Explore the vast range of titles available.

Emerging vulnerabilities demand new conversations With public and academic attention increasingly focused on the new role of machine learning in the health information economy, an unusual and no-longer-esoteric category of vulnerabilities in machine-learning systems could prove important. These vulnerabilities allow a small, carefully designed change in how inputs are presented to a system to completely alter its output, causing it to confidently arrive at manifestly wrong conclusions. These advanced techniques to subvert otherwise-reliable machine-learning systems—so-called adversarial attacks—have, to date, been of interest primarily to computer science researchers ( 1 ). However, the landscape of often-competing interests within health care, and billions of dollars at stake in systems' outputs, implies considerable problems. We outline motivations that various players in the health care system may have to use adversarial attacks and begin a discussion of what to do about them. Far from discouraging continued innovation with medical machine learning, we call for active engagement of medical, technical, legal, and ethical experts in pursuit of efficient, broadly available, and effective health care that machine learning will enable.

Due to the dominant position of deep learning (mostly deep neural networks) in various artificial intelligence applications, recently, ensemble learning based on deep neural networks (ensemble deep learning) has shown significant performances in improving the generalization of learning system. However, since modern deep neural networks usually have millions to billions of parameters, the time and space overheads for training multiple base deep learners and testing with the ensemble deep learner are far greater than that of traditional ensemble learning. Though several algorithms of fast ensemble deep learning have been proposed to promote the deployment of ensemble deep learning in some applications, further advances still need to be made for many applications in specific fields, where the developing time and computing resources are usually restricted or the data to be processed is of large dimensionality. An urgent problem needs to be solved is how to take the significant advantages of ensemble deep learning while reduce the required expenses so that many more applications in specific fields can benefit from it. For the alleviation of this problem, it is essential to know about how ensemble learning has developed under the era of deep learning. Thus, in this article, we present discussions focusing on data analyses of published works, methodologies, recent advances and unattainability of traditional ensemble learning and ensemble deep learning. We hope this article will be helpful to realize the intrinsic problems and technical challenges faced by future developments of ensemble learning under the era of deep learning.

Within the domain of image processing, a wide array of methodologies is dedicated to tasks including denoising, enhancement, segmentation, feature extraction, and classification. These techniques collectively address the challenges and opportunities posed by different aspects of image analysis and manipulation, enabling applications across various fields. Each of these methodologies contributes to refining our understanding of images, extracting essential information, and making informed decisions based on visual data. Traditional image processing methods and Deep Learning (DL) models represent two distinct approaches to tackling image analysis tasks. Traditional methods often rely on handcrafted algorithms and heuristics, involving a series of predefined steps to process images. DL models learn feature representations directly from data, allowing them to automatically extract intricate features that traditional methods might miss. In denoising, techniques like Self2Self NN, Denoising CNNs, DFT-Net, and MPR-CNN stand out, offering reduced noise while grappling with challenges of data augmentation and parameter tuning. Image enhancement, facilitated by approaches such as R2R and LE-net, showcases potential for refining visual quality, though complexities in real-world scenes and authenticity persist. Segmentation techniques, including PSPNet and Mask-RCNN, exhibit precision in object isolation, while handling complexities like overlapping objects and robustness concerns. For feature extraction, methods like CNN and HLF-DIP showcase the role of automated recognition in uncovering image attributes, with trade-offs in interpretability and complexity. Classification techniques span from Residual Networks to CNN-LSTM, spotlighting their potential in precise categorization despite challenges in computational demands and interpretability. This review offers a comprehensive understanding of the strengths and limitations across methodologies, paving the way for informed decisions in practical applications. As the field evolves, addressing challenges like computational resources and robustness remains pivotal in maximizing the potential of image processing techniques.

Machine learning methods strive to acquire a robust model during the training process that can effectively generalize to test samples, even in the presence of distribution shifts. However, these methods often suffer from performance degradation due to unknown test distributions. Test-time adaptation (TTA), an emerging paradigm, has the potential to adapt a pre-trained model to unlabeled data during testing, before making predictions. Recent progress in this paradigm has highlighted the significant benefits of using unlabeled data to train self-adapted models prior to inference. In this survey, we categorize TTA into several distinct groups based on the form of test data, namely, test-time domain adaptation, test-time batch adaptation, and online test-time adaptation. For each category, we provide a comprehensive taxonomy of advanced algorithms and discuss various learning scenarios. Furthermore, we analyze relevant applications of TTA and discuss open challenges and promising areas for future research. For a comprehensive list of TTA methods, kindly refer to https://github.com/tim-learn/awesome-test-time-adaptation .

