Image Difference Method Research Articles

160 MicroRNA-199a-5p is Elevated in Experimental Hepatic Fibrosis and Its Inhibition Attenuates Procollagen Type I Expression in Hepatic Stellate Cells

Background: Staging of liver fibrosis is an essential part in the management of chronic liver disease. Invasive screening tests like liver biopsy, hepatic venous pressure gradient (HVPG) measurement and Upper-GI-Endoscopy have a significant burden on patients. We compared two different noninvasive methods of liver stiffness measurement (LSM), Transient Elastography (TE; FibroScan®) and Acoustic Radiation Force Impulse Imaging (ARFI; ACUSON Siemens 2000®), with HVPG and histology. Methods: Forty-eight patients (mean age 53±13) with different etiologies of chronic liver diseases were included. Liver cirrhosis (F4) and fibrosis (≥F1) was histologically diagnosed in 75.0% and 22.9%, respectively. Clinically significant portal hypertension (CSPH; HVPG≥10mmHg) was present in 56.2%. Results: Invalid results of TE and ARFI were obtained in 14 patients (29.2%) and 1 patient (2.1%), respectively, because of ascites and obesity (BMI 29.00 vs. BMI 25.57; p<0.05) HVPG correlated significantly with TE (r=0.754; p<0.01) and ARFI (r=0.709; p<0.01). Furthermore, TE correlated significantly to ARFI (r=0.757, p<0.01). Patients with CSPH had significantly higher LSM via TE (47.03kPa vs. 15.92kPa; r=0.649; p<0.01) and via ARFI (2.92m/s vs. 1.91m/s; r=0.497; p<0.01) Area under the receiver operating curve (AUROC) for diagnosis of CSPH was 0.896 (CI 0.791-1.001; p<0.01) for TE and 0.874(CI 0.7480.999; p<0.01) for ARFI. Patients with cirrhosis had significantly higher LSM with TE (42.6kPa vs. 8.0kPa; r=0.806; p<0.01) and with ARFI (2.75m/s vs. 1.48m/s; r=0.536; p<0.01) than patients with fibrosis. AUROC for diagnosis of cirrhosis was 0.955 (CI 0.891-1.018; p<0.01) for TE and 0.849 (CI 0.703-0.995; p<0.01) for ARFI. Mean HVPG in cirrhosis was 16.32±8.81mmHg compared to 3.00±1.63mmHg in non-cirrhotic patients. For diagnosing cirrhosis, a cut-off value of 16.0kPa for TE shows a sensitivity of 86% , a specificity of 100% (PPV 100%; NPV 69%). A cut-off value of 1.71m/s for ARFI shows a sensitivity of 91%, a specificity of 82% (PPV 94%; NPV 80%). Conclusion: Both TE and ARFI correlated with HVPG and showed a similar performance for the evaluation of CSPH. TE may be superior to ARFI for the diagnosis of liver cirrhosis. Possible advantages of ARFI are superior feasibility in obese patients and patients with ascites.

Quantum diffraction of biphotons at a blazed grating

Correlations between photons are interesting for a number of applications and concepts in metrology in particular for resolution improvements in different methods of quantum imaging. Since Fock-states of N-photons of wavelength lambda in interference schemes acquire N-times the phase shift of single photons these states can appear as if they had a de Broglie wavelength of lambda/N. A biphoton beam diffracted at a blazed grating shows this reduced de Broglie wavelength. This experiment can be seen as a purification of biphotons of a certain correlation strength on the one hand. On the other hand the evaluation of the one-photon and two-photon rate distributions in the Fraunhofer far field of the grating allows for an analysis of the spatial correlation between the photons. An experimental demonstration of these ideas tested for a biphoton beam generated by parametric down conversion (PDC) is reported. We demonstrate in addition that the existence of higher order spatial modes is important to observe strong spatial correlations and to observe the photonic de Broglie wavelength of lambda/2 for biphotons.

