Search

Alzheimer's disease (AD) poses significant global health challenges due to its increasing prevalence and associated societal costs. Early detection and diagnosis of AD are critical for delaying progression and improving patient outcomes. Traditional diagnostic methods and single-modality data often fall short in identifying early-stage AD and distinguishing it from Mild Cognitive Impairment (MCI). This study addresses these challenges by introducing a novel approach: multImodal enseMble via class BALancing diversity for iMbalancEd Data (IMBALMED). IMBALMED integrates multimodal data from the Alzheimer's Disease Neuroimaging Initiative database, including clinical assessments, neuroimaging phenotypes, biospecimen and subject characteristics data. It employs an ensemble of model classifiers, each trained with different class balancing techniques, to overcome class imbalance and enhance model accuracy. We evaluate IMBALMED on two diagnostic tasks (binary and ternary classification) and four binary early detection tasks (at 12, 24, 36, and 48 months), comparing its performance with state-of-the-art algorithms and an unbalanced dataset method. IMBALMED demonstrates superior diagnostic accuracy and predictive performance in both binary and ternary classification tasks, significantly improving early detection of MCI at 48-month time point. The method shows improved classification performance and robustness, offering a promising solution for early detection and management of AD.

We utilize a method using frequency combs to construct waves that feature superoscillations - local regions of the wave that exhibit a change in phase that the bandlimits of the wave should not otherwise allow. This method has been shown to create superoscillating regions that mimic any analytic function - even ones well outside the bandlimits - to an arbitrary degree of accuracy. We experimentally demonstrate that these waves are extremely robust against noise, allowing for accurate reconstruction of a superoscillating target function thoroughly buried in noise. We additionally show that such a construction can be easily used to range-resolve a signal well below the commonly accepted fundamental limit.

We study conversion processes between gravitons and dark photons and reveal the effects of dark photons on the polarization of gravitational waves. Considering cosmological dark magnetic fields, we investigate the evolution of the intensity and polarization of gravitational waves through the conversion. Specifically, we demonstrate that for minimal coupling between gravitons and dark photons, the intensity, circular polarization, and linear polarization evolve separately. We derive explicit formulas for the statistical mean and variance of the intensity and polarization when the gravitational waves pass through magnetic fields with random orientation. The formulas capture how the initial polarization of dark photons will be imprinted on the observed gravitational wave background., Comment: 34 pages, 1 figure

Endoscopic surgery relies on two-dimensional views, posing challenges for surgeons in depth perception and instrument manipulation. While Monocular Visual Simultaneous Localization and Mapping (MVSLAM) has emerged as a promising solution, its implementation in endoscopic procedures faces significant challenges due to hardware limitations, such as the use of a monocular camera and the absence of odometry sensors. This study presents BodySLAM, a robust deep learning-based MVSLAM approach that addresses these challenges through three key components: CycleVO, a novel unsupervised monocular pose estimation module; the integration of the state-of-the-art Zoe architecture for monocular depth estimation; and a 3D reconstruction module creating a coherent surgical map. The approach is rigorously evaluated using three publicly available datasets (Hamlyn, EndoSLAM, and SCARED) spanning laparoscopy, gastroscopy, and colonoscopy scenarios, and benchmarked against four state-of-the-art methods. Results demonstrate that CycleVO exhibited competitive performance with the lowest inference time among pose estimation methods, while maintaining robust generalization capabilities, whereas Zoe significantly outperformed existing algorithms for depth estimation in endoscopy. BodySLAM's strong performance across diverse endoscopic scenarios demonstrates its potential as a viable MVSLAM solution for endoscopic applications., Comment: 16 pages, 7 figures

