Search

Numerous deep learning-based solutions have been proposed for histopathological image analysis over the past years. While they usually demonstrate exceptionally high accuracy, one key question is whether their precision might be affected by low-level image properties not related to histopathology but caused by microscopy image handling and pre-processing. In this paper, we analyze a popular NCT-CRC-HE-100K colorectal cancer dataset used in numerous prior works and show that both this dataset and the obtained results may be affected by data-specific biases. The most prominent revealed dataset issues are inappropriate color normalization, severe JPEG artifacts inconsistent between different classes, and completely corrupted tissue samples resulting from incorrect image dynamic range handling. We show that even the simplest model using only 3 features per image (red, green and blue color intensities) can demonstrate over 50% accuracy on this 9-class dataset, while using color histogram not explicitly capturing cell morphology features yields over 82% accuracy. Moreover, we show that a basic EfficientNet-B0 ImageNet pretrained model can achieve over 97.7% accuracy on this dataset, outperforming all previously proposed solutions developed for this task, including dedicated foundation histopathological models and large cell morphology-aware neural networks. The NCT-CRC-HE dataset is publicly available and can be freely used to replicate the presented results. The codes and pre-trained models used in this paper are available at https://github.com/gmalivenko/NCT-CRC-HE-experiments

Histopathological images are widely used for the analysis of diseased (tumor) tissues and patient treatment selection. While the majority of microscopy image processing was previously done manually by pathologists, recent advances in computer vision allow for accurate recognition of lesion regions with deep learning-based solutions. Such models, however, usually require extensive annotated datasets for training, which is often not the case in the considered task, where the number of available patient data samples is very limited. To deal with this problem, we propose a novel DeepCMorph model pre-trained to learn cell morphology and identify a large number of different cancer types. The model consists of two modules: the first one performs cell nuclei segmentation and annotates each cell type, and is trained on a combination of 8 publicly available datasets to ensure its high generalizability and robustness. The second module combines the obtained segmentation map with the original microscopy image and is trained for the downstream task. We pre-trained this module on the Pan-Cancer TCGA dataset consisting of over 270K tissue patches extracted from 8736 diagnostic slides from 7175 patients. The proposed solution achieved a new state-of-the-art performance on the dataset under consideration, detecting 32 cancer types with over 82% accuracy and outperforming all previously proposed solutions by more than 4%. We demonstrate that the resulting pre-trained model can be easily fine-tuned on smaller microscopy datasets, yielding superior results compared to the current top solutions and models initialized with ImageNet weights. The codes and pre-trained models presented in this paper are available at: https://github.com/aiff22/DeepCMorph

We present ANYU, a new virtually augmented version of the NYU depth v2 dataset, designed for monocular depth estimation. In contrast to the well-known approach where full 3D scenes of a virtual world are utilized to generate artificial datasets, ANYU was created by incorporating RGB-D representations of virtual reality objects into the original NYU depth v2 images. We specifically did not match each generated virtual object with an appropriate texture and a suitable location within the real-world image. Instead, an assignment of texture, location, lighting, and other rendering parameters was randomized to maximize a diversity of the training data, and to show that it is randomness that can improve the generalizing ability of a dataset. By conducting extensive experiments with our virtually modified dataset and validating on the original NYU depth v2 and iBims-1 benchmarks, we show that ANYU improves the monocular depth estimation performance and generalization of deep neural networks with considerably different architectures, especially for the current state-of-the-art VPD model. To the best of our knowledge, this is the first work that augments a real-world dataset with randomly generated virtual 3D objects for monocular depth estimation. We make our ANYU dataset publicly available in two training configurations with 10% and 100% additional synthetically enriched RGB-D pairs of training images, respectively, for efficient training and empirical exploration of virtual augmentation at https://github.com/ABrain-One/ANYU

It is well-known that a Lorenz curve derived from the distribution function of a random variable can be viewed itself as a probability distribution function of a new random variable [3]. We prove the surprising result that a sequence of consecutive iterations of this map leads to a non-corner case convergence independent of the starting random variable. In the primal case, the limiting distribution has a power law, with coefficient equal to the golden section. In the reflected case, the limiting distribution is classical Pareto, with a conjugate coefficient. Possible directions for applications are discussed.