Independent Component Analysis-Based Background Subtraction for Indoor Surveillance

In video surveillance, detection of moving objects from an image sequence is very important for target tracking, activity recognition, and behavior understanding. Background subtraction is a very popular approach for foreground segmentation in a still scene image. In order to compensate for illumination changes, a background model updating process is generally adopted, and leads to extra computation time. In this paper, we propose a fast background subtraction scheme using independent component analysis (ICA) and, particularly, aims at indoor surveillance for possible applications in home-care and health-care monitoring, where moving and motionless persons must be reliably detected. The proposed method is as computationally fast as the simple image difference method, and yet is highly tolerable to changes in room lighting. The proposed background subtraction scheme involves two stages, one for training and the other for detection. In the training stage, an ICA model that directly measures the statistical independency based on the estimations of joint and marginal probability density functions from relative frequency distributions is first proposed. The proposed ICA model can well separate two highly-correlated images. In the detection stage, the trained de-mixing vector is used to separate the foreground in a scene image with respect to the reference background image. Two sets of indoor examples that involve switching on/off room lights and opening/closing a door are demonstrated in the experiments. The performance of the proposed ICA model for background subtraction is also compared with that of the well-known FastICA algorithm.

Near-infrared and short-wavelength autofluorescence imaging in central serous chorioretinopathy

Aims:The aim of the study was to compare the results of short-wavelength (SW) and near-infrared (NIR) autofluorescence (AF) in acute central serous chorioretinopathy (CSC).Methods:Twenty-six eyes of the 26 patients diagnosed...

Vegetation cover changes and their relationship to climate variation in the source region of the Yellow River, China, 1990–2000

The change history of vegetation cover and its relations to growing season precipitation (GSP) and average growing season temperature (AGST) in the source region of the Yellow River (SRYR) during 1990–2000 was retrieved based on the 1 km Advanced Very High‐Resolution Radiometer (AVHRR) Normalized Difference Vegetation Index (NDVI) data and meteorological records. The results show an overall warming and drying trend of the climate and a common degradation tendency of the ecosystem, with a greening trend in higher rugged regions. The pixel‐by‐pixel correlations between NDVI and climate factors indicate that a decrease in GSP mainly affects ecosystems with low precipitation and worse vegetation condition, and superimposes on the effects of increasing AGST which further deteriorate the climate background of these ecosystems. However, the positive correlations between AGST and NDVI in some higher/rugged regions suggest that the raising temperature can ameliorate vegetation growth conditions in these areas. Comparison and combination of the results of three change detection algorithms, i.e. post‐classification comparison (PCC), principal components analysis (PCA) and a newly developed multi‐temporal image difference (MTID) method, show that the integration of different methods can give a more comprehensive understanding of vegetation changes than any single method.

MAGNETIC RESONANCE IMAGING OF MULTIPLE SCLEROSIS

ABSTRACT MRI has revolutionized the diagnosis of multiple sclerosis (MS). In addition, use of MRI has provided a means to monitor disease and the response to therapy and has been used extensively as an adjunct means to evaluate new therapies. Despite its high sensitivity to detect MS lesions, its specificity for the underlying pathology is low. Furthermore, correlation of MS disability with the extent of abnormalities using several different methods of imaging has been modest at best. This chapter discusses the main imaging techniques used for MS diagnosis, MS management and treatment monitoring, and in clinical trials.

Search by X-rays applied technology

X-rays and gamma-rays are part of different methods of imaging for control purpose. The main and most widely used method is conventional fluoroscopy with transmission images. Dual energy imaging, backscatter imaging and computed tomography allow visualizing of interesting items or of single layers without superposition; these methods have been developed recently or have been adapted to imaging for security reasons. Dual energy uses the specific absorption characteristics of a substance. Spectroscopy permits the identification of substances in packages or luggage, without direct access. Industry offers technical solutions which combine different methods. Comparing the different methods, one has to conclude that the disadvantage of the transmission image is due to superimposition, which limits the recognition of searched items. The combination with other methods increases the reliability of recognition. Future will bring more controls for security reasons and in consequence, more “control imaging” by X-rays. Imaging for security reasons will be combined with pattern recognition; politicians will answer the demand for security and impose more controls.

