RESUMO
Adversarial attacks inject imperceptible noise to images to deteriorate the performance of deep image classification models. However, most of the existing studies consider attacks in the digital (pixel) domain where an image acquired by an image sensor with sampling and quantization is recorded. This paper, for the first time, introduces a scheme for optical adversarial attack, which physically alters the light field information arriving at the image sensor so that the classification model yields misclassification. We modulate the phase of the light in the Fourier domain using a spatial light modulator placed in the photographic system. The operative parameters of the modulator for adversarial attack are obtained by gradient-based optimization to maximize cross-entropy and minimize distortion. Experiments based on both simulation and a real optical system demonstrate the feasibility of the proposed optical attack. We show that our attack can conceal perturbations in the image more effectively than the existing pixel-domain attack. It is also verified that the proposed attack is completely different from common optical aberrations such as spherical aberration, defocus, and astigmatism in terms of both perturbation patterns and classification results.
RESUMO
We present a novel, to the best of our knowledge, form of polarization microscopy capable of producing quantitative optic-axis and phase retardation maps of transparent and anisotropic materials. The proposed method operates on differential phase-contrast (DPC) microscopy that produces a phase image of a thin specimen using multi-axis intensity measurements. For polarization-sensitive imaging, patterned illumination light is circularly polarized to illuminate a specimen. The light transmitted through a specimen is split into two orthogonal polarization states and measured by an image sensor. Subsequent DPC computation based on the illumination patterns, acquired images, and the imaging model enables the retrieval of polarization-dependent quantitative phase images, which are utilized to reconstruct the orientation and retardation of the specimen. We demonstrate the validity of the proposed method by measuring the optic-axis and phase retardation maps of calibrated and various anisotropic samples.
RESUMO
Optical anisotropy, which is an intrinsic property of many materials, originates from the structural arrangement of molecular structures, and to date, various polarization-sensitive imaging (PSI) methods have been developed to investigate the nature of anisotropic materials. In particular, the recently developed tomographic PSI technologies enable the investigation of anisotropic materials through volumetric mappings of the anisotropy distribution of these materials. However, these reported methods mostly operate on a single scattering model, and are thus not suitable for three-dimensional (3D) PSI imaging of multiple scattering samples. Here, we present a novel reference-free 3D polarization-sensitive computational imaging technique-polarization-sensitive intensity diffraction tomography (PS-IDT)-that enables the reconstruction of 3D anisotropy distribution of both weakly and multiple scattering specimens from multiple intensity-only measurements. A 3D anisotropic object is illuminated by circularly polarized plane waves at various illumination angles to encode the isotropic and anisotropic structural information into 2D intensity information. These information are then recorded separately through two orthogonal analyzer states, and a 3D Jones matrix is iteratively reconstructed based on the vectorial multi-slice beam propagation model and gradient descent method. We demonstrate the 3D anisotropy imaging capabilities of PS-IDT by presenting 3D anisotropy maps of various samples, including potato starch granules and tardigrade.
RESUMO
Birefringence, an inherent characteristic of optically anisotropic materials, is widely utilized in various imaging applications ranging from material characterizations to clinical diagnosis. Polarized light microscopy enables high-resolution, high-contrast imaging of optically anisotropic specimens, but it is associated with mechanical rotations of polarizer/analyzer and relatively complex optical designs. Here, we present a form of lens-less polarization-sensitive microscopy capable of complex and birefringence imaging of transparent objects without an optical lens and any moving parts. Our method exploits an optical mask-modulated polarization image sensor and single-input-state LED illumination design to obtain complex and birefringence images of the object via ptychographic phase retrieval. Using a camera with a pixel size of 3.45 µm, the method achieves birefringence imaging with a half-pitch resolution of 2.46 µm over a 59.74 mm2 field-of-view, which corresponds to a space-bandwidth product of 9.9 megapixels. We demonstrate the high-resolution, large-area, phase and birefringence imaging capability of our method by presenting the phase and birefringence images of various anisotropic objects, including a monosodium urate crystal, and excised mouse eye and heart tissues.
RESUMO
Paper-based lateral-flow assays (LFAs) have achieved considerable commercial success and continue to have a significant impact on medical diagnostics and environmental monitoring. Conventional LFAs are typically performed by examining the color changes in the test bands by naked eye. However, for critical biochemical markers that are present in extremely small amounts in the clinical specimens, this readout method is not quantitative, and does not provide sufficient sensitivity or suitable detection limit for a reliable assay. Diverse technologies for high-sensitivity LFA detection have been developed and commercialization efforts are underway. In this review, we aim to provide a critical and objective overview of the recent progress in high-sensitivity LFA detection technologies, which involve the exploitation of the physical and chemical responses of transducing particles. The features and biomedical applications of the technologies, along with future prospects and challenges, are also discussed.
