Medical endoscopy has become a key technology for the purposes of minimally invasive diagnostic and therapeutic procedures. With the aid of an endoscope, and external observation of a location inside the human body which may be difficult or dangerous to reach or which should be subjected to minimal traumatization associated with the intended medical treatment is facilitated through a viewing channel. A television camera is usually attached to the distal end of an endoscope for image presentation. Of primary importance for the user are image quality, ease of application and safety. This communication is primarily concerned with aspects relating to image quality. First, the optical properties of various types of endoscopes are analyzed and their spatial resolution is determined. Second, it is shown that the inherent optical quality in terms of spatial, temporal and contrast resolution of most endoscopes exceeds the performance limits of presently used standard TV systems in part substantially. Accordingly, a special purpose advanced digital television system exhibiting high spatial resolution, furthermore true color and high contrast dynamics has been developed. Since endoscopic images usually are circular, its aspect ratio is quadratic with a spatial resolution of 1000 x 1000 pixels. The framing rate is variable from 1 Hz to 30 Hz noninterlaced. Some design aspects and the performance of the system are shown and further applications are indicated.

A novel digital high-definition TV (digital HDTV) system is presented which is adapted to the special characteristics of medical endoscopes. In particular, it has a quadratic image aspect ratio to accommodate round endoscopic images, furthermore, the spatial resolution is 1024 x 1024 pixels thereby approaching the diffraction limit of small endoscopes. It exhibits 24 bits true-color with an adjustable temporal resolution of up to 30 full frames per second in progressive scan. To ensure overall superior performance and high-quality image acquisition and reproduction, complex digital error correction and picture enhancing algorithms are integrated into custom ASICs (Application Specific Integrated Circuit). Additionally, an improved color space transformation is performed in real-time to match the spectral characteristics of the digital camera and the viewing device, allowing for a quantitative judgement of colors. Further features include the conversion of the digital video stream to standard video norms such as Pal or NTSC for recording on analog VCRs, the calculation and evaluation of a focus criterion, which is used to perform passive, stable and reliable auto-focusing and the implementation of an automated illumination control, ensuring proper picture brightness during the whole time of operation.

Small bowel endoscopy with existing endoscopes is limited by problems of discomfort and the technical difficulty of advancing far into the small-bowel. Our aim has been to develop and test wireless capsule endoscopy. Wireless endoscopes, in the form of capsules (11 x 33 mm), were constructed by Given Imaging. These were powered by silver oxide batteries and each contained a CMOS imaging chip and miniature processor, white light emitting diodes (LEDs), a short focal length lens, and a miniature transmitter and antenna. Two video frames per second were transmitted, using radio-frequency (approx. 410 MHz), to an array of aerials attached to the body. The array of aerials can also be used to calculate the position of the capsule in the body. The images were stored on a portable recorder carried on a belt and subsequently downloaded for analysis. The batteries allow more than 5 hours of recording, although the capsule generally passes through the whole small bowel in under two hours. Clear video images of the human bowel were recorded from the pylorus to the caecum. Wireless endoscopy, for the first time, allows painless optical imaging of the whole of the small bowel.

Background: Methods are required for propulsion of endoscopes through the small bowel and for propelling capsule endoscopes without cables. Aim: To test the hypothesis that electrical stimulation could propel an endoscope by stimulating muscular contraction. Methods: Prototype acrylic devices of ovoid shape were constructed with two stainless steel electrodes mounted on the tapered section. Five devices of 13 to 23 mm diameter with a taper of 16 degree(s) to 20 degree(s) (half angle) were tested. When in contact with the bowel wall electrostimulation was applied causing circular muscle contraction which when applied to the taper of the ovoid resulted in forward propulsion of the device. The method does not induce peristalsis but works by stimulating local contraction. The device was tested in small bowel and oesophagus of anaesthetized pigs. Results: Electrostimulation caused the ovoid to advance rapidly (6 mm/sec) up and down the oesophagus by inducing circular esophageal muscle contraction. When stimulated at 15 Hz with 30 ms pulses the threshold for movement was 12 mA; at 20 mA the device moved reliably in both directions in the small bowel at speeds of up to 4.5 mm/s, negotiating tight curves.

