畑 豊

新生児低酸素性脳虚血性脳の進行度合いの定量化に,脳容積や脳回表面積の測定は非常に有効である.本研究ではMR画像を用いた脳容積計測,脳回表面積計測が可能な計算機診断支援システムを提案する.本提案システムでは,全脳形状や画像特徴に関する知識に基づくファジィ動的モデルを用いて全脳領域を自動抽出する.また脳回分割においては,操作者が脳表上で描画した脳回境界線を基に,ファジィ動的モデルにより脳回境界を自動検出する.提案システムを修正齢3週間から4歳2ヶ月の被験者8名に適用した結果,Sensitivity 97.72%, FPR 0.31%で全脳領域の抽出が行えた.また全被験者に対し各脳回の表面積計測が行えた.

ACL(anterior cruciate ligament; 前十字靭帯)損傷により生じる膝の回旋不安定性の評価手法として,主にpivot shift testが用いられる.同検査方法は簡便でACL損傷の程度と相関があり,今後の治療方針の決定に有用な指標であるが,同検査法は検者の主観的評価に基づいており,客観性,定量性が低いという問題点を有する.本研究では,小型で高時間分解能な慣性センサを用いた非拘束膝回旋不安定性定量化システムを提案する.同提案システムは,pivot shift test時の大腿部,下腿部運動を計測した時系列データの周波数解析により大腿-下腿の相対加速度を求めることで,pivot shift test現象の定量化を行う.本システムをACL損傷例1例に対して適用した結果,同時撮影したビデオ画像との比較により,pivot shift現象を正しく検出できたことを確認した.また,pivot shift現象時の大腿部に対する下腿部の相対運動の最大加速度は2.19±0.69m/s^2であり,検者の主観的評価であるGrade値との相関も示唆された.

心拍は、身体の変化を顕著に表す生体情報として知られており、自覚症状の現れにくい疾病を予防するためにも、その変化を継続的に観察することは重要である。さらに、心拍のゆらぎには、自律神経活動を推定する上で有用な情報も含まれており、一部の睡眠障害の要因でもある自律神経活動の変異を知る上でも有効な指標といえる。本稿では、この心拍を簡便かつ非侵襲に計測する技術を示す。実験により、本技術が心電と同等の計測精度を持ち、さらに、そのゆらぎから、自律神経活動を推定することができる可能性も確認した。

高齢者がベッドから降りる際の転倒事故は大怪我に至る可能性があり、転倒事故の予防はQOLの観点で重要な課題である。 事故予防のためには、高齢者が離床するタイミングを介護者が正確に把握できることが望ましい。本研究では、空気圧センサと超音波振動センサを用いた離床前動作の検出手法を構築した。本手法を健康な成人男性に適用した結果，正確に離床前動作を検出することを確認した。

新生児の脳疾患は永久的な脳障害を残す可能性が大きい．脳疾患の障害部位は症状によって異なり，また症状の進行度合いにより委縮，変形することが報告されている．そのため脳回別の表面積を測定，評価することは，症状の進行度合いの定量的な診断に有効である．しかし，新生児を含む乳幼児を対象とした脳回分割法は未だ確立されていない．そこで本研究では，ファジィ領域境界最適化アルゴリズムによる脳回ラベリング法を提案する．修正齢3週間から4歳3ヶ月の被験者14例に対して提案法を適用した結果，すべての被験者の脳表を脳回に分割することができた．

心拍の測定は、日常生活における健康状態の変容を知る上で重要な指標の一つである。さらに、心拍のゆらぎは、自律神経活動を推定するための判断指標として有用であることが知られている。すなわち、心拍の変動を正確に測定する技術が望まれる。我々は、睡眠時における体の圧力変動を空気圧センサで測定し、その変動からファジィ論理により心拍を推定する測定法を提案してきた。本稿では、ファジィ論理の有効性を心拍の特性と関連付けて議論し、ファジィ論理に含まれるパラメータが推定精度に及ぼす影響を考察する。