The MEG II experiment, based at the Paul Scherrer Institut in Switzerland, reports the result of a search for the decay $\mu^+\to e^+\gamma$ from data taken in the first physics run in 2021. No excess of events over the expected background is observed, yielding an upper limit on the branching ratio of B($\mu^+\to e^+\gamma$) < $7.5 \times 10^{-13}$ (90% C.L.). The combination of this result and the limit obtained by MEG gives B($\mu^+\to e^+\gamma$) < $3.1 \times 10^{-13}$ (90% C.L.), which is the most stringent limit to date. A ten-fold larger sample of data is being collected during the years 2022-2023, and data-taking will continue in the coming years., Comment: 10 pages, 6 figures. To be published in EPJC

The cylindrical drift chamber is the most innovative part of the MEG~II detector, the upgraded version of the MEG experiment. The MEG~II chamber differs from the MEG one because it is a single volume cylindrical structure, instead of a segmented one, chosen to improve its resolutions and efficiency in detecting low energy positrons from muon decays at rest. In this paper, we show the characteristics and performances of this fundamental part of the MEG~II apparatus and we discuss the impact of its higher resolution and efficiency on the sensitivity of the MEG~II experiment. Because of its innovative structure and high quality resolution and efficiency the MEG~II cylindrical drift chamber will be a cornerstone in the development of an ideal tracking detector for future positron-electron collider machines., Comment: 27 pages, 42 figures, published on EPJC 84(2024)5,473

The MEG II experiment, located at the Paul Scherrer Institut (PSI) in Switzerland, is the successor to the MEG experiment, which completed data taking in 2013. MEG II started fully operational data taking in 2021, with the goal of improving the sensitivity of the mu+ -> e+ gamma decay down to 6e-14 almost an order of magnitude better than the current limit. In this paper, we describe the operation and performance of the experiment and give a new estimate of its sensitivity versus data acquisition time., Comment: 42 pages, 55 figures. Submitted to EPJC

A measurement of the 21-cm global signal would be a revealing probe of the Dark Ages, the era of first star formation, and the Epoch of Reionization. It has remained elusive owing to bright galactic and extra-galactic foreground contaminants, coupled with instrumental noise, ionospheric effects, and beam chromaticity. The simultaneous detection of a consistent 21-cm dipole signal alongside the 21-cm global signal would provide confidence in a claimed detection. We use simulated data to investigate the possibility of using drift-scan dipole antenna experiments to achieve a detection of both monopole and dipole. We find that at least two antennae located at different latitudes are required to localise the dipole. In the absence of foregrounds, a total integration time of $\sim 10^4$ hours is required to detect the dipole. With contamination by simple foregrounds, we find that the integration time required increases to $\sim 10^5$ hours. We show that the extraction of the 21-cm dipole from more realistic foregrounds requires a more sophisticated foreground modelling approach. Finally, we motivate a global network of dipole antennae that could reasonably detect the dipole in $\sim 10^3$ hours of integration time., Comment: 12 pages, 15 figures

Fundamental studies of the interaction of chiral light with chiral matter are important for the development of techniques that allow handedness-selective optical detection of chiral organic molecules. One approach to achieve this goal is the creation of a Fabry-P\'erot cavity that supports eigenmodes with a desired electromagnetic handedness, which interacts differently with left and right molecular enantiomers. In this paper, we theoretically study chiral Fabry-P\'erot cavities with mirrors comprising one-dimensional photonic crystal slabs made of van der Waals As$_2$S$_3$, a material with one of the highest known in-plane anisotropy. By utilizing the anisotropy degree of freedom provided by As$_2$S$_3$, we design Fabry-P\'erot cavities with constitutional and configurational geometrical chiralities. We demonstrate that in cavities with constitutional chirality, electromagnetic modes of left or right handedness exist due to the chirality of both mirrors, often referred to as handedness preserving mirrors in the literature. At the same time, cavities with configurational chirality support modes of both handednesses due to chiral morphology of the entire structure, set by the twist angle between the optical axes of the upper and lower non-chiral anisotropic mirrors. The developed chiral Fabry-P\'erot cavities can be tuned to the technologically available distance between the mirrors by properly twisting them, making such systems a prospective platform for the coupling of chiral light with chiral matter., Comment: 33 pages, 9 figures

The need for skilled medical support is growing in the era of digital healthcare. This research presents an innovative strategy, utilizing the RuBERT model, for categorizing user inquiries in the field of medical consultation with a focus on expert specialization. By harnessing the capabilities of transformers, we fine-tuned the pre-trained RuBERT model on a varied dataset, which facilitates precise correspondence between queries and particular medical specialisms. Using a comprehensive dataset, we have demonstrated our approach's superior performance with an F1-score of over 92%, calculated through both cross-validation and the traditional split of test and train datasets. Our approach has shown excellent generalization across medical domains such as cardiology, neurology and dermatology. This methodology provides practical benefits by directing users to appropriate specialists for prompt and targeted medical advice. It also enhances healthcare system efficiency, reduces practitioner burden, and improves patient care quality. In summary, our suggested strategy facilitates the attainment of specific medical knowledge, offering prompt and precise advice within the digital healthcare field., Comment: 16 pages, 5 figures

We propose an alternative method to extract the leading-order hadronic contribution to the muon g-2, $a_{\mu}^\text{HLO}$, with the MUonE experiment. In contrast to the traditional method based on the integral of the hadronic contribution to the running of the effective fine-structure constant $\Delta\alpha_{had}$ in the space-like region, our approach relies on the computation of the derivatives of $\Delta\alpha_{had}(t)$ at zero squared momentum transfer $t$. We show that this approach allows to extract $\sim 99\%$ of the total value of $a_{\mu}^\text{HLO}$ from the MUonE data, while the remaining $\sim 1\%$ can be computed combining perturbative QCD and data on $e^+e^-$ annihilation to hadrons. This leads to a competitive evaluation of $a_{\mu}^\text{HLO}$ which is robust against the parameterization used to model $\Delta\alpha_{had}(t)$ in the MUonE kinematic region, thanks to the analyticity properties of $\Delta\alpha_{had}(t)$, which can be expanded as a polynomial at $t\sim 0$., Comment: 12 pages, 8 figures, version accepted for publication in Phys. Lett. B

The cross section of the process $e^+e^-\to\pi^+\pi^-$ has been measured in the center-of-mass energy range from 0.32 to 1.2 GeV with the CMD-3 detector at the electron-positron collider VEPP-2000. The measurement is based on an integrated luminosity of about 88 pb$^{-1}$, of which 62 pb$^{-1}$ represent a complete dataset collected by CMD-3 at center-of-mass energies below 1 GeV. In the dominant region near the $\rho$ resonance a systematic uncertainty of 0.7% was achieved. The implications of the presented results for the evaluation of the hadronic contribution to the anomalous magnetic moment of the muon are discussed., Comment: 8 pages, 3 figures; as published in Phys. Rev. Lett. 132, 231903

The cross section of the process $e^+e^-\to\pi^+\pi^-$ has been measured in the center of mass energy range from 0.32 to 1.2 GeV with the CMD-3 detector at the electron-positron collider VEPP-2000. The measurement is based on a full dataset collected below 1 GeV during three data taking seasons, corresponding to an integrated luminosity of about 62 pb$^{-1}$. In the dominant $\rho$-resonance region, a systematic uncertainty of 0.7% has been reached. At energies around $\phi$-resonance the $\pi^+\pi^-$ production cross section was measured for the first time with high beam energy resolution. The forward-backward charge asymmetry in the $\pi^+\pi^-$ production has also been measured. It shows a strong deviation from the theoretical prediction based on the conventional scalar quantum electrodynamics framework, and it is in good agreement with the generalized vector-meson-dominance and dispersive-based predictions. The impact of the presented results on the evaluation of the hadronic contribution to the anomalous magnetic moment of muon is discussed., Comment: 54 pages, 36 figures; as published in Phys. Rev. D 109, 112002

While neural networks-based photo processing solutions can provide a better image quality compared to the traditional ISP systems, their application to mobile devices is still very limited due to their very high computational complexity. In this paper, we present a novel MicroISP model designed specifically for edge devices, taking into account their computational and memory limitations. The proposed solution is capable of processing up to 32MP photos on recent smartphones using the standard mobile ML libraries and requiring less than 1 second to perform the inference, while for FullHD images it achieves real-time performance. The architecture of the model is flexible, allowing to adjust its complexity to devices of different computational power. To evaluate the performance of the model, we collected a novel Fujifilm UltraISP dataset consisting of thousands of paired photos captured with a normal mobile camera sensor and a professional 102MP medium-format FujiFilm GFX100 camera. The experiments demonstrated that, despite its compact size, the MicroISP model is able to provide comparable or better visual results than the traditional mobile ISP systems, while outperforming the previously proposed efficient deep learning based solutions. Finally, this model is also compatible with the latest mobile AI accelerators, achieving good runtime and low power consumption on smartphone NPUs and APUs. The code, dataset and pre-trained models are available on the project website: https://people.ee.ethz.ch/~ihnatova/microisp.html, Comment: arXiv admin note: text overlap with arXiv:2211.06263