METHODS OF IMAGING IN THE DIAGNOSIS OF TEMPOROMANDIBULAR JOINT DISORDERS

Different methods of temporomandibular joint imaging are presented and discussed. Research reports published from 1979 to 2002 have been evaluated. The aim of this paper is to summarize the main findings from research. Basic X-ray examination is the most readily available method of imaging which usually does not have any contraindication. The use of computer tomography offers all advantages of tomographic scanning in different layers and projections, imaging soft tissues close to the joint and the possibility of "3D" reconstruction of bone structures. In case of joint dysfunctions and internal joint derangement, it is more preferable to use nuclear magnetic resonance for the depiction of the joint structures. To make a diagnosis more efficient, miniinvasive diagnostic methods are becoming necessary. Arthroscopy allows direct visual control of the joint space with the possibility of therapeutic help in cases when conservative treatment has failed. Ultrasonography, as a completely non-invasive procedure, is applied even in diagnosing functional temporomandibular defects. In these cases the diagnostic value of ultrasonography is almost comparable with this of nuclear magnetic resonance. So it is possible that, together with the improvement of the equipment, development of the diagnostic methods used during the treatment of temporomandibular joint defects could continue in this direction.

Which radiographic techniques should we use for research and clinical practice?

Based upon published data, the different methods of imaging the knee joint are evaluated with respect to the radiographic procedure and anatomical structures recorded in: (1) images from radiographs of the tibiofemoral joint that have been obtained from the non-weight-bearing and weight-bearing extended knee views and the standardized knee flexion positions using fluoroscopy (semi-flexed and Lyon Schuss views) and non-fluoroscopic methods (MTP and fixed flexion); and (2) images from radiographs of the patellofemoral joint obtained from the lateral and the different axial views of the joint. Non-weight-bearing and weight-bearing radiographs of the knee in extension were found to be of limited value in assessing disease status, whereas all standing flexed knee positions reliably imaged joint space width and bone changes in the tibiofemoral joint. Fluoroscopic positioning of the joint is more demanding on equipment and personnel than non-fluoroscopic methods. Skyline rather than lateral views of the patellofemoral joint were better at detecting joint changes in osteoarthritis. It is concluded that for both clinical practice and research purposes the non-fluoroscopic MTP and fixed flexion methods are recommended for imaging the tibiofemoral compartment, and the standing skyline view is recommended for the patellofemoral joint.

Two Unusual Projection Spaces

Two immersive projection environments, both unconventional, both exploring different methods of 3D and panoramic imaging, and both produced as art installations, are described. Displacements (1980–1984) recreated an interior living space using a panoramic motion picture method and relief projection. Be Now Here (1995–1997) recreated outdoor public plazas using a panoramic motion picture method, stereopsis, and four channel sound. Both installations were unusual in that no intentions existed for anything more general or useful than the installations themselves as individual artworks.