本文では，超音波による低位前方切除術のための腸の厚み決定法について述べる．低位前方切除術では，吻合器を用いた手術が行われる．この手術では，腸が吻合器により引っ張られて，厚みが薄くなりすぎると細胞が壊死するという問題がある．これを防ぐためには，手術中に腸の厚みを知り，腸の引っ張りを調整する必要がある．そこで，本文では15MHzの超音波プローブを用いた腸の厚みの決定法を提案する．本手法では，表面反射波と取得波形との相関係数を求めることによって、表面位置の決定を行う．次に，底面反射波の振幅値，底面反射波と表面反射波との相関係数，底面位置と表面位置との距離という3つの知識に関するファジィ所属度を用いて，底面位置の決定を行う．最後に，以上より決定された表面位置，底面位置を用いて厚みの算出を行う．実験の結果，5.09%の誤差率で厚みを測定することができた．

本文は，空気圧センサシステムを用いた自律神経系機能評価を提案する．自律神経系は心拍変動によって機能評価することができる．心拍変動を検出する一般的な方法として，心電計を用いる方法がある．しかし，心電計は複数のセンサを体表に直接固定するために被験者を拘束してしまう．空気圧センサシステムは，ベッドのマットレスの下に置くことによって，生体情報を非拘束に感知することができる．本文ではファジィ推論を用いた心拍変動法を提案し，自律神経系機能評価を行った．7人の被験者に提案法を適用した結果，心拍変動を心電計との相関係数0.851で検出することができた．また交感神経系の指標となるLF/HFを22.18%の誤差率で，副交感神経系の指標となるHF(n.u.)を11.98%の誤差率で評価することができた．

Horticultural therapy (HT) is gaining attention as a form of rehabilitations in medical fields especially such as occupational therapy and nursing care, although its effectiveness has not been proven yet. This paper uses a strictly medical point of view to assess whether or not HT is effective for improvement of functional activities in the brains of brain-damaged patients. Five patients in Ishikawa Hospital with cerebrovascular diseases were invited to participate in HT for a month in addition to their routine medication and physical therapy (PT). The HT program was designed by horticultural therapists. The original purpose of the HT program was to monitor its effects on mental healing, cognitive re-organization, and training of sensory-motor function. The Functional Independence Measure (FIM) and the Self-Rating Depression Scale (SDS) were performed before and after HT to assess the patients’ physical activities of daily living (ADL) and to determine the patients’ mental changes in depressive states, respectively. Functional magnetic resonance imaging (fMRI) during recognition tasks was also measured before and after HT. The ADL of all patients significantly improved after HT; however, the depressive states in all patients did not change remarkably after the HT. fMRI examinations showed that the visual area, the inferior temporal area, the fusiform gyros, and the supramarginal gyros (SMG), in addition to the motor area, the supplementary motor area (SMA), the sensory area, and the cerebellum were activated after HT. These findings suggest that HT can accelerate an improvement of activities in the “visual and color processing areas” and the “association areas” as well as the sensory-motor areas of the brain in the patients with cerebrovascular diseases. HT, therefore, stimulates parts of brain, that are not always evoked through routine physical rehabilitation. HT can complement the routine physical rehabilitation and help to improve damaged brain function. © 2008, TSI® Press Printed in the USA.

Many studies showed that the hemodynamic response (HR) to brief neural activity caused by the blood oxygen level-dependent effect is delayed some seconds, and that the HR latency (= time to peak from neural activation) varies among activation sites. This paper proposes a novel method for estimating HR latency by analyzing event-related functional magnetic resonance images based on a continuous wavelet transform. The proposed method can simultaneously detect activation areas and estimate the HR latency. We also studied variability of the HR latency across subjects within the same activation sites, across acquisition days within a subject, and among activation sites within a subject. The experiments were done on 10 healthy subjects with hand-gripping tasks. We found no significant difference of HR latencies at a given activation site within a subject across acquisition days. And, we found significant differences of HR latency between the supplementary motor area (SMA) and the primary motor cortex (M1) in individual subjects, and among subjects. © 2008, TSI® Press Printed in the USA.