Identifying risky driving behavior in real-world situations is essential for the safety of both drivers and pedestrians. However, integrating natural language models in this field remains relatively untapped. To address this, we created a novel multi-modal instruction tuning dataset and driver coaching inference system. Our primary use case is dashcam-based coaching for commercial drivers. The North American Dashcam Market is expected to register a CAGR of 15.4 percent from 2022 to 2027. Our dataset enables language models to learn visual instructions across various risky driving scenarios, emphasizing detailed reasoning crucial for effective driver coaching and managerial comprehension. Our model is trained on road-facing and driver-facing RGB camera footage, capturing the comprehensive scope of driving behavior in vehicles equipped with dashcams., Comment: On-going work

Deep learning has revolutionized biomedical research by providing sophisticated methods to handle complex, high-dimensional data. Multimodal deep learning (MDL) further enhances this capability by integrating diverse data types such as imaging, textual data, and genetic information, leading to more robust and accurate predictive models. In MDL, differently from early and late fusion methods, intermediate fusion stands out for its ability to effectively combine modality-specific features during the learning process. This systematic review aims to comprehensively analyze and formalize current intermediate fusion methods in biomedical applications. We investigate the techniques employed, the challenges faced, and potential future directions for advancing intermediate fusion methods. Additionally, we introduce a structured notation to enhance the understanding and application of these methods beyond the biomedical domain. Our findings are intended to support researchers, healthcare professionals, and the broader deep learning community in developing more sophisticated and insightful multimodal models. Through this review, we aim to provide a foundational framework for future research and practical applications in the dynamic field of MDL.

Long-tailed data is a special type of multi-class imbalanced data with a very large amount of minority/tail classes that have a very significant combined influence. Long-tailed learning aims to build high-performance models on datasets with long-tailed distributions, which can identify all the classes with high accuracy, in particular the minority/tail classes. It is a cutting-edge research direction that has attracted a remarkable amount of research effort in the past few years. In this paper, we present a comprehensive survey of latest advances in long-tailed visual learning. We first propose a new taxonomy for long-tailed learning, which consists of eight different dimensions, including data balancing, neural architecture, feature enrichment, logits adjustment, loss function, bells and whistles, network optimization, and post hoc processing techniques. Based on our proposed taxonomy, we present a systematic review of long-tailed learning methods, discussing their commonalities and alignable differences. We also analyze the differences between imbalance learning and long-tailed learning approaches. Finally, we discuss prospects and future directions in this field., Comment: Current Under Revision at IEEE TNNLS. [This is the long/Full-length version of our Long-Tailed Learning Survey paper]

Handling missing values in tabular datasets presents a significant challenge in training and testing artificial intelligence models, an issue usually addressed using imputation techniques. Here we introduce "Not Another Imputation Method" (NAIM), a novel transformer-based model specifically designed to address this issue without the need for traditional imputation techniques. NAIM employs feature-specific embeddings and a masked self-attention mechanism that effectively learns from available data, thus avoiding the necessity to impute missing values. Additionally, a novel regularization technique is introduced to enhance the model's generalization capability from incomplete data. We extensively evaluated NAIM on 5 publicly available tabular datasets, demonstrating its superior performance over 6 state-of-the-art machine learning models and 4 deep learning models, each paired with 3 different imputation techniques when necessary. The results highlight the efficacy of NAIM in improving predictive performance and resilience in the presence of missing data. To facilitate further research and practical application in handling missing data without traditional imputation methods, we made the code for NAIM available at https://github.com/cosbidev/NAIM., Comment: 12 pages, 8 figures

In the fight against the COVID-19 pandemic, leveraging artificial intelligence to predict disease outcomes from chest radiographic images represents a significant scientific aim. The challenge, however, lies in the scarcity of large, labeled datasets with compatible tasks for training deep learning models without leading to overfitting. Addressing this issue, we introduce a novel multi-dataset multi-task training framework that predicts COVID-19 prognostic outcomes from chest X-rays (CXR) by integrating correlated datasets from disparate sources, distant from conventional multi-task learning approaches, which rely on datasets with multiple and correlated labeling schemes. Our framework hypothesizes that assessing severity scores enhances the model's ability to classify prognostic severity groups, thereby improving its robustness and predictive power. The proposed architecture comprises a deep convolutional network that receives inputs from two publicly available CXR datasets, AIforCOVID for severity prognostic prediction and BRIXIA for severity score assessment, and branches into task-specific fully connected output networks. Moreover, we propose a multi-task loss function, incorporating an indicator function, to exploit multi-dataset integration. The effectiveness and robustness of the proposed approach are demonstrated through significant performance improvements in prognosis classification tasks across 18 different convolutional neural network backbones in different evaluation strategies. This improvement is evident over single-task baselines and standard transfer learning strategies, supported by extensive statistical analysis, showing great application potential.