다중 비디오카메라에서 색 정보를 이용한 특정 이동물체 추적 알고리듬

본 논문은 다중 비디오카메라에서 색 정보를 이용한 특정 이동물체 추적 이동물체 추적 알고리듬을 제안한다. 제안된 방법은 다중 비디오카메라로부터 입력되는 영상의 RGB 칼라 좌표계를 HSI 칼라 좌표계로 변환한 후, 영상의 색조 영역만을 가지고 배경영상과 물체가 존재하는 영상에서 차영상 기법과 가산투영 기법을 사용하여 이동물체를 검출한다. 검출된 이동물체 영역의 색조는 0도부터 360도 사이에서 24단계로 정규화 된다. 정규화된 이동물체의 색조 분포도를 구한 후, 가장 높은 분포를 갖는 3개의 정규화 레벨과 3개의 정규화 레벨 사이의 간격을 이동물체의 특징파라미터로 사용하였다. 각 카메라간의 이동물체 동일성 관별은 이동물체 특징파라미터를 가지고 판별하고, 추적 감시하였다. 제안된 방법의 타당성을 검토하기 위하여 실내에 각기 다른 장소에 4대의 카메라를 각각 설치하여 이동물체의 대상을 사람으로 놓고, 특정사람을 감시한 결과 각 카메라에서 검출된 특정사람의 색조분포도 변화는 10％내를 유지함을 보였고, 특징 파라미터로 4대의 카메라에서 특정사람이 자동 추적감시 됨을 확인하였다. In this paper, the interesting moving objects tracking algorithm using color information on Multi-Video camera is proposed Moving objects are detected by using difference image method and integral projection method to background image and objects image only with hue area, after converting RGB color coordination of image which is input from multi-video camera into HSI color coordination. Hue information of the detected moving area are normalized by 24 steps from 0<TEX>$^{\circ}$</TEX> to 360<TEX>$^{\circ}$</TEX> It is used for the feature parameters of the moving objects that three normalization levels with the highest distribution and distance among three normalization levels after obtaining a hue distribution chart of the normalized moving objects. Moving objects identity among four cameras is distinguished with distribution of three normalization levels and distance among three normalization levels, and then the moving objects are tracked and surveilled. To examine propriety of the proposed method, four cameras are set up indoor difference places, humans are targeted for moving objects. As surveillance results of the interesting human, hue distribution chart variation of the detected Interesting human at each camera in under 10%, and it is confirmed that the interesting human is tracked and surveilled by using feature parameters at four cameras, automatically.

Monitoring disease activity and progression in primary progressive multiple sclerosis using MRI: sub-voxel registration to identify lesion changes and to detect cerebral atrophy.

To explore the potential usefulness of two new magnetic resonance imaging (MRI) analysis techniques for assessment of progressive cerebral atrophy and T2 lesion activity in primary progressive multiple sclerosis (PPMS), and thereby assess the relationship between MRI activity and atrophy in this patient group. Measurements of cerebral atrophy and net change in T2 lesion volumes are currently used as surrogate markers of disease progression in multiple sclerosis (MS). However, manual implementation of these techniques is time-consuming and the pathological specificity of T2 lesion change is low. Advances in serial scan registration have facilitated the development of a new, fully-automated technique to measure cerebral volume (SIENA; Structural Image Evaluation, using Normalisation, of Atrophy), and a technique to measure the total new T2 lesion volume selectively (MRI difference imaging). SIENA measures changes in cerebral size based on sub-voxel detection of shifts in edge contours. The lesion difference imaging method measures differences in lesion volumes over time as defined by a semi-automated outlining technique. The two new methods were validated against the T2 lesion volume contour technique and a previously described measure of partial brain volume (which uses six slices centred on the presumed area of greatest change around the lateral ventricles). All were applied to serially acquired MR images from a cohort of 39 patients with PPMS, who also underwent scoring on the expanded disability status scale (EDSS) twice, two years apart. The two measures reflecting cerebral atrophy correlated strongly (r = 0.58, p < 0.001). T2 lesion load measurements using the two techniques correlated very highly (r = 0.999, p < 0.001). 91% of the total new T2 lesion volume was from enlargement of pre-existent lesions and only 9 % from new, discrete, lesions. No relationship was seen between the traditional measure of net gain in T2 lesion load and either measure of atrophy. However, the fully-automated measure of total new T2 load correlated with both measures of atrophy (SIENA technique, r= -0.37, p= 0.02; six slice measure, r = -0.41, p = 0.01). There was no relationship between the MRI measures and changes in the EDSS. Both of the new image analysis techniques appear to be promising as sensitive markers for disease progression in PPMS. The correlation of total new T2 lesion volume with the progression of cerebral atrophy (which is known to be a consequence of axonal loss in progressive disease), compared with a lack of correlation with the traditional net gain in T2 lesion load is interesting and suggests that the total new T2 lesion volume may ultimately be the most useful measure.