The increased importance of mobile photography created a need for fast and performant RAW image processing pipelines capable of producing good visual results in spite of the mobile camera sensor limitations. While deep learning-based approaches can efficiently solve this problem, their computational requirements usually remain too large for high-resolution on-device image processing. To address this limitation, we propose a novel PyNET-V2 Mobile CNN architecture designed specifically for edge devices, being able to process RAW 12MP photos directly on mobile phones under 1.5 second and producing high perceptual photo quality. To train and to evaluate the performance of the proposed solution, we use the real-world Fujifilm UltraISP dataset consisting on thousands of RAW-RGB image pairs captured with a professional medium-format 102MP Fujifilm camera and a popular Sony mobile camera sensor. The results demonstrate that the PyNET-V2 Mobile model can substantially surpass the quality of tradition ISP pipelines, while outperforming the previously introduced neural network-based solutions designed for fast image processing. Furthermore, we show that the proposed architecture is also compatible with the latest mobile AI accelerators such as NPUs or APUs that can be used to further reduce the latency of the model to as little as 0.5 second. The dataset, code and pre-trained models used in this paper are available on the project website: https://github.com/gmalivenko/PyNET-v2

As mobile cameras with compact optics are unable to produce a strong bokeh effect, lots of interest is now devoted to deep learning-based solutions for this task. In this Mobile AI challenge, the target was to develop an efficient end-to-end AI-based bokeh effect rendering approach that can run on modern smartphone GPUs using TensorFlow Lite. The participants were provided with a large-scale EBB! bokeh dataset consisting of 5K shallow / wide depth-of-field image pairs captured using the Canon 7D DSLR camera. The runtime of the resulting models was evaluated on the Kirin 9000's Mali GPU that provides excellent acceleration results for the majority of common deep learning ops. A detailed description of all models developed in this challenge is provided in this paper., Comment: arXiv admin note: substantial text overlap with arXiv:2211.03885; text overlap with arXiv:2105.07809, arXiv:2211.04470, arXiv:2211.05256, arXiv:2211.05910

Video super-resolution is one of the most popular tasks on mobile devices, being widely used for an automatic improvement of low-bitrate and low-resolution video streams. While numerous solutions have been proposed for this problem, they are usually quite computationally demanding, demonstrating low FPS rates and power efficiency on mobile devices. In this Mobile AI challenge, we address this problem and propose the participants to design an end-to-end real-time video super-resolution solution for mobile NPUs optimized for low energy consumption. The participants were provided with the REDS training dataset containing video sequences for a 4X video upscaling task. The runtime and power efficiency of all models was evaluated on the powerful MediaTek Dimensity 9000 platform with a dedicated AI processing unit capable of accelerating floating-point and quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 500 FPS rate and 0.2 [Watt / 30 FPS] power consumption. A detailed description of all models developed in the challenge is provided in this paper., Comment: arXiv admin note: text overlap with arXiv:2105.08826, arXiv:2105.07809, arXiv:2211.04470, arXiv:2211.03885

Image super-resolution is a common task on mobile and IoT devices, where one often needs to upscale and enhance low-resolution images and video frames. While numerous solutions have been proposed for this problem in the past, they are usually not compatible with low-power mobile NPUs having many computational and memory constraints. In this Mobile AI challenge, we address this problem and propose the participants to design an efficient quantized image super-resolution solution that can demonstrate a real-time performance on mobile NPUs. The participants were provided with the DIV2K dataset and trained INT8 models to do a high-quality 3X image upscaling. The runtime of all models was evaluated on the Synaptics VS680 Smart Home board with a dedicated edge NPU capable of accelerating quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 60 FPS rate when reconstructing Full HD resolution images. A detailed description of all models developed in the challenge is provided in this paper., Comment: arXiv admin note: text overlap with arXiv:2105.07825, arXiv:2105.08826, arXiv:2211.04470, arXiv:2211.03885, arXiv:2211.05256

The role of mobile cameras increased dramatically over the past few years, leading to more and more research in automatic image quality enhancement and RAW photo processing. In this Mobile AI challenge, the target was to develop an efficient end-to-end AI-based image signal processing (ISP) pipeline replacing the standard mobile ISPs that can run on modern smartphone GPUs using TensorFlow Lite. The participants were provided with a large-scale Fujifilm UltraISP dataset consisting of thousands of paired photos captured with a normal mobile camera sensor and a professional 102MP medium-format FujiFilm GFX100 camera. The runtime of the resulting models was evaluated on the Snapdragon's 8 Gen 1 GPU that provides excellent acceleration results for the majority of common deep learning ops. The proposed solutions are compatible with all recent mobile GPUs, being able to process Full HD photos in less than 20-50 milliseconds while achieving high fidelity results. A detailed description of all models developed in this challenge is provided in this paper.