Generative Adversarial Networks (GANs) have proved as a powerful framework for denoising applications in medical imaging. However, GAN-based denoising algorithms still suffer from limitations in capturing complex relationships within the images. In this regard, the loss function plays a crucial role in guiding the image generation process, encompassing how much a synthetic image differs from a real image. To grasp highly complex and non-linear textural relationships in the training process, this work presents a loss function that leverages the intrinsic multi-scale nature of the Gray-Level-Co-occurrence Matrix (GLCM). Although the recent advances in deep learning have demonstrated superior performance in classification and detection tasks, we hypothesize that its information content can be valuable when integrated into GANs' training. To this end, we propose a differentiable implementation of the GLCM suited for gradient-based optimization. Our approach also introduces a self-attention layer that dynamically aggregates the multi-scale texture information extracted from the images. We validate our approach by carrying out extensive experiments in the context of low-dose CT denoising, a challenging application that aims to enhance the quality of noisy CT scans. We utilize three publicly available datasets, including one simulated and two real datasets. The results are promising as compared to other well-established loss functions, being also consistent across three different GAN architectures. The code is available at: https://github.com/FrancescoDiFeola/DenoTextureLoss

Chromo-natural inflation is a model where non-abelian gauge fields are sustained by the coupling of the axion with the gauge field through the Chern-Simons term. While minimal warm inflation is a model where the axion produces a thermal bath of non-abelian gauge particles through the Chern-Simons term. Since both axion inflation models are based on the same action, a natural question is if those are compatible or not. We study axion inflation with the Chern-Simons term and find that chromo-natural inflation can accommodate radiation with a temperature much larger than the Hubble parameter during inflation, which is a characteristic feature of warm inflation. Thus, we conclude that chromo-natural warm inflation exists, which must have phenomenologically interesting consequences., Comment: 25 pages, 9 figures

We propose high-frequency gravitational wave (GW) detectors with Rydberg atoms. Rydberg atoms are ultra-sensitive detectors of electric fields. By setting up a constant magnetic field, a weak electric field is generated upon the arrival of GWs. The weak electric field signal is then detected by an electromagnetically induced transparency (EIT) in the system of the Rydberg atoms. Recently, the minimum detectable electric field with the Rydberg atoms is further improved by employing superheterodyne detection method. Hence, even the weak signal generated by GWs turns out to be detectable. We calculate the amplitude of Rabi frequency of the Rydberg atoms induced by the GWs and show that the sensitivity of the Rydberg atoms becomes maximum when the size of the Rydberg atoms is close to the wavelength of GWs. We evaluate the minimum detectable amplitude of GWs with Rubidium Rydberg atoms and find that the detector can probe GWs with a frequency $f=4.2$ GHz and an amplitude around $10^{-20}$., Comment: 31 pages, 9 figures

Astrophysical research into exoplanets has delivered thousands of confirmed planets orbiting distant stars. These planets span a wide ranges of size and composition, with diversity also being the hallmark of system configurations, the great majority of which do not resemble our own solar system. Unfortunately, only a handful of the known planets have been characterized spectroscopically thus far, leaving a gaping void in our understanding of planetary formation processes and planetary types. To make progress, astronomers studying exoplanets will need new and innovative technical solutions. Astrophotonics -- an emerging field focused on the application of photonic technologies to observational astronomy -- provides one promising avenue forward. In this paper we discuss various astrophotonic technologies that could aid in the detection and subsequent characterization of planets and in particular themes leading towards the detection of extraterrestrial life., Comment: 9 pages, 2 figures, SPIE Optics and Photonics conference