Neuromagnetic integrated methods tracking human brain mechanisms of sensorimotor areas ‘plastic’ reorganisation

The potential for reorganization in the adult brain has been largely underestimated in the past and we are just beginning to understand the organisational principles involved in functional recovery. A bulk of experimental evidences have been accumulated in support of the hypothesis that neuronal aggregates adjacent to a lesion in the cortical brain areas can be progressively vicarious to the function of the damaged neurones. Such a reorganisation, if occurring in the affected hemisphere of a patient with a monohemispheric lesion, should significantly modify the interhemispheric symmetry of somatotopic organisation of the sensorimotor cortices, both in terms of absolute surfaces and number of “recruited” neurons, as well as of spatial coordinates. In fact, a roughly symmetrical organisation of sensorimotor — particularly for the hand contorl — in the right and left hemisphere has been observed in healthy humans by different methods of functional brain imaging, including fMRI, TMS, MEG, HD-EEG. Not uniform results about the functional brain activity related to sensory, motor and cognitive functions in normal and diseased subjects are often due to differences in the experimental paradigm designed as well as in the spatial and temporal resolution of the neuroimaging techniques used. The multi-modal integration of data obtained with several neuroimaging techniques allowed a coherent modelling of human brain higher functions. Functional magnetic resonance imaging (fMRI) provided fine spatial details (millimetres) of the brain responses, which were compared with the cortical maps of the motor output to different body districts obtained with transcranial magnetic stimulation (TMS). Magnetoencephalography (MEG) ability to study sensorimotor areas by analysing cortical magnetic fields, is also complementary to the motor cortex topographical mapping provided by TMS. MEG high temporal resolution allows to detect relatively restricted functional neuronal pools activated during cerebral processing of external stimuli. Moreover, these brain responses can be investigated with magnetoencephalography (MEG) and high density electroencephalography (EEG) techniques, with elevated time resolution (ms). With respect to the high resolution EEG technique, the MEG technique allowed a more precise localisation of the sites of neural activity buried into the cortical sulci, but was unable to detect the response of the crown of the cortical giri and of the frontal-mesial cortex (including the supplementary motor area), because of its poor sensitivity to radially oriented dipoles. The integration of functional and anatomical information provide cues on the relationship between brain activity and anatomic sites where this takes place, allowing the characterisation of the physiological activity of the cortical brain layers as well as to study the plastic reorganisation of the brain in different pathological conditions following stroke, limb amputation, spinal cord injury, hemisperectomy.

Visualization of Retinal Proteins in the Cerebral Ganglion of Ascidian, Halocynthia roretzi

Abstract Retinal proteins in the cerebral ganglion of the ascidian, Halocynthia roretzi, were successfully visualized and their localization was determined by the time-resolved fluorescence difference imaging method. This visualizes retinal proteins in the tissue even though the concentration of the retinal proteins is low and there is considerable endogenous fluorescence. Heterogeneous retinal proteins in the same tissues could be distinguished by differential sensitivities to photobleaching, standardized by rhodopsin and retinochrome in octopus retina. Retinal proteins in the ocellus of ascidian larva, which is composed of only about 20 photoreceptor cells, were successfully visualized. Retinal proteins in the cerebral ganglion of adult Halocynthia roretzi, localized mainly at the surface of the anterodorsal root and the posterodorsal root. In the cross sections along the anteroposterio axis of the cerebral ganglion, the cells bearing retinal proteins were found in the peripheral cellular cortex mainly ...