Flexible endoscopes currently used in medicine have a fundamental tradeoff. Either resolution or field of view (FOV) is sacrificed when the scope diameter is less than 3 mm, since the minimum pixel size is usually greater than 4 microns in a pixel-array such as a camera or fiber bundle. Thus, the number of pixels within the image plane determines the minimum size of a conventional scope. However, an image plane is not required for image acquisition using a scanning single-fiber scope. Both high resolution and wide FOV are possible in a scanning single-fiber scope of 1 to 2 mm in diameter. The technical challenge is to produce a two- dimensional scanned beam of light at the distal tip of the scope. By manipulating a resonant fiberoptic cantilever as the optical scanner, various 2-D scan patterns can be produced. The general design concepts and analyses of the fiberoptic scanner for scaling to small size and high resolution/FOV are reviewed. In our initial experimental tests, the size of the photon detector in a fiberoptic scanning scope is demonstrated to not affect image resolution, unlike existing endoscopes with pixel-based detector systems.

Although there is an ongoing effort to use infrared- spectroscopy for tissue differentiation and classification, there is still a need for the development of a sensor for use in the quick and non-destructive detection of spectral changes resulting from, for example, the influence of temperature. It is our intention to investigate flat IR- sensors and their suitability as a tool for detection within bio-organic compounds. Initial investigations using bio- liquids, tissues in the native and coagulated state, and micro-organisms have already been carried out. According to the literature, major IR-spectroscopic differences can be expected in the fingerprint region (1500-1000 cm^-1). An experimental set up was based on an FTIR-spectrometer, fiber optic cables, sensor and an external detector. A sensor was used whereby a segment of 1.5 cm at the center of the fiber was flattened to approx. 150 micrometers . Transmission spectra of flattened fibers have been compared to those of the cylindrical fibers. Spectral differences were obtained indicating a possible use for the study of thermal damage in tissues. The fiber-optic based results are comparable to micro-ATR-FTIR-spectroscopy (attenuated total reflectance) measurements. IR-sensing was performed either using circular or flat silver halide wave guides.

Owing to their low losses in the 2-12 micrometers region, the Te- As-Se glass fibers are used for infrared light transportation as well as sensing element based on evanescent wave absorption mechanism. The efficiency of the system is improved by tapering the fiber diameter in the sensing zone. With this kind of sensor, infrared spectra of biological tissue have been obtained. Spectra of mouse liver have been especially recorded in order to detect spectral differences between the healthy and the tumoral states of mouse liver.

Background and objectives: Laser soldering of tissues is based on the application of a biological solder on the approximated edges of a cut. Our goal was to use laser soldering for sealing cuts in skin under temperature feedback control and compare the results with ones obtained using standard sutures. Materials and methods: Albumin solder was applied onto the approximated edges of cuts created in rat skin. A fiberoptic system was used to deliver the radiation of a CO₂ laser, to heat a spot near the cut edges, and to control the temperature. Laser soldering was carried out, spot by spot, where the temperature at each spot was kept at 65-70 degree(s)C for 10 seconds. Results: The tensile strength of laser-soldered cuts was measured after 3-28 days postoperatively and was found comparable to that of sutured cuts. Histopathological studies showed no thermal damage and less inflammatory reaction than that caused by standard sutures. Conclusions: Temperature controlled laser soldering of cuts in rat skin gave strong bonding. The cosmetic and histological results were very good, in comparison to those of standard sutures.

The authors introduce the design of a diode-pumped solid- state laser phototherapy unit, capable of fulfilling the recommendations of the clinical use of laser in phototherapy for the treatment of neonatal hyperbilirubinemia. This compact laser power unit has the advantages of easy application by the medical staff plus being more cost effective.

The authors introduce the design of a blue-green diode- pumped solid-state laser system for transcutaneous measurement of serum bilirubin level in jaundiced new born infant. The system follows the principles of optical bilirubinometry. The choice of wavelengths provides correction for the presence of hemoglobin. The new design is more compact and less expensive.