Machine learning has played an essential role in the past decades and has been in lockstep with the main advances in computer technology. Given the massive amount of data generated daily, there is a need for even faster and more effective machine learning algorithms that can provide updated models for real-time applications and on-demand tools. This paper presents FEMa—a finite element machine classifier—for supervised learning problems, where each training sample is the center of a basis function, and the whole training set is modeled as a probabilistic manifold for classification purposes. FEMa has its theoretical basis in the finite element method, which is widely used for numeral analysis in engineering problems. It is shown FEMa is parameterless and has a quadratic complexity for both training and classification phases when basis functions are used that satisfy certain properties. The proposed classifier yields very competitive results when compared to some state-of-the-art supervised pattern recognition techniques.

An ongoing challenge in neuromorphic computing is to devise general and computationally efficient models of inference and learning which are compatible with the spatial and temporal constraints of the brain. One increasingly popular and successful approach is to take inspiration from inference and learning algorithms used in deep neural networks. However, the workhorse of deep learning, the gradient descent Gradient Back Propagation (BP) rule, often relies on the immediate availability of network-wide information stored with high-precision memory during learning, and precise operations that are difficult to realize in neuromorphic hardware. Remarkably, recent work showed that exact backpropagated gradients are not essential for learning deep representations. Building on these results, we demonstrate an event-driven random BP (eRBP) rule that uses an error-modulated synaptic plasticity for learning deep representations. Using a two-compartment Leaky Integrate & Fire (I&F) neuron, the rule requires only one addition and two comparisons for each synaptic weight, making it very suitable for implementation in digital or mixed-signal neuromorphic hardware. Our results show that using eRBP, deep representations are rapidly learned, achieving classification accuracies on permutation invariant datasets comparable to those obtained in artificial neural network simulations on GPUs, while being robust to neural and synaptic state quantizations during learning.

Using deep learning to combine target sequence and chromatin accessibility data boosts the accuracy of CRISPR–Cpf1 guide RNA activity We present two algorithms to predict the activity of AsCpf1 guide RNAs. Indel frequencies for 15,000 target sequences were used in a deep-learning framework based on a convolutional neural network to train Seq-deepCpf1. We then incorporated chromatin accessibility information to create the better-performing DeepCpf1 algorithm for cell lines for which such information is available and show that both algorithms outperform previous machine learning algorithms on our own and published data sets.

Deep learning has recently been used for the analysis of neuroimages, such as structural magnetic resonance imaging (MRI), functional MRI, and positron emission tomography (PET), and it has achieved significant performance improvements over traditional machine learning in computer-aided diagnosis of brain disorders. This paper reviews the applications of deep learning methods for neuroimaging-based brain disorder analysis. We first provide a comprehensive overview of deep learning techniques and popular network architectures by introducing various types of deep neural networks and recent developments. We then review deep learning methods for computer-aided analysis of four typical brain disorders, including Alzheimer's disease, Parkinson's disease, Autism spectrum disorder, and Schizophrenia, where the first two diseases are neurodegenerative disorders and the last two are neurodevelopmental and psychiatric disorders, respectively. More importantly, we discuss the limitations of existing studies and present possible future directions.

HintergrundDank Mammographie-Screening-Programmen (MSP) kann Brustkrebs erwiesenermaßen in früheren Stadien entdeckt werden, was weniger eingreifende Therapien erlaubt und zu einem besseren Überleben führt. Kritisch beurteilt werden die beträchtliche Zahl der Intervallkarzinome (IBC) und zusätzlich notwendige Abklärungen, bei denen sich in der Mehrzahl erweist, dass kein Karzinom vorliegt.FragestellungIn den letzten Jahren wurden von Firmen und Universitäten mittels maschinellem Lernen (ML) leistungsfähige Algorithmen entwickelt, welche erstaunliche Fähigkeiten zum Lesen von Mammographien zeigen. Können dadurch MSP qualitativ verbessert werden?MethodeMittels der Software ProFound AI® (iCAD, Nashua, NH, USA) wurden retrospektiv die ursprünglichen Screening-Mammographien von 251 Fällen mit Intervallkarzinom untersucht und die Resultate (Case-Score, Risk-Score) mit denen einer Kontrollgruppe verglichen. Darüber hinaus wurde die relevante aktuelle Literatur studiert.ErgebnisseDie Verteilung des Case-Score wie auch des Risk-Score der Mammographien mit späterem IBC war signifikant zu höherem Risiko verschoben im Vergleich zur Kontrolle, ähnlich wie in anderen Studien.SchlussfolgerungRetrospektive Studien, wie auch eigene Daten zeigen, dass möglicherweise künstliche Intelligenz (KI) in Zukunft das Vorgehen bei MSP ändern wird in Richtung personalisiertem Screening, mit deutlicher Entlastung der Radiologen, weniger Abklärungen und einer verminderten Anzahl von IBC. Für eine solche Umsetzung braucht es die Resultate prospektiver Studien.