Contemporary role of X-ray devices in medical introscopy

Visual information accounts for at least 90% of the information received by human sensory systems. At least 65% of brain systems are involved in processing visual information. As a result, many trends in the development of medical diagnostic technologies are based on visual presentation of diagnostic information. The goal of making the patient's body "transparent" has given rise to speedy progress in medical introscopy, a branch of medicine that employs all types of physical techniques for the purpose of imaging of internal organs of the human body. Electromagnetic radiation and ultrasonic waves are most commonly used for obtaining images of nontransparent anatomical organs and systems inaccessible for direct visual inspection [4]. The extensive use of a broad range of physical methods in introscopic devices (X-ray apparatuses, ultrasonic scanners, endoscopes, ophthalmoscopes, thermal imagers, gamma chambers, and tomographs of various types) raises a number of important questions. I. Why are there so many types of imaging technologies? 2. Should any radiation diagnosis department use all these technologies? 3. What is the relative clinical value of various methods of introscopy, and X-ray introscopy, in particular? These are rather difficult questions, and there is no single conclusive answer to any of them. However, some points relevant to this issue can be clarified. Virtually all new methods of medical introscopy appear to supplement rather than to replace existing methods. As a matter of fact, different methods of imaging are based on different mechanisms of interaction between a physical field and biological tissue. Therefore, they provide information on different properties of biological structures. The main point is to distinguish visually between normal and pathological tissues. For many types of imaging, this is a problem that is presently far from being solved. It should be emphasized that different methods of imaging are mainly designed to supplement rather than replace each other. Only in individual cases are the results of imaging obtained by different methods are so closely correlated with each other that simultaneous use of the competitive methods should be avoided. For example, ultrasonic and endoscopic imaging in some cases impose limitation on X-ray examination. However, it should be clearly understood that in spite of adverse side effects of X-ray radiation, it is presently (and will probably be in the foreseeable future) indispensable in medical diagnosis. At the present time, about 80% of all medical images are obtained using X-ray radiation, and the diagnostic capacity of this radiation is far from being exhausted. An ideal X-ray converter should detect all incident quanta of information and provide accurate information on its position, energy, and time of detection. Also, the dynamic range of the ideal X-ray converter should be broad enough to provide accurate detection. No currently available commercially system of X-ray imaging meets all these requirements simultaneously. First, primary X-ray images are detected against a significant background of noise caused by radiation scattered within the object of examination. Second, some incident X-ray quanta are not detected. Third, in the absence of input signal, the system forms an image generated by internal noise. Fourth, the spatial resolution of modern X-ray imaging systems is limited by the characteristics of the device rather than by the cross-section of interaction between X-ray radiation and the object of examina-

553 Reorganization of somatosensory afferent projections after injury of the senso-motor cortex in men

The potential for reorganization in the adult brain has been largely underestimated in the past and we are just beginning to understand the organisational principles involved in functional recovery. A bulk of experimental evidences have been accumulated in support of the hypothesis that neuronal aggregates adjacent to a lesion in the cortical brain areas can be progressively vicarious to the function of the damaged neurones. Such a reorganisation, if occurring in the affected hemisphere of a patient with a monohemispheric lesion, should significantly modify the interhemispheric symmetry of somatotopic organisation of the sensorimotor cortices, both in terms of absolute surfaces and number of “recruited” neurons, as well as of spatial coordinates. In fact, a roughly symmetrical organisation of sensorimotor — particularly for the hand contorl — in the right and left hemisphere has been observed in healthy humans by different methods of functional brain imaging, including fMRI, TMS, MEG, HD-EEG. Not uniform results about the functional brain activity related to sensory, motor and cognitive functions in normal and diseased subjects are often due to differences in the experimental paradigm designed as well as in the spatial and temporal resolution of the neuroimaging techniques used. The multi-modal integration of data obtained with several neuroimaging techniques allowed a coherent modelling of human brain higher functions. Functional magnetic resonance imaging (fMRI) provided fine spatial details (millimetres) of the brain responses, which were compared with the cortical maps of the motor output to different body districts obtained with transcranial magnetic stimulation (TMS). Magnetoencephalography (MEG) ability to study sensorimotor areas by analysing cortical magnetic fields, is also complementary to the motor cortex topographical mapping provided by TMS. MEG high temporal resolution allows to detect relatively restricted functional neuronal pools activated during cerebral processing of external stimuli. Moreover, these brain responses can be investigated with magnetoencephalography (MEG) and high density electroencephalography (EEG) techniques, with elevated time resolution (ms). With respect to the high resolution EEG technique, the MEG technique allowed a more precise localisation of the sites of neural activity buried into the cortical sulci, but was unable to detect the response of the crown of the cortical giri and of the frontal-mesial cortex (including the supplementary motor area), because of its poor sensitivity to radially oriented dipoles. The integration of functional and anatomical information provide cues on the relationship between brain activity and anatomic sites where this takes place, allowing the characterisation of the physiological activity of the cortical brain layers as well as to study the plastic reorganisation of the brain in different pathological conditions following stroke, limb amputation, spinal cord injury, hemisperectomy.

