TY - INPR A1 - Ahmadi, Samim A1 - Kästner, L. A1 - Hauffen, Jan Christian A1 - Jung, P. A1 - Ziegler, Mathias T1 - Photothermal-SR-Net: A customized deep unfolding neural network for photothermal super resolution imaging N2 - This paper presents deep unfolding neural networks to handle inverse problems in photothermal radiometry enabling super resolution (SR) imaging. Photothermal imaging is a well-known technique in active thermography for nondestructive inspection of defects in materials such as metals or composites. A grand challenge of active thermography is to overcome the spatial resolution limitation imposed by heat diffusion in order to accurately resolve each defect. The photothermal SR approach enables to extract high-frequency spatial components based on the deconvolution with the thermal point spread function. However, stable deconvolution can only be achieved by using the sparse structure of defect patterns, which often requires tedious, hand-crafted tuning of hyperparameters and results in computationally intensive algorithms. On this account, Photothermal-SR-Net is proposed in this paper, which performs deconvolution by deep unfolding considering the underlying physics. This enables to super resolve 2D thermal images for nondestructive testing with a substantially improved convergence rate. Since defects appear sparsely in materials, Photothermal-SR-Net applies trained blocksparsity thresholding to the acquired thermal images in each convolutional layer. The performance of the proposed approach is evaluated and discussed using various deep unfolding and thresholding approaches applied to 2D thermal images. Subsequently, studies are conducted on how to increase the reconstruction quality and the computational performance of Photothermal-SR-Net is evaluated. Thereby, it was found that the computing time for creating high-resolution images could be significantly reduced without decreasing the reconstruction quality by using pixel binning as a preprocessing step. KW - Photothermal super resolution KW - Nondestructive testing KW - Deep unfolding KW - Deep learning KW - Deep imaging KW - Physics-based deep learning KW - Laser thermography KW - Elastic net KW - Iterative shrinkage thresholding algorithm PY - 2021 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-525371 DO - https://doi.org/10.48550/arXiv.2104.10563 SN - 2331-8422 SP - 1 EP - 10 PB - Cornell University CY - Ithaca, NY AN - OPUS4-52537 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Ahmadi, Samim A1 - Kästner, L. A1 - Hauffen, Jan Christian A1 - Jung, P. A1 - Ziegler, Mathias T1 - Photothermal-SR-Net: A customized deep unfolding neural network for photothermal super resolution imaging N2 - This article presents deep unfolding neural networks to handle inverse problems in photothermal radiometry enabling super-resolution (SR) imaging. The photothermal SR approach is a well-known technique to overcome the spatial resolution limitation in photothermal imaging by extracting high-frequency spatial components based on the deconvolution with the thermal point spread function (PSF). However, stable deconvolution can only be achieved by using the sparse structure of defect patterns, which often requires tedious, handcrafted tuning of hyperparameters and results in computationally intensive algorithms. On this account, this article proposes Photothermal-SR-Net, which performs deconvolution by deep unfolding considering the underlying physics. Since defects appear sparsely in materials, our approach includes trained block-sparsity thresholding in each convolutional layer. This enables to super-resolve 2-D thermal images for nondestructive testing (NDT) with a substantially improved convergence rate compared to classic approaches. The performance of the proposed approach is evaluated on various deep unfolding and thresholding approaches. Furthermore, we explored how to increase the reconstruction quality and the computational performance. Thereby, it was found that the computing time for creating high-resolution images could be significantly reduced without decreasing the reconstruction quality by using pixel binning as a preprocessing step. KW - Deep unfolding KW - Defect reconstruction KW - Elastic net KW - Inverse problems KW - Iterative shrinkage thresholding KW - Neural network KW - Nondestructive testing (NDT) KW - Photothermal imaging KW - Super resolution (SR) KW - Thermography PY - 2022 DO - https://doi.org/10.1109/tim.2022.3154803 SN - 1557-9662 VL - 71 SP - 1 EP - 9 PB - IEEE AN - OPUS4-54678 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Hauffen, J. C. A1 - Kästner, L. A1 - Ahmadi, Samim A1 - Jung, P. A1 - Caire, G. A1 - Ziegler, Mathias T1 - Learned block iterative shrinkage thresholding algorithm for photothermal super resolution imaging N2 - Block-sparse regularization is already well known in active thermal imaging and is used for multiple-measurement-based inverse problems. The main bottleneck of this method is the choice of regularization parameters which differs for each experiment. We show the benefits of using a learned block iterative shrinkage thresholding algorithm (LBISTA) that is able to learn the choice of regularization parameters, without the need to manually select them. In addition, LBISTA enables the determination of a suitable weight matrix to solve the underlying inverse problem. Therefore, in this paper we present LBISTA and compare it with state-of-the-art block iterative shrinkage thresholding using synthetically generated and experimental test data from active thermography for defect reconstruction. Our results show that the use of the learned block-sparse optimization approach provides smaller normalized mean square errors for a small fixed number of iterations. Thus, this allows us to improve the convergence speed and only needs a few iterations to generate accurate defect reconstruction in photothermal super-resolution imaging. KW - Thermography KW - Laser KW - Machine learning KW - Optimization KW - Non-destructive testing KW - NDT KW - Neural network KW - Defect reconstruction KW - Block-sparsity KW - Active thermal imaging KW - Regularization KW - Laser thermography PY - 2022 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-554598 DO - https://doi.org/10.3390/s22155533 SN - 1424-8220 VL - 22 IS - 15 SP - 1 EP - 15 PB - MDPI CY - Basel AN - OPUS4-55459 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Hiebl, B. A1 - Ascher, Lena A1 - Luetzow, K. A1 - Kratz, K. A1 - Gruber, C. A1 - Mrowietz, C. A1 - Nehring, M. E. A1 - Lendlein, A. A1 - Franke, R.- P. A1 - Jung, F. T1 - Albumin solder covalently bound to a polymer membrane: New approach to improve binding strength in laser tissue soldering in-vitro N2 - Laser tissue soldering (LTS) based on indocyanine green (ICG)-mediated heat- denaturation of proteins might be a promising alternative technique for micro-suturing, but up to now the Problem of too weak shear strength of the solder welds in comparison to sutures is not solved. Earlier reports gave promising results showing that solder supported by carrier materials can enhance the cohesive strength of the liquid solder. In these studies, the solder was applied to the carriers by dip coating. Higher reliability of the connection between the solder and the carrier material is expected when the solder is bound covalently to the carrier material. In the present study a poly(ether imide) (PEI) membrane served as carrier material and ICG- supplemented albumin as solder substrate. The latter was covalently coupled to the carrier membrane under physiological conditions to prevent structural protein changes. As laser source a diode continuous-wave laser emitting at 808 nm with intensities between 250mW and 1500mW was utilized. The Albumin functionalized carrier membrane was placed onto the tunica media of explanted pig thoracic aortae forming an overlapping area of approximately 0.5×0.5 cm2. All tests were performed in a dry state to prevent laser light absorption by water. Infrared spectroscopy, spectro-photometrical determination of the secondary and Primary amine groups after acid orange II staining, contact angle measurements, and atomic force microscopy proved the successful functionalization of the PEI membrane with albumin. A laser power of 450mW LTS could generate a membrane-blood vessel connection which was characterized by a shear strength of 0.08±0.002MPa, corresponding to 15% of the tensile strength of the native blood