Computer aided diagnosis (CAD) system is an effective system of systems engineering (SoSE) in the medical field. The CAD system using medical images is constructed by integrating medical image acquisition system, medical image processing system, and clinical data analysis system, and it presents both of quantitative and qualitative data for physicians. This article proposes a CAD system for diagnosing neonatal and infantile brain disorders. The CAD system acquires intracranial images using MR scanner system, and segments the cerebral volume and surface using medical image processing systems, and produces cerebral volume and area, and 3-D rendering of cerebral surface. Especially, this article proposes medical image processing systems based on expert knowledge systems for neonatal/infantile brain MR images. To validate the proposed CAD system, it has been applied to three neonatal and five infantile MR images.

This paper describes a fuzzy processing method for a respiratory rate monitoring system by an optical fiber sensor. This optical fiber sensor has a part of sensor sheet and a polarization detector. It noninvasively detects vital signal of human on the sheet by obtaining polarization variations. By using this sensor, we propose a fuzzy processing method of a respiratory rate for human in the bed. Our method was tested on five volunteers. We successfully detected a respiratory rate. In it, fuzzy logic plays a primary role in the detection. Our system determines fuzzy membership functions by using characteristic of respiration. Consequently, this system can noninvasively detect a respiratory rate by using a constraint-free device.

It is effective to measure the surface area of each gyrus for quantitative diagnosis of infantile brain diseases. This paper proposes an interactive gyrus labeling method for the infantile brain in magnetic resonance images. First, a user roughly gives guidelines of gyral boundaries on a 2-D projected image of the cerebral surface. The cerebral gyri are labeled by automatically determining gyral boundaries with respect to the user-given guidelines. The boundary deformation process is based on fuzzy rules. The automatically determined boundaries are validated by the user, and modified interactively. We applied the proposed method to 14 infantile subjects (3 weeks - 4 years 3 months old). The results showed that the cerebral gyri were successfully labeled with a mean sensitivity of 92.8% and a mean false positive rate of 0.1%. ©2008 IEEE.

Volumetric cerebrum and measurement of cerebral surface area using MR images plays a fundamental role in computer-aided diagnosis (CAD) of neonatal cerebral diseases such as hypoxic ischemic encephalopathy. There are many conventional methods for brain extraction from adult MR images. However, it is hard to apply these methods for neonatal MR images that image features are different from adult one. This paper proposes a novel method for extracting neonatal cerebral surface with sub-voxel accuracy using Thick Rubber Model (TRM). This method extracts a cerebral surface by deforming TRM so that pseudo MR images synthesized from TRM are identical to the given MR images. Experimental results showed that the proposed method extracted the cerebral surface with higher accuracy (mean RMS distance was 9.0 mm) in comparison with the conventional method (mean RMS distance was 15.7).

The aim of this paper is to assess whether horticultural therapy (HT) is effective for improvement of daily activity and brain functional activity in brain-damaged patients from a medical point of view. Five patients with cerebrovascular disorders were invited to participate in horticultural therapy for a month in Ishikawa Hospital. The Independence Scale in Activities of Daily Living (ISADL) and the Mini Mental State Examination (MMSE) were performed before and after a whole series of horticultural therapy program to assess a patient's activities of daily life and to determine the patient's cognitive function, respectively. Functional magnetic resonance imaging (fMRI) under recognition tasks were measured before and after the horticultural therapy. Independency of the daily activities and cognitive function tended to improve after the horticlutural therapy. fMRI examinations showed that the supramarginal gyrus (SMG), the motor area, the supplementary motor area (SMA), the sensory area, the visual area, the middle temporal gyrus, and the frontopolar area were activated after the therapy in most patients.