Various depth estimation models are now widely used on many mobile and IoT devices for image segmentation, bokeh effect rendering, object tracking and many other mobile tasks. Thus, it is very crucial to have efficient and accurate depth estimation models that can run fast on low-power mobile chipsets. In this Mobile AI challenge, the target was to develop deep learning-based single image depth estimation solutions that can show a real-time performance on IoT platforms and smartphones. For this, the participants used a large-scale RGB-to-depth dataset that was collected with the ZED stereo camera capable to generated depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the Raspberry Pi 4 platform, where the developed solutions were able to generate VGA resolution depth maps at up to 27 FPS while achieving high fidelity results. All models developed in the challenge are also compatible with any Android or Linux-based mobile devices, their detailed description is provided in this paper., Comment: arXiv admin note: substantial text overlap with arXiv:2105.08630, arXiv:2211.03885; text overlap with arXiv:2105.08819, arXiv:2105.08826, arXiv:2105.08629, arXiv:2105.07809, arXiv:2105.07825

The electronic parameters and, in particular, the isotropic electrostatic polarizability (IEP) of sub-nanoscale magnesium clusters were studied in an expanded set of 1237 structurally unique isomers found in the course of direct global DFT optimization of the structure of Mg2-Mg32 clusters at the BP86 level, as well as using global optimization based on DFT-calibrated MTP potential for some larger structures. The calculation of the polarizability at the same DFT level reveals an unusual property of the IEP - the dependence of the IEP of the most favorable isomers on the cluster nuclearity n is linear with a high correlation coefficient, and its value for each n is close to the minimum value among all found isomers of a given nuclearity. These features take place independently on the cluster structure which allows hypothesizing that the energetic favorability of a cluster structure is connected to their polarizability. A possible explanation of the observed dependence, its significance for quantum chemistry, and the possibility of practical application are discussed.

In this paper we count set closure systems (also known as Moore families) for the case when all single element sets are closed. In particular, we give the numbers of such strict (empty set included) and non-strict families for the base set of size $n=6$. We also provide the number of such inequivalent Moore families with respect to all permutations of the base set up to $n=6$. The search in OEIS and existing literature revealed the coincidence of the found numbers with the entry for D.\ M.~Davis' set union lattice (\seqnum{A235604}, up to $n=5$) and $|\mathcal L_n|$, the number of atomic lattices on $n$ atoms, obtained by S.\ Mapes (up to $n=6$), respectively. Thus we study all those cases, establish one-to-one correspondences between them via Galois adjunctions and Formal Concept Analysis, and provide the reader with two of our enumerative algorithms as well as with the results of these algorithms used for additional tests. Other results include the largest size of intersection free families for $n=6$ plus our conjecture for $n=7$, an upper bound for the number of atomic lattices $\mathcal L_n$, and some structural properties of $\mathcal L_n$ based on the theory of extremal lattices.

We present a new movie and TV show recommendation dataset collected from the real users of MTS Kion video-on-demand platform. In contrast to other popular movie recommendation datasets, such as MovieLens or Netflix, our dataset is based on the implicit interactions registered at the watching time, rather than on explicit ratings. We also provide rich contextual and side information including interactions characteristics (such as temporal information, watch duration and watch percentage), user demographics and rich movies meta-information. In addition, we describe the MTS Kion Challenge - an online recommender systems challenge that was based on this dataset - and provide an overview of the best performing solutions of the winners. We keep the competition sandbox open, so the researchers are welcome to try their own recommendation algorithms and measure the quality on the private part of the dataset., Comment: Accepted at ACM RecSys CARS workshop 2022

The widespread availability of satellite images has allowed researchers to model complex systems such as disease dynamics. However, many satellite images have missing values due to measurement defects, which render them unusable without data imputation. For example, the scanline corrector for the LANDSAT 7 satellite broke down in 2003, resulting in a loss of around 20\% of its data. Inpainting involves predicting what is missing based on the known pixels and is an old problem in image processing, classically based on PDEs or interpolation methods, but recent deep learning approaches have shown promise. However, many of these methods do not explicitly take into account the inherent spatiotemporal structure of satellite images. In this work, we cast satellite image inpainting as a natural meta-learning problem, and propose using convolutional neural processes (ConvNPs) where we frame each satellite image as its own task or 2D regression problem. We show ConvNPs can outperform classical methods and state-of-the-art deep learning inpainting models on a scanline inpainting problem for LANDSAT 7 satellite images, assessed on a variety of in and out-of-distribution images.