Visibility enhancement of fluorescent objects hidden in animal tissues using spectral fluorescence difference method

The difference between the fluorescence spectra emitted from Cardio Green dye and chicken breast tissue in the near-infrared (NIR) region was measured and used to select a pair of fingerprint wavelengths for imaging. Using spectral fluorescence difference imaging method, an 1-mm tissue object stained with Cardio Green dye located 3.1 cm underneath the surface of the host chicken tissue was clearly imaged and identified. The spatial resolution and the contrast of the fluorescence difference image were found to be significantly improved over the conventional single wavelength imaging approach.

Improved video-based eye-gaze detection method

Recently, some video-based eye-gaze detection methods used in eye-slaved support systems for the severely disabled have been studied. In these methods, infrared light was irradiated to an eye, two feature areas (the corneal reflection light and pupil) were detected in the image obtained from a video camera and then the eye-gaze direction was determined by the relative positions between the two. However, there were problems concerning stable pupil detection under various room light conditions. In this paper, methods for precisely detecting the two feature areas are consistently mentioned. First, a pupil detection technique using two light sources and the image difference method is proposed. Second, for users wearing eye glasses, a method for eliminating the images of the light sources reflected in the glass lens is proposed. The effectiveness of these proposed methods is demonstrated by using an imaging board. Finally, the feasibility of implementing hardware for the proposed methods in real time is discussed.

Sophisticated instrumentation and ophthalmic ultrasonography.

Formerly, ophthalmic ultrasonography was leading in view of high-performance apparatuses and transducer probes: e.g., the first array-scanner in the world was built for ophthalmic use. Within the past 2 decades, however, high-tech innovations were merely developed for other medical specialties. These were studied in view of their use for ultrasonography of eye and orbit. The combination of B-scan and Doppler techniques facilitates detection of orbital vessels. The resolution of ophthalmic digital B-scan video images proved poorer than crt-B-scans. A digital memory, however, is advantageous. But one high-resolution crt-type B-scan needs more than one disc storage capacity. "Frontline digitalization" could help to reduce the amount of data. Array transducers are now available in small sizes and could better show structure movements, but they were not yet adapted to ophthalmic use. This applies as well to annular arrays and dynamic focusing. Different methods of 3-dimensional scanning and (Pseudo-) 3-dimensional imaging might renew Baum's and Coleman's early work.

Unconstrained Pupil Detection TechniqueUsing Two Light Sources And The ImageDifference Method

For developing a human-computer interface applying eye-gaze, we have already proposed a noncontact, unconstrained video-based pupil detection technique using two light sources and the image difference method. The detected pupil position in the difference image is utilized together with the glint (corneal reflection light of an infrared light source) position for eyegaze position determination. In this paper, the hardware for real-time image differentiation was developed. This image differenciator made realtime pupil possible, by applying the pupil detector including the noise reducer, which had been already developed. For stably detecting the glint and the pupil, it was clarified that the pupil brightness is influenced by the pupil area and the power of infrared light irradiated for the eye. For stabilizing the pupil brightness in the difference images, a method which utilized this characteristics was proposed. This method made pupil and glint center detection stabler.