vessel. Theoretically, an overlapping zone of 4.1mmaround the entire circumference of the blood vessel could have provided shear strength of the PEI membrane-blood vessel compound identical to the tensile strength of the native blood vessel. These in-vitro results confirmed the beneficial effects of solder reinforcement by carrier membranes, and suggest LTS with covalently bound solders on PEI substrates for further studies in animal models. KW - Tissue soldering KW - Albumin KW - Polymer membrane KW - Laser PY - 2018 DO - https://doi.org/10.3233/CH-189108 SN - 1386-0291 SN - 1875-8622 VL - 69 IS - 1-2 SP - 317 EP - 326 PB - IOS Press AN - OPUS4-45149 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Augustin, Sven A1 - Hieronymus, J. A1 - Jung, P. A1 - Hübers, H.-W. T1 - Compressed sensing in a fully non-mechanical 350 GHz imaging setting N2 - We investigate a single-pixel camera (SPC) that relies on non-mechanical scanning with a terahertz (THz) spatial light modulator (SLM) and Compressed Sensing (CS) for image generation. The camera is based on a 350 GHz multiplier source and a Golay cell detector. The SLM consists of a Germanium disc, which is illuminated by a halogen lamp. The light of the lamp is transmitted through a thin-film transistor (TFT) liquid crystal display (LCD). This enables the generation of light patterns on the Germanium disc, which in turn produce reflecting patterns for THz radiation. Using up to 1000 different patterns the pseudo-inverse reconstruction algorithm and the CS algorithm CoSaMP are evaluated with respect to image quality. It is shown that CS allows a reduction of the necessary measurements by a factor of three without compromising the image quality. KW - Compressed Sensing (CS) KW - Terahertz KW - Imaging KW - Single-pixel camera KW - Non-mechanical scanning KW - TFT-LCD KW - THz KW - Spatial light modulator PY - 2015 DO - https://doi.org/10.1007/s10762-014-0141-5 SN - 1866-6892 SN - 1866-6906 VL - 36 IS - 5 SP - 496 EP - 512 PB - Springer CY - New York, NY AN - OPUS4-33022 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Walther, M. A1 - Jung, C. M. A1 - Bergmann, R. A1 - Pietzsch, J. A1 - Rode, K. A1 - Fahmy, K. A1 - Mirtschink, P. A1 - Stehr, S. A1 - Heintz, A. A1 - Wunderlich, G. A1 - Kraus, Werner A1 - Pietzsch, H.-J. A1 - Kropp, J. A1 - Deussen, A. A1 - Spies, H. T1 - Synthesis and Biological Evaluation of a New Type of 99mTechnetium-Labeled Fatty Acid for Myocardial Metabolism Imaging PY - 2007 SN - 1043-1802 SN - 1520-4812 VL - 18 SP - 216 EP - 230 CY - Washington, DC AN - OPUS4-14514 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Park, J. A1 - Kim, G. B. A1 - Lippitz, Andreas A1 - Kim, Y. M. A1 - Jung, D. A1 - Unger, Wolfgang A1 - Kim, Y.-P. A1 - Lee, T. G. T1 - Plasma-polymerized antifouling biochips for label-free measurement of protease activity in cell culture media N2 - We report polyethylene glycol (PEG)-grafting antifouling surfaces using a plasma copolymerized (PcP) technique to monitor protease activity in complex media. By varying the mixing ratio of the PEG and ethylenediamine (EDA) precursors, the PcP-PEG-EDA (PcP-PE) film was able to easily control surface amine density with good preservation of the internal PEG structure. We found that nonspecific protein adsorption was dramatically reduced in serum-containing media on the PcP-PE films, as opposed to that on plasma polymerized-EDA (PP-E) films without PEG. When SPR sensor chips coated with PcP-PE film were employed to detect protease activity, biotinylated luciferase probes (luciferase-peptide-biotin) on streptavidin-conjugated SPR chips enabled real-time and label-free measurement of matrix metalloproteinase activity in cell culture media. Owing to its excellent antifouling ability, this newly developed method boasts minimal nonspecific binding and can serve as a biochip platform to promote a wide range of applications in the biological field. KW - Polyethylene glycol (PEG)-grafting antifouling surface KW - Biochip KW - ATR-FTIR spectroscopy KW - XPS KW - NEXAFS KW - Plasma PY - 2019 DO - https://doi.org/10.1016/j.snb.2018.10.123 SN - 0925-4005 VL - 281 SP - 527 EP - 534 PB - Elsevier B.V. AN - OPUS4-46463 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -