Biblio

In this paper, a closely packed Biosensor construction by using a two-dimensional structure is described. This structure uses air-holes slab constructed on silicon material. By removing certain air holes in the slab, waveguides are constructed. By carrying out simulation, it is proved that the harmonic guided wave changes to lengthier wavelengths with reagents, pesticides, proteins & DNA capturing. A Biosensor is constructed with an improved Quality factor & wavelength. This gives high Quality Factor (HQF) resolution Biosensor. The approach used for Simulation purpose is Finite Difference Time Domain(FDTD).

Advances in deep neural networks (DNNs) have shown tremendous promise in the medical domain. However, the deep learning tools that are helping the domain, can also be used against it. Given the prevalence of fraud in the healthcare domain, it is important to consider the adversarial use of DNNs in manipulating sensitive data that is crucial to patient healthcare. In this work, we present the design and implementation of a DNN-based image translation attack on biomedical imagery. More specifically, we propose Jekyll, a neural style transfer framework that takes as input a biomedical image of a patient and translates it to a new image that indicates an attacker-chosen disease condition. The potential for fraudulent claims based on such generated `fake' medical images is significant, and we demonstrate successful attacks on both X-rays and retinal fundus image modalities. We show that these attacks manage to mislead both medical professionals and algorithmic detection schemes. Lastly, we also investigate defensive measures based on machine learning to detect images generated by Jekyll.

Summary form only given. Authentication of people or objects using physical keys is insecure against secret duplication. Physical unclonable functions (PUF) are special physical keys that are assumed to be unclonable due to the large number of degrees of freedom in their manufacturing [1]. Opaque scattering media, such as white paint and teeth, comprise of millions of nanoparticles in a random arrangement. Under coherent light illumination, the multiple scattering from these nanoparticles gives rise to a complex interference resulting in a speckle pattern. The speckle pattern is seemingly random but highly sensitive to the exact position and size of the nanoparticles in the given piece of opaque scattering medium [2], thereby realizing an ideal optical PUF. These optical PUFs enabled applications such as quantum-secure authentication (QSA) and communication [3, 4].

Fourier domain mode locked (FDML) lasers, in which the sweep period of the swept bandpass filter is synchronized with the roundtrip time of the optical field, are broadband and rapidly tunable fiber ring laser systems, which offer rich dynamics. A detailed understanding is important from a fundamental point of view, and also required in order to improve current FDML lasers which have not reached their coherence limit yet. Here, we study the formation of localized patterns in the intensity trace of FDML laser systems based on a master equation approach [1] derived from the nonlinear Schrödinger equation for polarization maintaining setups, which shows excellent agreement with experimental data. A variety of localized patterns and chaotic or bistable operation modes were previously discovered in [2–4] by investigating primarily quasi-static regimes within a narrow sweep bandwidth where a delay differential equation model was used. In particular, the formation of so-called holes which are characterized by a dip in the intensity trace and a rapid phase jump are described. Such holes have tentatively been associated with Nozaki-Bekki holes which are solutions to the complex Ginzburg-Landau equation. In Fig. 1 (b) to (d) small sections of a numerical solution of our master equation are presented for a partially dispersion compensated polarization maintaining FDML laser setup. Within our approach, we are able to study the full sweep dynamics over a broad sweep range of more than 100 nm. This allows us to identify different co-existing intensity patterns within a single sweep. In general, high frequency distortions in the intensity trace of FDML lasers [5] are mainly caused by synchronization mismatches caused by the fiber dispersion or a detuning of the roundtrip time of the optical field to the sweep period of the swept bandpass filter. This timing errors lead to rich and complex dynamics over many roundtrips and are a major source of noise, greatly affecting imaging and sensing applications. For example, the imaging quality in optical coherence tomography where FDML lasers are superior sources is significantly reduced [5].

Wireless Capsule Endoscopy (WCE) is a noninvasive device for detection of gastrointestinal problems especially small bowel diseases, such as polyps which causes gastrointestinal bleeding. The quality of WCE images is very important for diagnosis. In this paper, a new method is proposed to improve the quality of WCE images. In our proposed method for improving the quality of WCE images, Removing Noise and Contrast Enhancement (RNCE) algorithm is used. The algorithm have been implemented and tested on some real images. Quality metrics used for performance evaluation of the proposed method is Structural Similarity Index Measure (SSIM), Peak Signal-to-Noise Ratio (PSNR) and Edge Strength Similarity for Image (ESSIM). The results obtained from SSIM, PSNR and ESSIM indicate that the implemented RNCE method improve the quality of WCE images significantly.

Optical Coherence Tomography (OCT) has shown a great potential as a complementary imaging tool in the diagnosis of skin diseases. Speckle noise is the most prominent artifact present in OCT images and could limit the interpretation and detection capabilities. In this work we evaluate various denoising filters with high edge-preserving potential for the reduction of speckle noise in 256 dermatological OCT B-scans. Our results show that the Enhanced Sigma Filter and the Block Matching 3-D (BM3D) as 2D denoising filters and the Wavelet Multiframe algorithm considering adjacent B-scans achieved the best results in terms of the enhancement quality metrics used. Our results suggest that a combination of 2D filtering followed by a wavelet based compounding algorithm may significantly reduce speckle, increasing signal-to-noise and contrast-to-noise ratios, without the need of extra acquisitions of the same frame.