Neuroimage processing tasks like segmentation, reconstruction, and registration are central to the study of neuroscience. Robust deep learning strategies and architectures used to solve these tasks are often similar. Yet, when presented with a new task or a dataset with different visual characteristics, practitioners most often need to train a new model, or fine-tune an existing one. This is a time-consuming process that poses a substantial barrier for the thousands of neuroscientists and clinical researchers who often lack the resources or machine-learning expertise to train deep learning models. In practice, this leads to a lack of adoption of deep learning, and neuroscience tools being dominated by classical frameworks. We introduce Neuralizer, a single model that generalizes to previously unseen neuroimaging tasks and modalities without the need for re-training or fine-tuning. Tasks do not have to be known a priori, and generalization happens in a single forward pass during inference. The model can solve processing tasks across multiple image modalities, acquisition methods, and datasets, and generalize to tasks and modalities it has not been trained on. Our experiments on coronal slices show that when few annotated subjects are available, our multi-task network outperforms task-specific baselines without training on the task.
In this article, we propose a framework for contactless human-computer interaction (HCI) using novel tracking techniques based on deep learning-based super-resolution and tracking algorithms. Our system offers unprecedented high-resolution tracking of hand position and motion characteristics by leveraging spatial and temporal features embedded in the reflected radar waveform. Rather than classifying samples from a predefined set of hand gestures, as common in existing work on deep learning with mmWave radar, our proposed imager employs a regressive full convolutional neural network (FCNN) approach to improve localization accuracy by spatial super-resolution. While the proposed techniques are suitable for a host of tracking applications, this article focuses on their application as a musical interface to demonstrate the robustness of the gesture sensing pipeline and deep learning signal processing chain. The user can control the instrument by varying the position and velocity of their hand above the vertically-facing sensor. By employing a commercially available multiple-input-multiple-output (MIMO) radar rather than a traditional optical sensor, our framework demonstrates the efficacy of the mmWave sensing modality for fine motion tracking and offers an elegant solution to a host of HCI tasks. Additionally, we provide a freely available software package and user interface for controlling the device, streaming the data to MATLAB in real-time, and increasing accessibility to the signal processing and device interface functionality utilized in this article.
The issue of Automatic License Plate Recognition (ALPR) has been one of the most challenging issues in recent years. Weather conditions, camera angle of view, lighting conditions, different characters written on license plates, and many other factors are among the challenges for the issue of ALPR. Given the advances that have been made in recent years in the field of deep neural networks, some types of neural networks and models based on them can be used to perform the task of Iranian license plate recognition. In the proposed method presented in this paper, the license plate recognition is done in two steps. The first step is to detect the rectangles of the license plates from the input image. In the second step, these license plates are cropped from the image and their characters are recognized. For the first step, 3065 images including license plates and for the second step, 3364 images including characters of license plates have been prepared and considered as the desired datasets. In the first step, license plates are detected using the YOLOv4-tiny model, which is based on Convolutional Neural Network (CNN). In the next step, the characters of these license plates are recognized using Convolutional Recurrent Neural Network (CRNN), and Connectionist Temporal Classification (CTC). In the second step, there is no need to segment and label the characters separately, only one string of numbers and letters is enough for the labels.
With the advent of deep learning methods, Neural Machine Translation (NMT) systems have become increasingly powerful. However, deep learning based systems are susceptible to adversarial attacks, where imperceptible changes to the input can cause undesirable changes at the output of the system. To date there has been little work investigating adversarial attacks on sequence-to-sequence systems, such as NMT models. Previous work in NMT has examined attacks with the aim of introducing target phrases in the output sequence. In this work, adversarial attacks for NMT systems are explored from an output perception perspective. Thus the aim of an attack is to change the perception of the output sequence, without altering the perception of the input sequence. For example, an adversary may distort the sentiment of translated reviews to have an exaggerated positive sentiment. In practice it is challenging to run extensive human perception experiments, so a proxy deep-learning classifier applied to the NMT output is used to measure perception changes. Experiments demonstrate that the sentiment perception of NMT systems' output sequences can be changed significantly with small imperceptible changes to input sequences.
Deep-learning-based super-resolution photoacoustic angiography (PAA) is a powerful tool that restores blood vessel images from under-sampled images to facilitate disease diagnosis. Nonetheless, due to the scarcity of training samples, PAA super-resolution models often exhibit inadequate generalization capabilities, particularly in the context of continuous monitoring tasks. To address this challenge, we propose a novel approach that employs a super-resolution PAA method trained with forged PAA images. We start by generating realistic PAA images of human lips from hand-drawn curves using a diffusion-based image generation model. Subsequently, we train a self-similarity-based super-resolution model with these forged PAA images. Experimental results show that our method outperforms the super-resolution model trained with authentic PAA images in both original-domain and cross-domain tests. Specially, our approach boosts the quality of super-resolution reconstruction using the images forged by the deep learning model, indicating that the collaboration between deep learning models can facilitate generalization, despite limited initial dataset. This approach shows promising potential for exploring zero-shot learning neural networks for vision tasks.
Using knowledge graphs to assist deep learning models in making recommendation decisions has recently been proven to effectively improve the model's interpretability and accuracy. This paper introduces an end-to-end deep learning model, named RKGCN, which dynamically analyses each user's preferences and makes a recommendation of suitable items. It combines knowledge graphs on both the item side and user side to enrich their representations to maximize the utilization of the abundant information in knowledge graphs. RKGCN is able to offer more personalized and relevant recommendations in three different scenarios. The experimental results show the superior effectiveness of our model over 5 baseline models on three real-world datasets including movies, books, and music.
Epilepsy affects more than 50 million people worldwide, making it one of the world's most prevalent neurological diseases. The main symptom of epilepsy is seizures, which occur abruptly and can cause serious injury or death. The ability to predict the occurrence of an epileptic seizure could alleviate many risks and stresses people with epilepsy face. We formulate the problem of detecting preictal (or pre-seizure) with reference to normal EEG as a precursor to incoming seizure. To this end, we developed several supervised deep learning approaches to identify preictal EEG from normal EEG. We further develop novel unsupervised deep learning approaches to train the models on only normal EEG, and detecting pre-seizure EEG as an anomalous event. These deep learning models were trained and evaluated on two large EEG seizure datasets in a person-specific manner. We found that both supervised and unsupervised approaches are feasible; however, their performance varies depending on the patient, approach and architecture. This new line of research has the potential to develop therapeutic interventions and save human lives.
The field of speech processing has undergone a transformative shift with the advent of deep learning. The use of multiple processing layers has enabled the creation of models capable of extracting intricate features from speech data. This development has paved the way for unparalleled advancements in speech recognition, text-to-speech synthesis, automatic speech recognition, and emotion recognition, propelling the performance of these tasks to unprecedented heights. The power of deep learning techniques has opened up new avenues for research and innovation in the field of speech processing, with far-reaching implications for a range of industries and applications. This review paper provides a comprehensive overview of the key deep learning models and their applications in speech-processing tasks. We begin by tracing the evolution of speech processing research, from early approaches, such as MFCC and HMM, to more recent advances in deep learning architectures, such as CNNs, RNNs, transformers, conformers, and diffusion models. We categorize the approaches and compare their strengths and weaknesses for solving speech-processing tasks. Furthermore, we extensively cover various speech-processing tasks, datasets, and benchmarks used in the literature and describe how different deep-learning networks have been utilized to tackle these tasks. Additionally, we discuss the challenges and future directions of deep learning in speech processing, including the need for more parameter-efficient, interpretable models and the potential of deep learning for multimodal speech processing. By examining the field's evolution, comparing and contrasting different approaches, and highlighting future directions and challenges, we hope to inspire further research in this exciting and rapidly advancing field.
Image reconstruction in optoacoustic tomography (OAT) is a trending learning task highly dependent on measured physical magnitudes present at sensing time. The large number of different settings, and also the presence of uncertainties or partial knowledge of parameters, can lead to reconstructions algorithms that are specifically tailored and designed to a particular configuration which could not be the one that will be ultimately faced in a final practical situation. Being able to learn reconstruction algorithms that are robust to different environments (e.g. the different OAT image reconstruction settings) or invariant to such environments is highly valuable because it allows to focus on what truly matters for the application at hand and discard what are considered spurious features. In this work we explore the use of deep learning algorithms based on learning invariant and robust representations for the OAT inverse problem. In particular, we consider the application of the ANDMask scheme due to its easy adaptation to the OAT problem. Numerical experiments are conducted showing that, when out-of-distribution generalization (against variations in parameters such as the location of the sensors) is imposed, there is no degradation of the performance and, in some cases, it is even possible to achieve improvements with respect to standard deep learning approaches where invariance robustness is not explicitly considered.
Pile-up signals are frequently produced in experimental physics. They create inaccurate physics data with high uncertainty and cause various problems. Therefore, the correction to pile-up signals is crucially required. In this study, we implemented a deep learning method to restore the original signals from the pile-up signals. We showed that a deep learning model could accurately reconstruct the original signal waveforms from the pile-up waveforms. By substituting the pile-up signals with the original signals predicted by the model, the energy and timing resolutions of the data are notably enhanced. The model implementation significantly improved the quality of the particle identification plot and particle tracks. This method is applicable to similar problems, such as separating multiple signals or correcting pile-up signals with other types of noises and backgrounds.
Providing online adaptive lightweight time series anomaly detection without human intervention and domain knowledge is highly valuable. Several such anomaly detection approaches have been introduced in the past years, but all of them were only implemented in one deep learning library. With the development of deep learning libraries, it is unclear how different deep learning libraries impact these anomaly detection approaches since there is no such evaluation available. Randomly choosing a deep learning library to implement an anomaly detection approach might not be able to show the true performance of the approach. It might also mislead users in believing one approach is better than another. Therefore, in this paper, we investigate the impact of deep learning libraries on online adaptive lightweight time series anomaly detection by implementing two state-of-the-art anomaly detection approaches in three well-known deep learning libraries and evaluating how these two approaches are individually affected by the three deep learning libraries. A series of experiments based on four real-world open-source time series datasets were conducted. The results provide a good reference to select an appropriate deep learning library for online adaptive lightweight anomaly detection.
In this work, we focus on the inverse medium scattering problem (IMSP), which aims to recover unknown scatterers based on measured scattered data. Motivated by the efficient direct sampling method (DSM) introduced in [23], we propose a novel direct sampling-based deep learning approach (DSM-DL)for reconstructing inhomogeneous scatterers. In particular, we use the U-Net neural network to learn the relation between the index functions and the true contrasts. Our proposed DSM-DL is computationally efficient, robust to noise, easy to implement, and able to naturally incorporate multiple measured data to achieve high-quality reconstructions. Some representative tests are carried out with varying numbers of incident waves and different noise levels to evaluate the performance of the proposed method. The results demonstrate the promising benefits of combining deep learning techniques with the DSM for IMSP.
In the past decade, deep learning became the prevalent methodology for predictive modeling thanks to the remarkable accuracy of deep neural networks in tasks such as computer vision and natural language processing. Meanwhile, the structure of neural networks converged back to simpler representations based on piecewise constant and piecewise linear functions such as the Rectified Linear Unit (ReLU), which became the most commonly used type of activation function in neural networks. That made certain types of network structure $\unicode{x2014}$such as the typical fully-connected feedforward neural network$\unicode{x2014}$ amenable to analysis through polyhedral theory and to the application of methodologies such as Linear Programming (LP) and Mixed-Integer Linear Programming (MILP) for a variety of purposes. In this paper, we survey the main topics emerging from this fast-paced area of work, which bring a fresh perspective to understanding neural networks in more detail as well as to applying linear optimization techniques to train, verify, and reduce the size of such networks.
3D shape generation techniques leveraging deep learning have garnered significant interest from both the computer vision and architectural design communities, promising to enrich the content in the virtual environment. However, research on virtual architectural design remains limited, particularly regarding designer-AI collaboration and deep learning-assisted design. In our survey, we reviewed 149 related articles (81.2% of articles published between 2019 and 2023) covering architectural design, 3D shape techniques, and virtual environments. Through scrutinizing the literature, we first identify the principles of virtual architecture and illuminate its current production challenges, including datasets, multimodality, design intuition, and generative frameworks. We then introduce the latest approaches to designing and generating virtual buildings leveraging 3D shape generation and summarize four characteristics of various approaches to virtual architecture. Based on our analysis, we expound on four research agendas, including agency, communication, user consideration, and integrating tools. Additionally, we highlight four important enablers of ubiquitous interaction with immersive systems in deep learning-assisted architectural generation. Our work contributes to fostering understanding between designers and deep learning techniques, broadening access to designer-AI collaboration. We advocate for interdisciplinary efforts to address this timely research topic, facilitating content designing and generation in the virtual environment.