We proposed a use of hollow fibers for delivery of harmonic pulses of high-peak power, Nd:YAG laser light Hollow fibers with an internal silver and polymer layer were fabricated by a liquid phase coating technique and, in an experiment using the 2nd-order Nd:YAG laser, it was shown that the fiber has low-loss property because of the smooth inner surface and proper coating materials. High-power capability of the fiber is also shown when the hollow core is purged with helium gas.

Cyclic olefin polymer-coated silver hollow glass waveguides with inner diameters of 540 micrometers , 700 micrometers , and 1000 micrometers were used for transmission of single pulses in picosecond region generated by a high-power Nd:YAG laser system (wavelengths 1.06 micrometers -first (fundamental) harmonic and 0.53 micrometers -second harmonic). Maximum first and second harmonic intensities delivered by 1000 micrometers inner diameter waveguide were 185 GW/cm² and 48 GW/cm², respectively. For fundamental radiation guided by the waveguide with the inner diameters 540 micrometers , and 700 micrometers the maximum measured transmitted intensities were 397 GW/cm² and 331 GW/cm², respectively. Temporal development with picosecond resolution, spatial profile, and divergence of delivered laser pulses were determined for each waveguide.

We present a novel all-optical-waveguide method for ultraviolet (UV), visible (VIS) and infrared (IR) laser delivery including a lens-free method of laser-to-fiber coupling using a simple uncoated glass hollow taper. Based on the grazing incidence effect, the hollow taper provides a way of direct launching, without any intermediate focusing elements, high power laser radiation into delivery fibers. Because of the mutual action of the nearly parallel laser excitation, the mode coupling process, and mode filtering effect, the hollow taper serves as a mode converter that transforms the highly multimode profile of the input laser emission into a high-quality Gaussian-shaped profile at the taper output. When the grazing incidence effect of the taper is applied to laser delivery, the maintenance of high reflectance coefficients in a wide spectral region allows to utilize the same uncoated hollow taper for laser radiation in the UV, VIS and IR ranges. Applying the experimental hollow-taper based delivery systems, we obtain high laser- to-taper and taper-to-fiber coupling efficiencies.

The effect of the mid-infrared laser radiation propagation through waveguides and fibers on the quality of the laser beam is investigated. Fluorocarbon fibers and metallic hollow waveguides with dielectric coating were investigated, using an Er:YAG laser (2.94 micrometers ), an HF laser (2.78 micrometers ) and a CO₂ laser (10.6 micrometers ). A comparison was made between the different waveguides and fibers, regarding the quality of the output beam.

A comprehensive theoretical model was developed to describe mid IR laser radiation propagation in an optical cylindrical waveguide. The effects of bending the waveguide on the transmission and beam shape were studied. The same parameters were used in the theoretical calculations as well as in the experiments. Good agreement between the theoretical and experimental results was found.

Chalcogenide glasses based on suplhides and selenides are very promising materials for various photonic applications, particularly for applications in medicine. Most of current optical fibers have been developed form ultrapure silica. While silica glasses are suitable for optical components in telecommunications they exhibit high losses beyond 2 micrometers - a wavelength range important for clinical practice. Thus special glass materials 9from which fibers could be drawn) should be developed for optical power transmission beyond 2 micrometers . The investigation and preparation of vitreous materials that include sulphide, selenide and selenide- tellurite glass systems together with fluoride and heavy metal oxide glasses on the base of ZrF₄, HfF₄ and GeO₂, TeO₂, PbO, respectively are being pursued in our laboratory. This research is aimed at the development of both passive and active (rare-earth doped) optical fibers. In this contribution we concentrate on the doping of chalcogenide glasses by rare earth elements (Er, Pr, Nd). Although the major role of these glasses is assumed in the development of laser power delivery systems for applications in surgery, dentistry, dermatology and ophthalmology, they can equally be used for the diagnostics of human tissues. An example of colon tissue autofluorescence will be given.

