An end-to-end tumor diagnosis framework including resolution enhancement and tumor classification is proposed. The U-Net + EDSR network enables a significant improvement of PSNR and enhances the resolution beyond physical limitations. Moreover, the subsequent tumor discrimination can benefit from the enhancement. Multi-image as network input and advanced models like generative adversarial networks are expected to bring a further improvement for the imaging. Our proposed novel method first time realizes intraoperative lensless CFB imaging with high resolution in the near-field. The technique builds a bridge to many techniques like optical biopsies, multi-modal imaging, virtual staining, and computer-assisted disease diagnostics for neuron signal monitoring as well as neurosurgery.
Fiber-based lensless endoscopy is powerful tool for minimally invasive tissue in clinical practice. However, the inherent honeycomb-artifact reduce the resolution and increases diagnosis difficulty. We proposed an end-to-end resolution enhancement and classification network for fiber bundle imaging. Comparing with conventional interpolation and filtering methods, the average peak signal to noise ratio (PSNR) can be improved 2~6 dB. Then we trained a VGG-19 classification network on label-free multiphoton images of 382 human braintumor 28 nontumor brain samples. The results show the classification accuracy of enhanced images is up to 91%, while the fiber bundle images are only 67% accurate. The method paves the way to in vivo histologic imaging through miniaturized endoscopic probes, and gives rapid and accurate determination for intraoperative diagnosis.
Endoscopic optical coherence tomography (eOCT), a depth-resolved, non-invasive, three-dimensional imaging technique, enables morphologic and functional examinations of the human tympanic membrane (TM) in vivo. To overcome the limitation of intensity-based OCT in tissue differentiation, light polarization is used here to assess collagen fibers as birefringent tissue. A fiber-based polarization-sensitive eOCT system was realized for first in vivo examinations of the TM’s collagen fiber layer, on which the acoustic and mechanical properties depend. By evaluating the local retardation, distinguishing the TM’s different tissues was possible and a comparison to microscopic ex vivo measurements was made.
Sandra Tamosaityte, Roberta Galli, Ortrud Uckermann, Kerim Sitoci-Ficici, Maria Koch, Robert Later, Gabriele Schackert, Edmund Koch, Gerald Steiner, Matthias Kirsch
Spinal cord injury (SCI) triggers several lipid alterations in nervous tissue. It is characterized by extensive demyelination and the inflammatory response leads to accumulation of activated microglia/macrophages, which often transform into foam cells by accumulation of lipid droplets after engulfment of the damaged myelin sheaths. Using an experimental rat model, Raman microspectroscopy was applied to retrieve the modifications of the lipid distribution following SCI. Coherent anti-Stokes Raman scattering (CARS) and endogenous two-photon fluorescence (TPEF) microscopies were used for the detection of lipid-laden inflammatory cells. The Raman mapping of CH2 deformation mode intensity at 1440 cm−1 retrieved the lipid-depleted injury core. Preserved white matter and inflammatory regions with myelin fragmentation and foam cells were localized by specifically addressing the distribution of esterified lipids, i.e., by mapping the intensity of the carbonyl Raman band at 1743 cm−1, and were in agreement with CARS/TPEF microscopy. Principal component analysis revealed that the inflammatory regions are notably rich in saturated fatty acids. Therefore, Raman spectroscopy enabled to specifically detect inflammation after SCI and myelin degradation products.
Roberta Galli, Sandra Tamosaityte, Maria Koch, Kerim Sitoci-Ficici, Robert Later, Ortrud Uckermann, Rudolf Beiermeister, Michael Gelinsky, Gabriele Schackert, Matthias Kirsch, Edmund Koch, Gerald Steiner
The treatment of spinal cord injury by using implants that provide a permissive environment for axonal growth is in the focus of the research for regenerative therapies. Here, Raman-based label-free techniques were applied for the characterization of morphochemical properties of surgically induced spinal cord injury in the rat that received an implant of soft unfunctionalized alginate hydrogel. Raman microspectroscopy followed by chemometrics allowed mapping the different degenerative areas, while multimodal multiphoton microscopy (e.g. the combination of coherent anti-Stokes Raman scattering (CARS), endogenous two-photon fluorescence and second harmonic generation on the same platform) enabled to address the morphochemistry of the tissue at cellular level. The regions of injury, characterized by demyelination and scarring, were retrieved and the distribution of key tissue components was evaluated by Raman mapping. The alginate hydrogel was detected in the lesion up to six months after implantation and had positive effects on the nervous tissue. For instance, multimodal multiphoton microscopy complemented the results of Raman mapping, providing the micromorphology of lipid-rich tissue structures by CARS and enabling to discern lipid-rich regions that contained myelinated axons from degenerative regions characterized by myelin fragmentation and presence of foam cells. These findings demonstrate that Raman-based imaging methods provide useful information for the evaluation of alginate implant effects and have therefore the potential to contribute to new strategies for monitoring degenerative and regenerative processes induced in SCI, thereby improving the effectiveness of therapies.
Coherent anti-Stokes Raman scattering (CARS) microscopy is an emerging multiphoton technique for the label-free histopathology of the central nervous system, by imaging the lipid content within the tissue. In order to apply the technique on standard histology sections, it is important to know the effects of tissue fixation on the CARS image. Here, we report the effects of two common fixation methods, namely with formalin and methanol–acetone, on mouse brain and human glioblastoma tissue. The variations induced by fixation on the CARS contrast and intensity were compared and interpreted using Raman microspectroscopy. The results show that, whenever unfixed cryosections cannot be used, fixation with formalin constitutes an alternative which does not deteriorate substantially the contrast generated by the different brain structures in the CARS image. Fixation with methanol–acetone strongly modifies the tissue lipid content and is therefore incompatible with the CARS imaging.
Nonlinear optical microscopy offers a series of techniques that have the potential to be applied in vivo, for
intraoperative identification of tumor border and in situ pathology. By addressing the different content of lipids
that characterize the tumors with respect to the normal brain tissue, CARS microscopy enables to discern
primary and secondary brain tumors from healthy tissue. A study performed in mouse models shows that the
reduction of the CARS signal is a reliable quantity to identify brain tumors, irrespective from the tumor type.
Moreover it enables to identify tumor borders and infiltrations at a cellular resolution. Integration of CARS with
autogenous TPEF and SHG adds morphological and compositional details about the tissue. Examples of
multimodal CARS imaging of different human tumor biopsies demonstrate the ability of the technique to
retrieve information useful for histopathological diagnosis.
The interest to use laser surface processing in microtechnology as a friendly method from the technologic and environmental point of view lead our studies about laser radiation interaction with photo-resist and metallic thin films. In this view we have tried in our experiments to process metallic thin films deposited on silicon substrate by using laser radiation. To obtain a good quality of the metallic thin film removal from the silicon surface a careful selection of the incident laser intensity, number of pulses and irradiation geometry is needed. The threshold value for the laser cleaning intensity depends on the number of incident laser pulses. A careful experimental estimation of the cleaning conditions from the point of view of incident laser energy, fluence, intensity and irradiation geometry was realized for aluminum, copper, and chromium thin films.
The applicability of fiber optic sensors to the automotive field was evaluated within a national project, in which the Fiat Research Center, RTM and CNR-IROE are involved. The measurement of the mechanical deformations on a vehicle windshield was considered as a case study. This paper refers mainly to the validation phase of the project and in particular to the deformation measurements on a windshield of a FIAT 'Brava' using fiber Bragg gratings. The results were compared with data collected by using strain gauges and thermocouples. A brief description of the manufacturing and characterization of the grating and of the interrogation system is also provided.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.