To diagnose total hip arthroplasty (THA) patients, it is important to analyze in-vivo kinematics of the implanted hip joint. There are some conventional analysis methods for estimating the pose/position of the implanted hip joint using 2-D/3-D image registration technique for 3-D multidetector raw CT (MDCT) images and 2-D X-ray radiograph images. One of the remained problems is that estimation results depend on the initial pose/position given manually. This paper proposes a fully automated method for estimating pose/position of the implants and the hip joint without giving initial pose/position. The proposed method releases us from manual interactions of giving initial pose/position by intervening between 3-D MDCT images and 2-D X-ray radiograph images with 3-D computer-aided design (CAD) model.

We propose an ultrasonic system to determine thickness and wave speed of an object. In it, we develop a triangle probe system composed of three probes; a central probe and two side probes. The central probe straightly transmits/receives ultrasonic waves, and the right probe receives ultrasonic waves transmitted from the left one. We move the triangle probe to depth direction of an object. Then, two peak intensities are revealed at the focal points of the surface and the bottom of the object. We employ fuzzy logic to extract the two focal points, and determine the wave speed and thickness of the object from these focal point data. The experimental results on two hard phantoms show that the method can successfully determine the wave speed and the thickness of the object with high accuracy.

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 often volunteers with over 90% accuracy.

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. In the quantitative assessment, the mean of the maximum acceleration during the pivot shift was 2.19±O.69 m/s2 , and the maximum acceleration values were correlated with grade scores based on the examiner's subjective evaluation.

This paper describes a human health monitoring system by an air pressure sensor and an ultrasonic sensor system. The ultrasonic sensor system can obtain the state of a patient in bed by placing it under a bed frame. That is, we can determine whether or not a patient in the bed. The air pressure sensor system can detect heart beats by placing it to the mattress on the bed. By using two sensors complementary, we can detect the behavior before getting out of bed aided by fuzzy inference technique. The system of the two sensors can complementary detects the state, heart beat, a behavior before getting out of bed. Thus, the system of the two sensor systems can noninvasively and unconsciously provide human health information with high accuracy aided by fuzzy logic. © 2008 IEEE.

Estimating 3-D pose/position of knee implants in-vivo is one of important methods for evaluation of total knee arthroplasty (TKA). It has been performed with 2-D/3-D image registration of knee implants' 3-D geometrical model (i.e., CAD; computer-aided design) and 2-D X-ray radiography image, which can take a projection image of TKA implants in-vivo. Although they have been producing valuable results, they cannot be applied without 3-D geometrical models. It is the critical limitation for clinical situations because most of physicians cannot have 3-D geometrical model. This paper proposes a novel system for evaluating knee implants without 3-D CAD models, and it uses 3-D digital radiology (DR) images. The system is based on a new method, called fuzzy visual hull, which is an extension of the conventional visual hull by introducing fuzzy logic. The new method can reconstruct 3-D object shape from 2-D silhouette images with vague contour of objects. The new method was evaluated by using TKA implants in-vitro and computer synthesized images. The results showed that fuzzy visual hull has high robustness for noises independently added to silhouette images.

In this paper, a fuzzy identification method for cellular quantity of Bone Marrow Stromal Cells (BMSCs) in artificial culture bones is proposed. We attempt to identify for cellular quantity with an ultrasonic system and fuzzy inference approach. We employ two characteristics: the amplitude and the frequency; the amplitude is obtained from the raw ultrasonic wave, and the frequency is calculated from frequency spectrum obtained by applying cross-spectrum method. The experimental results show that the proposed method identifies cellular quantity in artificial culture bone with high accuracy. © 2008 IEEE.

In order to realize heart rate monitoring on a bed, there are mainly two types of approaches similar to other signal processing applications: frequency domain analysis and time-series domain analysis. In frequency domain analysis, FFT is widely used to extract heart rate from obtained signals. Since FFT assumes constant frequency, it cannot be used for extracting microscopic variability of heart rate. In time-series domain analysis, pattern matching based on autocorrelation is commonly used. The method is not only advanced in sensitivity to heart rate variability, but it is also sensitive to unexpected noise. In response to these problems, heart rate monitoring technology is proposed by using air pressure sensor. In this paper, a heart rate estimation algorithm employing fuzzy logic is proposed and effectiveness of fuzzy logic applied to biomedical sensing is discussed. The experiments were conducted to validate the effectiveness of the proposed technology by comparing it with other methods such as pattern matching based on autocorrelation. © 2008 IEEE.

We propose a thickness determination method of the intestine with the ultrasonic probe of 15MHz. We determine the surface point by calculating correlation coefficient between surface echo and an acquisition waveform. Next, we determine the bottom point by calculating amplitude of bottom echo, correlation with surface echo and bottom echo, and interval distance between surface echo and bottom echo. Finally, we calculate the thickness between the surface and the bottom points. As the result, we have obtained the thickness within an error rate of 5.09%. © 2008 IEEE.

To diagnose total hip arthroplasty (THA) patients, it is important to analyze in-vivo kinematics of the implanted hip joint. There are some conventional analysis methods for estimating the pose/position of the implanted hip joint using 2-D/3-D image registration technique for 3-D multidetector raw CT (MDCT). images and 2-D X-ray radiograph images. One of the remained problems is that estimation results depend on the initial pose/position given manually. This paper proposes a fully automated method for estimating pose/position of the implants and the hip joint without giving initial pose/position. The proposed method releases us from manual interactions of giving initial pose/position by intervening between 3-D MDCT images and 2-D X-ray radiograph images with 3-D computer-aided design (CAD) model.

This paper proposes a functional assessment system of autonomic nervous system using an air pressure sensor. The air pressure sensor can unconstraintly detect vital information by placing it under the mattress in bed. We perform functional assessment of autonomic nervous system by heart rate variability obtained by the system. In this system, we employ fuzzy membership functions with dynamic parameter to detect RR intervals. The experimental results show that we detect RR intervals with the correlation coefficient of 0.851 with comparison to that of electrocardiograph. Then the errors of the HF (index of parasympathetic system) and the LF/HF (index of sympathetic system) are 11.98% and 22.18%, respectively. © 2008 IEEE.

近年,動画像から物体の3次元形状を復元するための追跡手法が盛んに研究されている.現在,勾配法やブロックマッチングに基づく様々な追跡手法が提案されているが,数多くの特徴点を正確に追跡する事は困難である.本報告では,射影ボクセル空間を用いる事で密な点集合を追跡し,それらを用いて3次元形状を復元する手法を提案する.ここでは,形状復元において,復元した密な点集合から自動的に三角網を生成し,テクスチャマッピングに基づいて形状を表現する.これにより,従来法に比べ物体の形状はより少ない欠落で復元される.実験により,提案法の有効性を評価する.

We describe a new ultrasonography system, which can identify an implant position in bone. Although conventional X-ray fluoroscopy can visualize implants, it has the serious disadvantage of X-ray exposure. Therefore, we developed a system for orthopedic surgery that involves no X-ray exposure. Barriers to the development of the system were overcome using an ultrasonic instrument and fuzzy logic techniques. We located distal transverse screw holes in an intramedullary nail during surgery for femur fracture. The screw hole positions are identified by calculating two fuzzy degrees of intensity and the variance. Results allow this system to identify the screw hole positions within an error of 1.43 mm, an error ratio adequate for clinical surgical practice. Copyright © 2007 The Institute of Electronics.

In this paper, we describe a new ultrasonic oscillosensor and its application in a biological information measurement system. This ultrasonic sensor has a cylindrical tank of 26 mm (diameter) × 20 mm (height) filled with water and an ultrasonic probe. It detects the vibration of the target object by obtaining echo signals reflected from the water surface. This sensor can noninvasively detect the vibration of a patient by placing it under a bed frame. We propose a recognition system for humans in bed. Using this sensor, we could determine whether or not a patient is in the bed. Moreover, we propose a heart rate monitoring system using this sensor. When our system was tested on four volunteers, we successfully detected a heart rate comparable to that in the case of using an electrocardiograph. Fuzzy logic plays a primary role in the recognition. Consequently, this system can noninvasively determine whether a patient is in the bed as well as their heart rate using a constraint-free and compact device. Copyright © 2007 The Institute of Electronics, Information and Communication Engineers.