Tunneling processes in de Sitter spacetime are studied by using the stochastic approach. We exploit the Martin-Siggia-Rose-Janssen-de Dominicis (MSRJD) functional integral to obtain the tunneling rate. The applicability conditions of this method are clarified using the Schwinger-Keldysh formalism. In the case of a shallow potential barrier, we reproduce the Hawking-Moss (HM) tunneling rate. Remarkably, in contrast to HM picture, the configuration derived from the MSRJD functional integral satisfies physically natural boundary conditions. We also discuss the case of a steep potential barrier and find an interesting Coleman-de Luccia (CDL) bubble-like configuration. Our results demonstrate how the bubble nucleation process could be described in the stochastic approach. Our method turns out to be useful for investigating various tunneling processes during inflation., Comment: 39 pages, 10 figures

Contrast Enhanced Spectral Mammography (CESM) is a dual-energy mammographic imaging technique that first needs intravenously administration of an iodinated contrast medium; then, it collects both a low-energy image, comparable to standard mammography, and a high-energy image. The two scans are then combined to get a recombined image showing contrast enhancement. Despite CESM diagnostic advantages for breast cancer diagnosis, the use of contrast medium can cause side effects, and CESM also beams patients with a higher radiation dose compared to standard mammography. To address these limitations this work proposes to use deep generative models for virtual contrast enhancement on CESM, aiming to make the CESM contrast-free as well as to reduce the radiation dose. Our deep networks, consisting of an autoencoder and two Generative Adversarial Networks, the Pix2Pix, and the CycleGAN, generate synthetic recombined images solely from low-energy images. We perform an extensive quantitative and qualitative analysis of the model's performance, also exploiting radiologists' assessments, on a novel CESM dataset that includes 1138 images that, as a further contribution of this work, we make publicly available. The results show that CycleGAN is the most promising deep network to generate synthetic recombined images, highlighting the potential of artificial intelligence techniques for virtual contrast enhancement in this field.

Pulsar timing arrays (PTAs) provide a way to detect gravitational waves (GWs) at nanohertz frequencies. To ensure the detection of GWs, observational data must exhibit the Hellings-Downs angular correlation. It is also known that PTAs can probe ultralight dark matter. In this paper, we consider possible contamination of the Hellings-Downs angular correlation by the ultralight dark matter. We find that ultralight vector dark matter can give rise to the deformation of the Hellings-Downs correlation curve. Thus, the Hellings-Downs correlation curve could contain information on ultralight dark matter with a spin., Comment: 13pages, 3 figures

In the field of lung cancer research, particularly in the analysis of overall survival (OS), artificial intelligence (AI) serves crucial roles with specific aims. Given the prevalent issue of missing data in the medical domain, our primary objective is to develop an AI model capable of dynamically handling this missing data. Additionally, we aim to leverage all accessible data, effectively analyzing both uncensored patients who have experienced the event of interest and censored patients who have not, by embedding a specialized technique within our AI model, not commonly utilized in other AI tasks. Through the realization of these objectives, our model aims to provide precise OS predictions for non-small cell lung cancer (NSCLC) patients, thus overcoming these significant challenges. We present a novel approach to survival analysis with missing values in the context of NSCLC, which exploits the strengths of the transformer architecture to account only for available features without requiring any imputation strategy. More specifically, this model tailors the transformer architecture to tabular data by adapting its feature embedding and masked self-attention to mask missing data and fully exploit the available ones. By making use of ad-hoc designed losses for OS, it is able to account for both censored and uncensored patients, as well as changes in risks over time. We compared our method with state-of-the-art models for survival analysis coupled with different imputation strategies. We evaluated the results obtained over a period of 6 years using different time granularities obtaining a Ct-index, a time-dependent variant of the C-index, of 71.97, 77.58 and 80.72 for time units of 1 month, 1 year and 2 years, respectively, outperforming all state-of-the-art methods regardless of the imputation method used., Comment: 24 pages, 4 figures

