Search

In contrast to non-medical image denoising, where enhancing image clarity is the primary goal, medical image denoising warrants preservation of crucial features without introduction of new artifacts. However, many denoising methods that improve the clarity of the image, inadvertently alter critical information of the denoised images, potentially compromising classification performance and diagnostic quality. Additionally, supervised denoising methods are not very practical in medical image domain, since a \emph{ground truth} denoised version of a noisy medical image is often extremely challenging to obtain. In this paper, we tackle both of these problems by introducing a novel neural network based method -- \emph{Contextual Checkerboard Denoising}, that can learn denoising from only a dataset of noisy images, while preserving crucial anatomical details necessary for image classification/analysis. We perform our experimentation on real Optical Coherence Tomography (OCT) images, and empirically demonstrate that our proposed method significantly improves image quality, providing clearer and more detailed OCT images, while enhancing diagnostic accuracy., Comment: Under review in Springer Journal of Medical Systems. Code available: https://github.com/AbtahiMajeed/CheckerBoardDenoiser/tree/main

Noninvasive optical imaging modalities can probe patient's tissue in 3D and over time generate gigabytes of clinically relevant data per sample. There is a need for AI models to analyze this data and assist clinical workflow. The lack of expert labelers and the large dataset required (>100,000 images) for model training and tuning are the main hurdles in creating foundation models. In this paper we introduce FoundationShift, a method to apply any AI model from computational pathology without retraining. We show our method is more accurate than state of the art models (SAM, MedSAM, SAM-Med2D, CellProfiler, Hover-Net, PLIP, UNI and ChatGPT), with multiple imaging modalities (OCT and RCM). This is achieved without the need for model retraining or fine-tuning. Applying our method to noninvasive in vivo images could enable physicians to readily incorporate optical imaging modalities into their clinical practice, providing real time tissue analysis and improving patient care.

In this paper, we propose a new approach to train deep learning models using game theory concepts including Generative Adversarial Networks (GANs) and Adversarial Training (AT) where we deploy a double-oracle framework using best response oracles. GAN is essentially a two-player zero-sum game between the generator and the discriminator. The same concept can be applied to AT with attacker and classifier as players. Training these models is challenging as a pure Nash equilibrium may not exist and even finding the mixed Nash equilibrium is difficult as training algorithms for both GAN and AT have a large-scale strategy space. Extending our preliminary model DO-GAN, we propose the methods to apply the double oracle framework concept to Adversarial Neural Architecture Search (NAS for GAN) and Adversarial Training (NAS for AT) algorithms. We first generalize the players' strategies as the trained models of generator and discriminator from the best response oracles. We then compute the meta-strategies using a linear program. For scalability of the framework where multiple network models of best responses are stored in the memory, we prune the weakly-dominated players' strategies to keep the oracles from becoming intractable. Finally, we conduct experiments on MNIST, CIFAR-10 and TinyImageNet for DONAS-GAN. We also evaluate the robustness under FGSM and PGD attacks on CIFAR-10, SVHN and TinyImageNet for DONAS-AT. We show that all our variants have significant improvements in both subjective qualitative evaluation and quantitative metrics, compared with their respective base architectures.

Continual learning (CL) adapt the deep learning scenarios with timely updated datasets. However, existing CL models suffer from the catastrophic forgetting issue, where new knowledge replaces past learning. In this paper, we propose Continual Learning with Task Specialists (CLTS) to address the issues of catastrophic forgetting and limited labelled data in real-world datasets by performing class incremental learning of the incoming stream of data. The model consists of Task Specialists (T S) and Task Predictor (T P ) with pre-trained Stable Diffusion (SD) module. Here, we introduce a new specialist to handle a new task sequence and each T S has three blocks; i) a variational autoencoder (V AE) to learn the task distribution in a low dimensional latent space, ii) a K-Means block to perform data clustering and iii) Bootstrapping Language-Image Pre-training (BLIP ) model to generate a small batch of captions from the input data. These captions are fed as input to the pre-trained stable diffusion model (SD) for the generation of task samples. The proposed model does not store any task samples for replay, instead uses generated samples from SD to train the T P module. A comparison study with four SOTA models conducted on three real-world datasets shows that the proposed model outperforms all the selected baselines

A canonical wavy surface exposed to a Mach 2 flow is investigated through high-frequency Pressure Sensitive Paint (PSP), Kulite measurements, and shadowgraph imaging. The wavy surface features a compression and expansion region, two shock-boundary layer interactions, and two shock-separation regions. The unsteady characteristics of the wall pressure and shock angles are presented, demonstrating an increase in amplitude of the instabilities when traveling through the shock systems. Three-dimensional flow features, observed in PSP data, reveal a two-dimensional flow pattern. Higher-Order Dynamic Mode Decomposition and Spectral Proper Orthogonal Decomposition are implemented to dissect the different flow features, revealing several dominant low-frequency and medium-frequency phenomena. The separation region appears at frequencies with Strouhal numbers between 0.01 and 0.2, confirmed by the frequency content in the local pressure measurement using Kulites.

Molecular optimization is a key challenge in drug discovery and material science domain, involving the design of molecules with desired properties. Existing methods focus predominantly on single-property optimization, necessitating repetitive runs to target multiple properties, which is inefficient and computationally expensive. Moreover, these methods often lack transparency, making it difficult for researchers to understand and control the optimization process. To address these issues, we propose a novel framework, Explainable Multi-property Optimization of Molecules (XMOL), to optimize multiple molecular properties simultaneously while incorporating explainability. Our approach builds on state-of-the-art geometric diffusion models, extending them to multi-property optimization through the introduction of spectral normalization and enhanced molecular constraints for stabilized training. Additionally, we integrate interpretive and explainable techniques throughout the optimization process. We evaluated XMOL on the real-world molecular datasets i.e., QM9, demonstrating its effectiveness in both single property and multiple properties optimization while offering interpretable results, paving the way for more efficient and reliable molecular design.

Continual learning (CL) models are designed to learn new tasks arriving sequentially without re-training the network. However, real-world ML applications have very limited label information and these models suffer from catastrophic forgetting. To address these issues, we propose an unsupervised CL model with task experts called Unsupervised Task Expert Lifelong Learning (U-TELL) to continually learn the data arriving in a sequence addressing catastrophic forgetting. During training of U-TELL, we introduce a new expert on arrival of a new task. Our proposed architecture has task experts, a structured data generator and a task assigner. Each task expert is composed of 3 blocks; i) a variational autoencoder to capture the task distribution and perform data abstraction, ii) a k-means clustering module, and iii) a structure extractor to preserve latent task data signature. During testing, task assigner selects a suitable expert to perform clustering. U-TELL does not store or replay task samples, instead, we use generated structured samples to train the task assigner. We compared U-TELL with five SOTA unsupervised CL methods. U-TELL outperformed all baselines on seven benchmarks and one industry dataset for various CL scenarios with a training time over 6 times faster than the best performing baseline., Comment: Accepted by International Conference on Image Processing 2024 (ICIP2024)

In the realm of sequential decision-making tasks, the exploration capability of a reinforcement learning (RL) agent is paramount for achieving high rewards through interactions with the environment. To enhance this crucial ability, we propose SAQN, a novel approach wherein a self-evolving autoencoder (SA) is embedded with a Q-Network (QN). In SAQN, the self-evolving autoencoder architecture adapts and evolves as the agent explores the environment. This evolution enables the autoencoder to capture a diverse range of raw observations and represent them effectively in its latent space. By leveraging the disentangled states extracted from the encoder generated latent space, the QN is trained to determine optimal actions that improve rewards. During the evolution of the autoencoder architecture, a bias-variance regulatory strategy is employed to elicit the optimal response from the RL agent. This strategy involves two key components: (i) fostering the growth of nodes to retain previously acquired knowledge, ensuring a rich representation of the environment, and (ii) pruning the least contributing nodes to maintain a more manageable and tractable latent space. Extensive experimental evaluations conducted on three distinct benchmark environments and a real-world molecular environment demonstrate that the proposed SAQN significantly outperforms state-of-the-art counterparts. The results highlight the effectiveness of the self-evolving autoencoder and its collaboration with the Q-Network in tackling sequential decision-making tasks., Comment: 11 pages, 9 figures, 3 tables