A brain tumor, whether benign or malignant, can potentially be life threatening and requires painstaking efforts in order to identify the type, origin and location, let alone cure one. Manual segmentation by medical specialists can be time-consuming, which calls out for the involvement of technology to hasten the process with high accuracy. For the purpose of medical image segmentation, we inspected and identified the capable deep learning model, which shows consistent results in the dataset used for brain tumor segmentation. In this study, a public MRI imaging dataset contains 3064 TI-weighted images from 233 patients with three variants of brain tumor, viz. meningioma, glioma, and pituitary tumor. The dataset files were converted and preprocessed before indulging into the methodology which employs implementation and training of some well-known image segmentation deep learning models like U-Net & Attention U-Net with various backbones, Deep Residual U-Net, ResUnet++ and Recurrent Residual U-Net. with varying parameters, acquired from our review of the literature related to human brain tumor classification and segmentation. The experimental findings showed that among all the applied approaches, the recurrent residual U-Net which uses Adam optimizer reaches a Mean Intersection Over Union of 0.8665 and outperforms other compared state-of-the-art deep learning models. The visual findings also show the remarkable results of the brain tumor segmentation from MRI scans and demonstrates how useful the algorithm will be for physicians to extract the brain cancers automatically from MRI scans and serve humanity.
Tooth segmentation from intraoral scans is a crucial part of digital dentistry. Many Deep Learning based tooth segmentation algorithms have been developed for this task. In most of the cases, high accuracy has been achieved, although, most of the available tooth segmentation techniques make an implicit restrictive assumption of full jaw model and they report accuracy based on full jaw models. Medically, however, in certain cases, full jaw tooth scan is not required or may not be available. Given this practical issue, it is important to understand the robustness of currently available widely used Deep Learning based tooth segmentation techniques. For this purpose, we applied available segmentation techniques on partial intraoral scans and we discovered that the available deep Learning techniques under-perform drastically. The analysis and comparison presented in this work would help us in understanding the severity of the problem and allow us to develop robust tooth segmentation technique without strong assumption of full jaw model.
Recent research demonstrates that deep learning models are capable of precisely extracting bio-information (e.g. race, gender and age) from patients' Chest X-Rays (CXRs). In this paper, we further show that deep learning models are also surprisingly accurate at recognition, i.e., distinguishing CXRs belonging to the same patient from those belonging to different patients. These findings suggest potential privacy considerations that the medical imaging community should consider with the proliferation of large public CXR databases.
In low-resource computing contexts, such as smartphones and other tiny devices, Both deep learning and machine learning are being used in a lot of identification systems. as authentication techniques. The transparent, contactless, and non-invasive nature of these face recognition technologies driven by AI has led to their meteoric rise in popularity in recent years. While they are mostly successful, there are still methods to get inside without permission by utilising things like pictures, masks, glasses, etc. In this research, we present an alternate authentication process that makes use of both facial recognition and the individual's distinctive temporal facial feature motions while they speak a password. Because the suggested methodology allows for a password to be specified in any language, it is not limited by language. The suggested model attained an accuracy of 96.1% when tested on the industry-standard MIRACL-VC1 dataset, demonstrating its efficacy as a reliable and powerful solution. In addition to being data-efficient, the suggested technique shows promising outcomes with as little as 10 positive video examples for training the model. The effectiveness of the network's training is further proved via comparisons with other combined facial recognition and lip reading models.
We present a novel application of category theory for deep learning. We show how category theory can be used to understand and work with the linear layer functions of group equivariant neural networks whose layers are some tensor power space of $\mathbb{R}^{n}$ for the groups $S_n$, $O(n)$, $Sp(n)$, and $SO(n)$. By using category theoretic constructions, we build a richer structure that is not seen in the original formulation of these neural networks, leading to new insights. In particular, we outline the development of an algorithm for quickly computing the result of a vector that is passed through an equivariant, linear layer for each group in question. The success of our approach suggests that category theory could be beneficial for other areas of deep learning.
In the recent years, hyperspectral imaging (HSI) has gained considerably popularity among computer vision researchers for its potential in solving remote sensing problems, especially in agriculture field. However, HSI classification is a complex task due to the high redundancy of spectral bands, limited training samples, and non-linear relationship between spatial position and spectral bands. Fortunately, deep learning techniques have shown promising results in HSI analysis. This literature review explores recent applications of deep learning approaches such as Autoencoders, Convolutional Neural Networks (1D, 2D, and 3D), Recurrent Neural Networks, Deep Belief Networks, and Generative Adversarial Networks in agriculture. The performance of these approaches has been evaluated and discussed on well-known land cover datasets including Indian Pines, Salinas Valley, and Pavia University.