The optical sensor based on the dependence of the phosphorescence lifetime of palladium meso- tetraphenyltetrabenzoporphine on the oxygen concentration was developed. The polystyrene film doped with the dye is deposited on the tip of the fiber wave-guide (d=190 micrometers ) placed in medical needle (d=330 micrometers ). As an excitation source He-Ne laser is used with power 50 mW. The rectangular excitation pulses (pulse durability is of 20 microsecond(s) , durability of fronts- not higher than 1 microsecond(s) , pulse repetition is 0.5-1 Hz) are formed using mechanical obturators and are introduced in wave-guide. The outgoing back from wave-guide phosphorescence radiation is registered using photo detector. The formed electrical signal is processed using computer. The sensor is suitable for invasive oxygen tension measurements. The kinetics of the tumor oxygenation during light irradiation in photodynamic therapy (photosensitizer- sulphonated aluminium phthalocyanine) as well as of the oxygen consumption rate in the catalytic oxidation of sodium ascorbate in the aqueous buffer solution (pH=7.4) in presence of cobalt octacarboxyphthalocyanine as the catalyst are presented.

The results of technology research on high resolution fiber optic image guides (used in medical needle endoscopes) are presented. Work has been concentrated on optimization of image contrast and resolution of the image guide. Special attention has been paid to contrast enhancement and cross- talk phenomenon between pixels. The measurements of diffusion properties of different extra mural absorption (EMA) glasses have been carried out to find proper technology of the image guides manufacturing. As a result a comparison of transmission properties of different image guides is presented.

In the rapidly evolving field of intravascular ultrasound (IVUS) for tissue characterization and visualization, the assessment of vessel morphology still lacks a geometrically correct 3D reconstruction. The IVUS frames are usually stacked up to form a straight vessel, neglecting curvature and the axial twisting of the catheter during the pullback. This paper presents a comprehensive system for geometrically correct reconstruction of IVUS images by fusion with biplane angiography, thus combining the advantages of both modalities. Vessel cross-section and tissue characteristics are obtained form IVUS, while the 3D locations are derived by geometrical reconstruction from the angiographic projections. ECG-based timing ensures a proper match of the image data with the respective heart phase. The fusion is performed for each heart phase individually, thus yielding the 4-D data as a set of 3-D reconstructions.

The application accuracy of frameless stereotaxy depends partly on the accuracy of the patient-to-image registration procedure. We compared the application accuracy of registration procedures based on anatomical landmarks, surface matching, and adhesive markers. After acquisition of a 3D-MRI volume, 30 patients were subjected to all three registration procedures. Frameless stereotaxy was performed with the STN system (Carl Zeiss, Germany). Following each registration procedure, the root-mean-squared-error (RMSE) and the target registration error (TRE) of an extra adhesive marker (target) were recorded. The first represents the goodness-of-fit of the registration procedure (not available in surface matching) while the second represents the application accuracy. The mean TRE+/- SD for each type of registration was 5.3+/- 2.1mm, 9.4+/- 6.6mm, and 3.6+/- 1.6mm (paired t-tests: p<0.01). When anatomical landmarks were used, anterior targets generated smaller TREs than posterior targets (4.6+/- 2.0mm and 6.8+/- 1.3mm respectively, t-test: p<0.01). There was no significant correlation between the RMSE and the TRE (anatomical landmarks: R₂=0.071, adhesive markers: R₂=0.004). A more detailed evaluation of surface matching, using a plastic skull phantom, also could not demonstrate an improvement in application accuracy due to surface matching. In conclusion, our results indicate that adhesive markers offer the most accurate alternative to bonescrews. However, we believe anatomical landmarks to provide sufficient accuracy for many neurosurgical procedures concerning frontally located targets, reducing the need for extra preoperative imaging.

A volume rendering library is presented to interactively analyze volume data from modalities like CT, MR, PET, SPECT< and fMRI for the planning of nuerosurgical procedures. Current applications are logging of Penfield procedures, fMRI visualization, blood vessel visualization and interactive localization of EEG electrodes implanted in the subdural space of a patient with epilepsy.