近年,カメラ画像処理による実物体3次元復元が盛んに研究されている.これまで画像中の照度情報から形状復元を行う "Shape from Shading",動画像の特徴点軌跡から特徴点の3次元位置を復元する因子分解法などが提案されている.本報告では光源付ビデオカメラを用いて多面構造をもつ対象を3次元復元する手法を提案する.提案法ではカメラに固定したライトの光を対象に照射して撮影を行う.そして,対象を異なる角度から撮影した動画像中の照度情報を用いて対象の面法線とカメラ視線方向を対応付ける.対象の面法線に対応するカメラ視線方向を因子分解法により復元したカメラ運動から推定することで対象の多面構造を復元する.これにより動画像のみから簡便に実物体の構造を実現する.実験を行い,その有効性を示す.

睡眠状態と眼球運動に強い関係があることはよく知られている.しかしながら,睡眠時において,脳活動の解析が可能な機能的MRIの撮影と同時に眼球運動を計測することは困難である.我々は以前,赤外線ビデオを用いた睡眠時における眼球運動計測方法を提案した.同方法は,ニューラルネットワークを用いて濃度プロファイルのモデル化を行うものである.しかしながら,ニューラルネットワークの学習データとして,赤外線ビデオから測定者が手動で眼球運動方向を計測しなければならず,非常に煩わしい作業であった.そこで,本研究では赤外線ビデオのあるフレームと同期した眼部MR画像を用いて閉眼時における視線を定義し,その視線角度を従来法のニューラルネットワークへの学習データとして用いる方法を提案する.提案法を覚醒閉眼状態の被験者に適応した結果,閉眼時においても赤外線ビデオから右眼0.89±2.91度,左眼1.10±3.60度の精度で視線角度推定が可能であることが確認できた.

ヒトの健康を維持するためには、継続的な生体情報の計測が非常に重要である。これまで、様々な生体情報の計測手法が提案されている。生体情報の中でも、特に睡眠中の心拍は注目を集めている。これは、心拍が健康状態を把握する上で複数の情報を含有しているためである。本誌では、簡便で低コストな空気圧センサを使用し、無意識/非侵襲で心拍を計測する手法を議論する。検証の結果、心拍の計測において、提案手法が十分な精度を示すことを確認した。

前十字靱帯(ACL; Anterior Cruciate Ligament)再建術後の動態解析による術後評価を行うため，我々は手術時に用いられるインプラントをランドマークとして利用する事で，画像位置合わせに必要な初期位置を自動で推定する手法を提案した．これにより，システムを全自動化でき，操作者に依存する誤差を削減する事に成功した．しかし，全自動であるが故に，位置合わせ手法自体の誤差要因による影響が大きくなってしまう．そこで本研究では，ファジィ関心領域を用いた画像位置合わせ手法を開発し，そのシステムにより膝の動態解析を行う．ファジィ理論を用いる事で医師の知識を導入し，従来手法よりも精度の高い位置合わせを行う．また，ファジィ関心領域情報を用いた位置合わせを行う事で，従来のランドマークベースの位置合わせ手法では合わせにくい領域に対しても良好な位置合わせを行う．

近年再生医療では親和性の高い人工培養骨と，骨芽細胞に分化する骨髄幹細胞(BMSCs)を組み合わせた混合物を骨補填剤として用いることが考えられている．しかしこの混合物内の細胞量を測定する方法は，BMSCsを注入した人工培養骨を粉砕し，電子顕微鏡を使用し目視で細胞を数えるのみであった．そのためこれらの混合物は測定以後臨床，研究に使用することができず，治療に最適な細胞量を知ることができなかった．そこで本研究では超音波装置を用いた人工培養骨内細胞量推定システムを提案する．本手法では上記の混合物に対して超音波を透過し，取得した波形から振幅と周波数の特徴値を得る．振幅値は取得波形から直接抽出し，周波数はクロススペクトル法により得られた伝達関数の周波数特性から算出する．これらの特徴値と従来法による真値の関係よりファジィ推論を構築し，細胞量を推定した結果高い成功率で推定を行うことができた．