Assuntos
Técnicas Biossensoriais/instrumentação , Dispositivos Lab-On-A-Chip , Papel , Sistemas Automatizados de Assistência Junto ao Leito , Animais , Técnicas Biossensoriais/métodos , Desenho de Equipamento , HumanosRESUMO
Assessing the surgical margin during breast lumpectomy operations can avoid the need for additional surgery. Optical coherence tomography (OCT) is an imaging technique that has been proven to be efficient for this purpose. However, to avoid overloading the surgeon during the operation, automatic cancer detection at the surface of the removed tissue is needed. This work explores automated margin assessment on a sample of patient data collected at the Pathology Department, Severance Hospital (Seoul, South Korea). Some methods based on the spatial statistics of the images have been developed, but the obtained results are still far from human performance. In this work, we investigate the possibility to use deep neural networks (DNNs) for real time margin assessment, demonstrating performance significantly better than the reported literature and close to the level of a human expert. Since the goal is to detect the presence of cancer, a patch-based classification method is proposed, as it is sufficient for detection, and requires training data that is easier and cheaper to collect than for other approaches such as segmentation. For that purpose, we train a DNN architecture that was proved to be efficient for small images on patches extracted from images containing only cancer or only normal tissue as determined by pathologists in a university hospital. As the number of available images in all such studies is by necessity small relative to other deep network applications such as ImageNet, a good regularization method is needed. In this work, we propose to use a recently introduced function norm regularization that attempts to directly control the function complexity, in contrast to classical approaches such as weight decay and DropOut. As neither the code nor the data of previous results are publicly available, the obtained results are compared with reported results in the literature for a conservative comparison. Moreover, our method is applied to locally collected data on several data configurations. The reported results are the average over the different trials. The experimental results show that the use of DNNs yields significantly better results than other techniques when evaluated in terms of sensitivity, specificity, F1 score, G-mean and Matthews correlation coefficient. Function norm regularization yielded higher and more robust results than competing regularization methods. We have demonstrated a system that shows high promise for (partially) automated margin assessment of human breast tissue, Equal error rate (EER) is reduced from approximately 12% (the lowest reported in the literature) to 5%â¯-â¯a 58% reduction. The method is computationally feasible for intraoperative application (less than 2â¯s per image) at the only cost of a longer offline training time.
Assuntos
Neoplasias da Mama/cirurgia , Margens de Excisão , Intensificação de Imagem Radiográfica/métodos , Algoritmos , Feminino , Humanos , Rede Nervosa , Tomografia Computadorizada por Raios XRESUMO
Paper-based lateral-flow assay (LFA) is a simple and inexpensive point-of-care device that has become commonplace in medicine, environmental monitoring, and over-the-counter personal use. Some LFAs have demonstrated comparable analytical performance with laboratory-based methods, but the detection limit or sensitivity of most LFAs is significantly inferior to other molecular techniques by 10-100â¯×â¯. Consequently, LFAs are not viable for the early detection of disease-relevant biomarkers that are present in extremely small amounts in clinical specimens. Herein, we present a simple, cost-effective, and highly sensitive LFA sensor based on photothermal laser speckle imaging (PT-LSI). Under the illumination of a photothermal excitation light, gold nanoparticles (AuNPs), a common signal transduction medium in LFAs, absorb the light energy to produce heat, which subsequently induces modulation of the optical property and thermal deformation of the membrane. We measured these fluctuations through laser speckle imaging to quantify the concentration of AuNP-biomarker complexes. We experimentally demonstrate that the detection limit of our technique is superior to that of colorimetric detector by 68-125â¯×â¯. The capability of our sensor for highly sensitive detection of disease biomarkers is validated by using U.S. FDA-approved LFA kits for cryptococcal antigens (CrAg).
Assuntos
Técnicas Biossensoriais/instrumentação , Imunoensaio/instrumentação , Nanopartículas Metálicas/química , Antígenos de Bactérias/metabolismo , Cryptococcus/metabolismo , Ouro/química , Lasers , Limite de Detecção , Imagem ÓpticaRESUMO
BACKGROUND: The stratum corneum is an almost impermeable barrier. Recently, microneedles have been used to increase drug delivery passing the stratum corneum by incorporating the drug within the microneedle or by coating the surface of the microneedle with the drug. OBJECTIVE: This study was performed to investigate whether applying a biodegradable microneedle patch after topical steroid application increases penetration of the steroid in vitro, as well as treatment efficacy in patients with prurigo nodularis. MATERIALS & METHODS: In vitro penetration of topical steroids after biodegradable microneedle patch application was measured using a 3D skin model. To evaluate the treatment efficacy of the combination of biodegradable microneedle and topical steroids, a split-body clinical study was performed. RESULTS: Penetration of topical steroid in the in vitro skin model was significantly greater in the microneedle-applied skin. In a split-body clinical study with prurigo nodularis patients, the area and height of skin lesions decreased after four weeks of treatment on both sides, however, the microneedle patch side exhibited a significantly greater decrease in both area and height, compared to the control side. The pruritus visual analogue scale was also significantly lower on the microneedle side. CONCLUSION: We suggest that simply applying a microneedle patch after topical steroid application could be a useful strategy for treating refractory skin diseases such as prurigo nodularis.
Assuntos
Anti-Inflamatórios/administração & dosagem , Valerato de Betametasona/administração & dosagem , Ácido Hialurônico/administração & dosagem , Prurigo/tratamento farmacológico , Implantes Absorvíveis , Adulto , Anti-Inflamatórios/farmacocinética , Valerato de Betametasona/farmacocinética , Implantes de Medicamento , Feminino , Humanos , Ácido Hialurônico/farmacocinética , Masculino , Agulhas , Prurigo/metabolismo , Prurido/tratamento farmacológico , Tomografia de Coerência ÓpticaRESUMO
Gross total resection (GTR) of glioma is critical for improving the survival rate of glioma patients. One of the greatest challenges for achieving GTR is the difficulty in discriminating low grade tumor or peritumor regions that have an intact blood brain barrier (BBB) from normal brain tissues and delineating glioma margins during surgery. Here we present a highly sensitive, label-free terahertz reflectometry imaging (TRI) that overcomes current key limitations for intraoperative detection of World Health Organization (WHO) grade II (low grade), and grade III and IV (high grade) gliomas. We demonstrate that TRI provides tumor discrimination and delineation of tumor margins in brain tissues with high sensitivity on the basis of Hematoxylin and eosin (H&E) stained image. TRI may help neurosurgeons to remove gliomas completely by providing visualization of tumor margins in WHO grade II, III, and IV gliomas without contrast agents, and hence, improve patient outcomes.