The full-field thickness distribution, three-dimensional surface model and general morphological data of six human tympanic membranes are presented. Crosssectional images were taken perpendicular through the membranes using a high-resolution optical coherence tomography setup. Five normal membranes and one membrane containing a pathological site are included in this study. The thickness varies strongly across each membrane, and a great deal of interspecimen variability can be seen in the measurement results, though all membranes show similar features in their respective relative thickness distributions. Mean thickness values across the pars tensa ranged between 79 and 97 μm; all membranes were thinnest in the central region between umbo and annular ring (50-70 μm), and thickness increased steeply over a small distance to approximately 100-120 μm when moving from the central region either towards the peripheral rim of the pars tensa or towards the manubrium. Furthermore, a local thickening was noticed in the antero-inferior quadrant of the membranes, and a strong linear correlation was observed between inferior-posterior length and mean thickness of the membrane. These features were combined into a single three-dimensional model to form an averaged representation of the human tympanic membrane. 3D reconstruction of the pathological tympanic membrane shows a structural atrophy with retraction pocket in the inferior portion of the pars tensa. The change of form at the pathological site of the membrane corresponds well with the decreased thickness values that can be measured there.

The acquisition of high-resolution, real-time three-dimensional surface data of dynamically moving objects has large applicability in many fields. When additional restrictions such as noninvasiveness and non-contact measurement are imposed on the employed profilometry technique, the list of possible candidates is reduced mainly to the broad range of structured light profilometry methods. In this manuscript, the current state-of-the-art in structured light profilometry systems is described, as well as the main advancements in hardware technology and coding strategy that have led to their successful development. A chronological overview of optical profilometry systems that have been reported to perform real-time acquisition, digital signal processing and display of full-field 3D surface maps is presented. The respective operating principles, strengths and weaknesses of these setups are reviewed and the main limitations and future challenges in high-speed optical profilometry are discussed.

Abstract. Fourier domain optical coherence tomography ͑FD-OCT͒ requires either a linear-in-wavenumber spectrometer or a computationally heavy software algorithm to recalibrate the acquired optical signal from wavelength to wavenumber. The first method is sensitive to the position of the prism in the spectrometer, while the second method drastically slows down the system speed when it is implemented on a serially oriented central processing unit. We implement the full resampling process on a commercial graphics processing unit ͑GPU͒, distributing the necessary calculations to many stream processors that operate in parallel. A comparison between several recalibration methods is made in terms of performance and image quality. The GPU is also used to accelerate the fast Fourier transform ͑FFT͒ and to remove the background noise, thereby achieving full GPU-based signal processing without the need for extra resampling hardware. A display rate of 25 frames/ sec is achieved for processed images ͑1024ϫ 1024 pixels͒ using a line-scan charge-coupled device ͑CCD͒ camera operating at 25.6 kHz. Optical coherence tomography ͑OCT͒ is a noncontact, noninvasive interferometric technique allowing micrometer resolution imaging of tissues. OCT can be classified into two categories: time domain ͑TD͒ OCT and spectral domain ͑SD͒ OCT. SD-OCT eliminates the need for mechanical scanning in TD-OCT to produce reflectivity profiles ͑A-scans͒ by recording the interference signal as a function of wavelength. This can be done either by using a spectrometer when performing Fourier domain ͑FD͒-OCT, to spatially encode the different optical frequencies, or by encoding them in time using a frequency scanning source, when performing swept source OCT. Both SD-OCT schemes have advantages of a better signal-to-noise ratio than that of the TD-OCT and a significantly improved imaging speed.1 The increased imaging speed requires very fast digital signal processing that is able to keep up with over tens of megahertz video rate data acquisitions. OCT digital signal processing mechanisms based on current central processing units ͑CPUs͒ are not capable of coping with such data processing speeds, and limit the sample rate dramatically.The rapid development of graphics processing units ͑GPUs͒ has introduced a revolution in numerical calculations. A GPU consists of many stream processors acting concurrently, thus favoring parallel programming. A recent report describes a GPU-based FD-OCT setup, where signal processing was performed partly by optical hardware and partly by the GPU 2 to enable real-time OCT imaging. We show here that similar data rates are possible by transferring all processing tasks to the GPU.FD-OCT employs a spectrometer to disperse the output of the interferometer onto a linear digital camera. A fast Fourier transform ͑FFT͒ of the linear camera signal provides the A-scan. However, the FFT operates in the optical frequency domain, while the equally spaced camera pixels produce an approximately linear representation of the spectrum in waveleng...

No abstract

A real-time microscopic profilometry system based on digital fringe projection and parallel programming has been developed and experimentally tested. Structured light patterns are projected onto an object through one pathway of a stereoscopic operation microscope. The patterns are deformed by the shape of the object and are then recorded with a high-speed CCD camera placed in the other pathway of the microscope. As the optical pathways of both arms are separated and reach the same object point at a relative angle, the recorded patterns allow the full-field object height variations to be calculated and the three-dimensional shape to be reconstructed by employing standard triangulation techniques. Applying proper hardware triggering, the projector-camera system is synchronized to capture up to 120 unique deformed line patterns per second. Using standard four-step phase-shifting profilometry techniques and applying graphics processing unit programming for fast phase wrapping, scaling, and visualization, we demonstrate the capability of the proposed system to generate 30 microscopic height maps per second. This allows the qualitative depth perception of the stereomicroscope operator to be enhanced by live quantitative height measurements with depth resolutions in the micrometer range.