MR画像を用いた脳診断において，頭蓋除去は基本的な画像処理技術の1つである．従来，成人のMR脳画像を対象とした頭蓋除去法が多く提案されている．しかし，新生児および幼児のMR画像は成人のMR画像と画像特徴が大きく異なり，頭蓋は脳表近くにある．このため，成人脳を対象とした従来法では，新生児や幼児脳に適用することができない．提案手法では，新生児および幼児MR画像から自動頭蓋除去法を提案する．提案手法は，ファジィIF-THENルールを用いた医学知識に基づくFuzzy rule-based Active Surface Model (FASM)を適用する．8例の新生児および幼児頭部MR画像に対して提案手法を適用した結果，提案手法はSensitivityが98,84%とFPR (false-positive rate)が1.21%の精度で脳領域を抽出することが確認できた．

経頭蓋超音波装置は非侵襲かつリアルタイムに脳画像を作成できるが，超音波は頭蓋骨内で屈折するため任意の箇所からの画像化は行えていない．そこで本論文では，牛の肩甲骨と擬似生態ファントムを用いて，超音波の屈折を考慮した画像化法を提案する．肩甲骨中での超音波の屈折を求めるためには，各箇所における肩甲骨の厚みを求める必要がある．しかし，肩甲骨表面からの反射波は容易に決定できるが，底面からの反射波を一意的に決定することは難しく，直接厚みを推定することは困難である．そこで，本手法では肩甲骨底面からの反射波を仮定し，注目波形と肩甲骨の表面反射波との相関，注目波形の振幅値，表面反射波から注目波形までの経過時間を考慮し，ファジィ推論を用いることで肩甲骨の厚みを推定する．本方法を適用した結果，肩甲骨の厚みを全ての箇所において推定することができ，任意の箇所からファントムの表面形状を画像化することができた．

現在，バイオメトリクス個人認証はセキュリティや個人サービスの分野で急速に発展している．ロードセルやカメラ，圧力センサなどを用いた歩行パターンによる個人認証は利用者の負担を減らすことのできる認証法である．特に，マット型の圧力センサは足のサイズや圧力分布などの多くの情報を得ることができる．本論文では，マット型圧力センサを用いた足底圧変化による個人認証を提案する．本手法では，圧力分布区域をいくつかの区間に分割し，それぞれの圧力が最大になる時間や踏み込み，踏み抜き時間のような特徴を用いて個人認証を行う．認証にはユークリッド距離と，ニューラルネットワークを用いる．結果より，ユークリッド距離を用いた認証では4分割で10人を70%の精度で認証を行うことができ，ニューラルネットワークによる認証では5分割で85%の精度で認証を行うことができた．

大腿骨を骨折した際，骨髄中に髄内釘を挿入し固定することで治療を行う．髄内釘の横止め穴にスクリューを差し込むことで固定を行う．現在では，X線を用いて骨内部を画像化して横止め穴位置を決定しているが被曝の危険性がある．そこで本論文では超音波アレイプローブを用いたフリーハンド走査による横止め穴位置決定法を提案する．まず超音波波形より振幅値を特徴値とした特徴値マップを作成する．その画像からファジィ推論を用いて横止め穴領域を抽出する．ファジィ推論には特徴値マップにおいて，8近傍の輝度値の平均値及び髄内定の長軸方向に沿った軸ごとの輝度値の分散という2つの知識を用いる．そして，取得画像と髄内釘の真値となる登録先画像を用いて位置合わせを行い，横止め穴位置を決定する．本手法を適用した結果， 2つの横止め穴の中心間距離を1mm以下の誤差で検出でき，それぞれの横止め穴位置を髄内釘の中心軸上で検出できた．

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. © 2007 IEEE.