Laparoscopic surgery is becoming increasingly popular as a form of minimally invasive surgery. A major problem in it is the smooth manipulation of the laparoscope. A surgeon is usually using both hands to manipulate long instruments, and it is difficult for a camera assistant to hold the laparoscope steady while quickly aiming the scope at the point required by the surgeon. A laparoscopic manipulator, which holds and manipulates the laparoscope according to direct commands from the surgeon, could solve these problems. Therefore, we have developed a laparoscopic manipulator system using a five-bar linkage mechanism and an optical zoom. A man-machine interface (MMI) is an important element in the laparoscopic manipulator system and it has special requirements as follows; Because the surgeon commands the manipulator at the same time with operating using his/her both hands, MMI, should be simple and intuitive method of command. The system should be safe as a medical device. Therefore, the aim of this study is to seek for the most suitable MMI for a laparoscopic manipulator system considering the requirements described above, for the safety and usability of the system. To achieve it, we develop several MMI and perform a quantitative experiment and in-vivo experiment.

The visualization of brain vessels on the cortex helps the neurosurgeon in two ways: To avoid blood vessels when specifying the trepanation entry, and to overcome errors in the surgical navigation system due to brain shift. We compared 3D T1 MR, 3D T1 MR with gadolinium contrast, MR venography and MR phase contrast angiography as scanning techniques, mutual information as registration technique, and thresholding and multi-vessel enhancement as image processing techniques. We evaluated the volume rendered results based on their quality and correspondence with photos took during surgery. It appears that with 3D T1 MR scans, gadolinium is required to show cortical veins. The visibility of small cortical veins is strongly enhanced by subtracting a 3D T1 MR baseline scan, which should be registered to the scan with gadolinium contrast, even when the scans are made during the same session. Multi-vessel enhancement helps to clarify the view on small vessels by reducing the noise level, but strikingly does not reveal more. MR venography does show intracerebral veins with high detail, but is, as is, unsuited to show cortical veins due to the low contrast with CSF. MR phase contrast angiography can perform equally well as the subtraction technique, but its quality seems to show more inter-patient variability.

Characterization of the proliferative activity of a tumor has been the subject of research for many years. The majority of the studies presented so far in the field of cytology and histology relates to the analysis of information from a limited number of cells, which are often easily distinguishable from the background and as well as from each other. The present paper introduces an automated image analysis technique for classification of cancer cell nuclei stained with proliferative markers. The images under processing were characterized by a high degree of complexity, containing considerable histological noise. The first step of the method aims to identify nuclear features of proliferating cells only, contained in large-scale histological images, using Principal Components Analysis (PCA). The histogram of the component that demonstrates the best contrast is processed appropriately for generating a binary image. Some standard morphological operations are then applied to remove any irrelevant structures and detect touching and/or overlapping nuclei. Two separate methods, Skeleton by Influence Zone and heuristic processing, are presented for segmentation of clustered cells. The algorithm was tested on tissue section images encountered in routine clinical practice with very encouraging results, after comparing image analysis and human observer cell counting.

This study concerns the identification of the state of human peripheral vascular tissue by using Artificial Neural Networks. The fluorescence spectra, obtained by dual wavelength excitation of the tissue samples, were passed through a non-linear filter, based on a High Order Neural Network (HONN). Then a classical Multi-Layer Perceptron was employed to serve as the classifier of the feature vector. The above process resulted in the successful discrimination between normal and different types of pathological tissue.

We propose navigational aid systems for the blind relying on active laser profilometry and infrared proximetry with a real time vibrotactile interface. The Teletact and the Vigitact are small hand held or badge worn devices to improve the spatial perception, the mobility and the security of blind people. The Teletact is a hand held laser telemeter and gives an accurate 3D spatial perception up to ten meters. The Vigitact is an infrared scanner and provides an automatic vigilance from knees to head up to two meters. Both devices are now commercially available. We will report on the basic functional parts of these devices, the results of everyday use by blind people, and future technological improvements.