We give a constructive approach for the study of integral representations of classical solutions to Poisson equations under some integrability conditions on data functions., Comment: 17

A planet's Lyman-{\alpha} (Ly{\alpha}) emission is sensitive to its thermospheric structure. Here, we report joint Hubble Space Telescope (HST) and Cassini cross-calibration observations of the Saturn Ly{\alpha} emission made two weeks before the Cassini grand finale. To investigate the long-term Saturn Ly{\alpha} airglow observed by different ultraviolet instruments, we cross-correlate their calibration, finding that while the official Cassini/UVIS sensitivity should be lowered by ~75%, the Voyager 1/UVS sensitivities should be enhanced by ~20% at the Ly{\alpha} channels. This comparison also allowed us to discover a permanent feature of the Saturn disk Ly{\alpha} brightness that appears at all longitudes as a brightness excess (Ly{\alpha} bulge) of ~30% (~12{\sigma}) extending over the latitude range ~5-35N compared to the regions at equator and ~60N. This feature is confirmed by three distinct instruments between 1980 & 2017 in the Saturn north hemisphere. To analyze the Ly{\alpha} observations, we use a radiation transfer (RT) model of resonant scattering of solar and interplanetary Ly{\alpha} photons, and a latitude-dependent photochemistry model of the upper atmosphere constrained by occultation and remote-sensing observations. For each latitude, we show that the Ly{\alpha} observations are sensitive to the temperature profile in the upper stratosphere and lower thermosphere, thus providing useful information in a region of the atmosphere that is difficult to probe by other means. In the Saturn Ly{\alpha} bulge region, at latitudes between ~5 to ~35{\deg}, the observed brightening and line broadening support seasonal effects, variation of the temperature vertical profile, and potential superthermal atoms that require confirmation., Comment: Published in Planetary Science Journal. Added reference to grants

Quantum metrology takes advantage of quantum correlations to enhance the sensitivity of sensors and measurement techniques beyond their fundamental classical limit given by the shot noise limit. The use of both temporal and spatial correlations present in quantum states of light can extend quantum-enhanced sensing to a parallel configuration that can simultaneously probe an array of sensors or independently measure multiple parameters. To this end, we use multi-spatial mode twin beams of light, which are characterized by independent quantum-correlated spatial subregions in addition to quantum temporal correlations, to probe a four-sensor quadrant plasmonic array. We show that it is possible to independently and simultaneously measure local changes in refractive index for all four sensors with a quantum enhancement in sensitivity in the range of $22\%$ to $24\%$ over the corresponding classical configuration. These results provide a first step towards highly parallel spatially resolved quantum-enhanced sensing techniques and pave the way toward more complex quantum sensing and quantum imaging platforms.

We study Dirichlet problems for fractional Laplace equations of the form $(-\Delta)^{\frac{\alpha}{2}} u = f(x,u)$ in $\mathbb{R}^{n}$ for $0<\alpha

We solve the existence problem for the minimal positive solutions $u\in L^{p}(\Omega, dx)$ to the Dirichlet problems for sublinear elliptic equations of the form \[ \begin{cases} Lu=\sigma u^q+\mu\qquad \quad \text{in} \quad \Omega, \\ \liminf\limits_{x \rightarrow y}u(x) = 0 \qquad y \in \partial_{\infty}\Omega, \end{cases} \] where $0