Data Augmentation (DA) is a technique to increase the quantity and diversity of the training data, and by that alleviate overfitting and improve generalisation. However, standard DA produces synthetic data for augmentation with limited diversity. Generative Adversarial Networks (GANs) may unlock additional information in a dataset by generating synthetic samples having the appearance of real images. However, these models struggle to simultaneously address three key requirements: fidelity and high-quality samples; diversity and mode coverage; and fast sampling. Indeed, GANs generate high-quality samples rapidly, but have poor mode coverage, limiting their adoption in DA applications. We propose LatentAugment, a DA strategy that overcomes the low diversity of GANs, opening up for use in DA applications. Without external supervision, LatentAugment modifies latent vectors and moves them into latent space regions to maximise the synthetic images' diversity and fidelity. It is also agnostic to the dataset and the downstream task. A wide set of experiments shows that LatentAugment improves the generalisation of a deep model translating from MRI-to-CT beating both standard DA as well GAN-based sampling. Moreover, still in comparison with GAN-based sampling, LatentAugment synthetic samples show superior mode coverage and diversity. Code is available at: https://github.com/ltronchin/LatentAugment.

Accurate forecasting of renewable generation is crucial to facilitate the integration of RES into the power system. Focusing on PV units, forecasting methods can be divided into two main categories: physics-based and data-based strategies, with AI-based models providing state-of-the-art performance. However, while these AI-based models can capture complex patterns and relationships in the data, they ignore the underlying physical prior knowledge of the phenomenon. Therefore, in this paper we propose MATNet, a novel self-attention transformer-based architecture for multivariate multi-step day-ahead PV power generation forecasting. It consists of a hybrid approach that combines the AI paradigm with the prior physical knowledge of PV power generation of physics-based methods. The model is fed with historical PV data and historical and forecast weather data through a multi-level joint fusion approach. The effectiveness of the proposed model is evaluated using the Ausgrid benchmark dataset with different regression performance metrics. The results show that our proposed architecture significantly outperforms the current state-of-the-art methods. These findings demonstrate the potential of MATNet in improving forecasting accuracy and suggest that it could be a promising solution to facilitate the integration of PV energy into the power grid.

The current deep learning approaches for low-dose CT denoising can be divided into paired and unpaired methods. The former involves the use of well-paired datasets, whilst the latter relaxes this constraint. The large availability of unpaired datasets has raised the interest in deepening unpaired denoising strategies that, in turn, need for robust evaluation techniques going beyond the qualitative evaluation. To this end, we can use quantitative image quality assessment scores that we divided into two categories, i.e., paired and unpaired measures. However, the interpretation of unpaired metrics is not straightforward, also because the consistency with paired metrics has not been fully investigated. To cope with this limitation, in this work we consider 15 paired and unpaired scores, which we applied to assess the performance of low-dose CT denoising. We perform an in-depth statistical analysis that not only studies the correlation between paired and unpaired metrics but also within each category. This brings out useful guidelines that can help researchers and practitioners select the right measure for their applications.

We study a scattering problem of gravitational waves (GWs) by an axion domain wall in Chern-Simons (CS) gravity. We find that circular polarization of GWs is produced after passing through the domain wall. It turns out that the circular polarization is sizable if the frequency of the GW is comparable to a critical value determined by the characteristic CS length scale and the energy scale of the axion domain wall. Thus, observations of the circular polarization could give a stringent constraint on the characteristic CS length scale or could be a new avenue to search for axion., Comment: 17 pages, 2 figures

Axions are candidates for dark matter in the universe.We develop an accurate Boltzmann code to calculate the linear growth of the plasma. As an interesting example, we investigate a mixed dark matter model consisting of cold dark matter (CDM) and two-axion dark matter. We analyze the growth of the structure numerically and analytically. We find that an effective single axion with an effective mass and an effective abundance is useful to characterize the two-axion cosmology. Moreover, we generalize the effective single axion description to multi-axion dark matter cosmology. We also compare the results with those of warm dark matter (WDM) model. Moreover, we calculate halo mass functions for the mixed model by using the Press-Schechter model and linear perturbations and then determine the mass function as a function of masses and axion abundance., Comment: 33 pages, 17 figures