A new digital video imaging system was developed and its performance was evaluated to analyze the spiral wave dynamics during polymorphic ventricular tachycardia (PVT) with high spatio-temporal resolution (1 ms, 0.1 mm). The epicardial surface of isolated rabbit heart stained with di- 4-ANEPPS was illuminated by 72 high-power bluish-green light emitting diodes (BGLED: (lambda) ₀ 500 nm, 10mw). The emitted fluorescence image (256x256 pixels) passing through a long-pass filter ((lambda) c 660nm) was monitored by a high-speed digital video camera recorder (FASTCAM-Ultima- UV3, Photron) at 1125 fps. The data stored in DRAM were processed by PC for background subtraction. 2D images of excitation wave and single-pixel action potentials at target sites during PVT induced by DC shocks (S2: 10 ms, 20 V) were displayed for 4.5 s. The wave form quality is high enough to observe phase 0 upstroke and to identify repolarization timing. Membrane potentials at the center of spiral were characterized by double-peak or oscillatory depolarization. Singular points during PVT were obtained from isophase mapping. Our new digital video-BGLED system has an advantage over previous ones for more accurate and longer time action potential analysis during spiral wave reentry.

The aim of the present study was to test the feasibility of the 980 nm diode laser for LEP (Laser Evoked Potentials) studies. Human subjects were exposed to laser stimulation. After the pain thresholds of the subjects were determined with respect to laser power level, 1.5 times the threshold value was applied and laser evoked potentials were recorded using standard EEG techniques. LEPs were obtained due to right hand stimulation. Latency and amplitude values of LEPs were found in accordance with those reported in the literature. Statistical evaluation showed differences in the LEPs at C3 and C4 locations as a function of the sex of the subjects. The power levels used in the present study was three times less than the levels applied for Nd: YAG laser in the literature. The evoked potential parameters measured were in consistence with the data reported by earlier researchers. Moreover, it was found that, LEPs due to 980 nm wavelength irradiation can be recorded by applying less energy when compared to Nd:YAG laser. This result indicated the potential of diode laser for LEP studies.

A new method of low contrast multispectral image sensitive analysis is outlined. The method is based on consideration of an analyzed image as a virtual rough surface. The one is illuminated with virtual pseudocoherent optical wave. Interaction of a reflected wave with a reference one gives possibility for synthesis of a new interferometric complex image. Analysis of a new image allows to detect very small brightness variations which are invisible for direct visual observations. Information possibilities of the new method are compare with known methods of histogram equalization and fuzzy C-means clustering. Experimental results are shown that new method provides a significant increasing in sensitivity and space resolving power of low contrast image analysis in comparison with known methods.

Detection of antigen is carried out by means of an optical immunosensor that uses laser reflectometry technique. It consists in a silicon wafer with antibody adhered on surface. Measurements were performed in samples containing different concentration of antigen A. This technique is very sensitive, quick and independent of cellular properties.

Many biological objects, including blood cells, are optically transparent. They don't absorb light and only change optical path length by variations of object's thickness or its refractive index variations. The phase shifting interferometry is used for the visualization of the phase information and the obtaining quantitative properties of cells. Cells are fixed by glutaraldehyde and mounted into immersion liquid for influences of refraction reduction. Blood cell's interferogram contains the information about space distribution refractive index. The phase-shift method was used for its reconstruction. This method includes record of four interferograms used different values of phase-shift of the reference wave. Many blood cells' interferograms are obtained (lymphocytes and erythrocytes). The shape, size and density of cells are measurement. We suppose that these date will be ones of the early diagnostically criteria of hematological diseases. Experimental results for interferometric microscopy and tomography of blood cells are presented.

In this work we have developed a new effective method to increase the probity of tissue structure classification based on X-Ray and X-RAY CT images. The proposed method requires much smaller resolution and it is also more sensitive to the small changes in the topology of tissue structure. As initial data, we used X-ray and X-ray CT images of coxa, affected with osteoporosis. The wave packet decompositions of a cut form the image were interpreted as an input and output of virtual non-linear system. The peculiarities of non-linear transfer characteristics were considered for performing a qualitative classification. The method helps to classify about 84% of data visually. A neuron network, which was used for classification based on this technique, has given an opportunity to classify correctly about 94% of all reviewed data. The described method was successfully applied to classification of other tissue.

Development of methods for indirect measurement of substance's consistence and characteristics is highly actual problem of medical diagnostics. Many diseases bring about changes of tissue density or appearances of alien bodies (e.g. stones in kidneys or gallbladders). Propose to use wavelet-analysis and neural nets for indirect measurement of relative density of tissue by images of internal organs. It shall allow to reveal a disease on early stage.