection and acceptance rates in identification and verification. © 2010 IEEE.

This paper describes a method for a heartbeat and respiratory rate monitoring system using air pressure sensors and ultrasonic oscillosensor. By using these sensors, we propose a detection method of the state of human and an extraction method of heartbeat and respiratory rate in bed by fuzzy logic. Our method was examined on four healthy volunteers. We successfully detected the state of human and extracted heartbeat and respiratory signals. In our method, fuzzy logic plays a primary role in the detection of the state and extraction of heartbeat and respiratory signals. An experiment on four healthy volunteers was done. Consequently, our proposed method noninvasively and successfully detects the state of human and extracted heartbeat and respiratory rate in the bed by using the unconstrained sensors. © 2010 IEEE.

Automated morphometric analysis using human brain magnetic resonance (MR) images is an effective approach to investigate the morphological changes of the brain. However, even though many methods for adult brain have been studied, there are few studies for infantile brain. Same as the adult brain, it is effective to measure cerebral surface and for quantitative diagnosis of neonatal and infantile brain diseases. This article proposes a skull stripping method that can be applied to the neonatal and infantile brain. The proposed method can be applied to both of T1 weighted and T2 weighted MR images. First, the proposed method estimates intensity distribution of white matter, gray matter, cerebrospinal fluid, fat, and others using a priori knowledge based Bayesian classification with Gaussian mixture-model. The priori knowledge is embedded by representing them with fuzzy membership functions. Second, the proposed method optimizes the whole brain by using fuzzy active surface model, which evaluates the deforming model with fuzzy rules. The proposed method was applied to 26 neonatal and infantile subjects between -4 weeks and 4 years 1 month old. The results showed that the proposed method stripped skull well from any neonatal and infantile MR images. © 2010 IEEE.

For estimating 3-D pose position of artificial knee implants in vivo, there are some studies based on 2-D/3-D image registration of 2-D fluoroscopy images and 3-D geometrical model. Knee implant mainly consists of femoral component and tibial component. Most conventional studies estimate 3-D pose position of femoral component and tibial component individually. Rather, they don't evaluate relative position between the femoral and tibial components. This paper proposes a method for estimating 3-D pose position of implanted knee based on particle filter. A priori knowledge on the relational position of the components are utilized by using fuzzy membership functions. The experimental results for a patient and simulation DR images showed that the proposed method adequately estimate 3-D pose position of the femoral and tibial components with respect to relational position between the components. © 2010 IEEE.

This paper describes soft computing identification methods for ellular quantity of Bone Marrow Stromal Cells in artificial culture bones. We attempt to identify cellular quantity with an ultrasonic system and approaches of a neural network and a fuzzy inference. We employ two features; amplitude and frequency. Amplitude is obtained from the raw ultrasonic wave, and frequency is calculated from frequency spectrum obtained by applying cross-spectrum method. A comparison was done with the multi regression method. The neural network approach identifies the cellular quantity with the highest accuracy. © 2010 IEEE.

This paper describes a method for distribution of people monitoring system using a 3D camera. This camera measures distance and intensity distribution based on time of flight (TOF). From the obtained distance data, we perform clustering processing to detect the distribution of people. In our method, fuzzy logic plays a primary role to decide the cluster, i.e., people number. Our proposed method applied for 120 sec in a room. We successfully detected distribution of people in a room for moving 3 people distribution. Consequently, our proposed method successfully detected distribution of people using automated cluster number determination. © 2010 TSI Press.

This paper propose a biometric personal identification method based on a pair of right and left sole pressure distribution change. We acquire the sole pressure distribution change by load distribution sensor and use it for a personal identification. We employ twelve features based on shape of footprint, and twenty seven features based on movement of weight during walking for each sole pressure data. We make these fuzzy if-then rules. We calculate a fuzzy degree of a pair of right and left sole pressure data for one person, and identify person by this fuzzy degree. We evaluated our method by five-hold cross validation method. The low false rejection and acceptance rates are evaluated from 20 to 90 persons. © 2010 TSI Press.

The pivot shift test has been performed to assess the instability of the knee caused by anterior cruciate ligament (ACL) injuries. However, the test depends on the clinician's subjective feeling. In this study, inertial and magnetic sensors have been introduced into quantitative evaluation of the pivot shift test. The analysis method extracts the knee movement of the pivot shift by using wavelet transformation. In the result of applying the proposed method to the ACL injured subject, pivot shift phenomenon was detected correctly in comparison with reference video images taken simultaneously, the mean of the maximum accelerations of the pivot shift was 2.19±O.69 m/s2, and the maximum accelerations were correlated with grade scores based on the clinician's subjective feeling. © 2010 TSI Press.

Conventional eye gaze tracking are based on special contact, optical equipment, Electrooculogram, etc. Although they have been produced valuable results in many areas, they cannot obtain the absolute eye gaze but obtain relative one based on calibration. This study proposes a method for detecting absolute eye gaze using MR images. Because MR images can obtain the sectional images of the eye ball, the method needs no calibration. The proposed method detects the eye gaze by segmenting the Vitreous, and lens of the eye ball. The method was applied to healthy subjects successfully. The estimated results were validated by comparing with the conventional method based on optical equipment.

Analyzing knee kinematics of implanted knee after Total Knee Arthroplasty (TKA) is affective approach in the research area of orthopaedics. There are some studies based on 2-D/3-D image registration of 2-D fluoroscopy images and 3-D geometric model. However, these conventional studies are designed for still statics image analysis, and there are a few studies for dynamic image of continuously knee movement. In addition, the another problem are to fall into local maxima in high-dimensional search space, and not to consider continuous knee movement. This paper proposes an analysis method of the continuous implanted knee kinematics based on Particle Filter. The proposed method estimates 3-D pose position of the implanted knee joint using 2-D/3-D image registration based on Particle filter. Particle filter is able to estimate optimal solution from multiple candidate pose position. The experimental results showed that the proposed method analyzed the continuous movement of the knee joint and estimated the knee joint angles. © 2010 TSI Press.