We use a result of Hawking and Gilkey to define a Euclidean path integral of gravity and matter which has the special property of being independent of the choice of basis in the space of fields. This property allows the path integral to describe also physical regimes that do not admit position bases. These physical regimes are pre-geometric in the sense that they do not admit a mathematical representation of the physical degrees of freedom in terms of fields that live on a spacetime. In regimes in which a spacetime representation does emerge, the geometric properties of the emergent spacetime, such as its dimension and volume, depend on the balance of fermionic pressure and bosonic and gravitational pull. That balance depends, at any given energy scale, on the number of bosonic and fermionic species that contribute, which in turn depends on their masses. This yields an explicit mechanism by which the effective spacetime dimension can depend on the energy scale., Comment: 6 pages, 3 figures

Dark gauge fields have been discussed as candidates for dark matter recently. If they existed, primordial dark magnetic fields during inflation would have existed. It is believed that primordial gravitational waves (PGWs) arise out of quantum fluctuations during inflation. We study the graviton-dark photon conversion process in the presence of background primordial dark magnetic fields and find that the process induces the tachyonic instability of the PGWs. As a consequence, a peak appears in the power spectrum of PGWs. It turns out that the peak height depends on the direction of observation. The peak frequency could be in the range from $10^{-5}$ to $10^{3}$ Hertz for GUT scale inflation. Hence, the observation of PGWs could provide a new window for probing primordial dark magnetic fields., Comment: 19 pages, 3 figures. arXiv admin note: text overlap with arXiv:2211.05576

We are witnessing a widespread adoption of artificial intelligence in healthcare. However, most of the advancements in deep learning in this area consider only unimodal data, neglecting other modalities. Their multimodal interpretation necessary for supporting diagnosis, prognosis and treatment decisions. In this work we present a deep architecture, which jointly learns modality reconstructions and sample classifications using tabular and imaging data. The explanation of the decision taken is computed by applying a latent shift that, simulates a counterfactual prediction revealing the features of each modality that contribute the most to the decision and a quantitative score indicating the modality importance. We validate our approach in the context of COVID-19 pandemic using the AIforCOVID dataset, which contains multimodal data for the early identification of patients at risk of severe outcome. The results show that the proposed method provides meaningful explanations without degrading the classification performance.

Detecting gravitational waves with frequencies higher than 10 kHz requires new strategies. In previous papers, we proposed magnon gravitational wave detectors and gave the first limit on GHz gravitational waves by reinterpreting the existing data from axion dark matter experiments. In this paper, we show that the sensitivity can be improved by constructing the detector specific to gravitational waves. In particular, we employ an infinite sum of terms in the expansion of Fermi normal coordinates to probe gravitational waves with a wavelength comparable to the detector size. As a consequence, we obtain sensitivity of around $h_c \sim 10^{-20}$., Comment: 14 pages, 2 figures

It is known that magnetic fields exist near black holes and photons can go around black holes due to strong gravity. Utilizing these facts, we can probe hypothetical pseudoscalar particles, so-called axions. In fact, photons can be converted into axions when they propagate in a magnetic field. The conversion of such photons into axions leads to a dimming of the photon ring around the black hole shadow. We show that photon ring dimming can occur efficiently for supermassive black holes. Remarkably, it turns out that the maximal dimming rate of the photon ring is 25%. In the case of M87*, the dimming of 10% will be observed in the X-ray and gamma-ray bands if the angular resolution of $10^{-5}$ arcsec is achieved. The frequency band and the magnitude of the dimming depend on the axion-photon coupling and axion mass. Hence, the distorted spectrum of the photon ring provides a novel tool for detecting axions., Comment: 18 pages, 4 figures