Search Results (1,292)

As artificial intelligence increasingly automates the recognition and analysis of visual content, it poses significant risks to privacy, security, and autonomy. Computer vision systems can surveil and exploit data without consent. With these concerns in mind, we introduce a novel method to control whether images can be recognized by computer vision systems using reversible adversarial examples. These examples are generated to evade unauthorized recognition, allowing only systems with permission to restore the original image by removing the adversarial perturbation with zero-bit error. A key challenge with prior methods is their reliance on merely restoring the examples to a state in which they can be correctly recognized by the model; however, the restored images are not fully consistent with the original images, and they require excessive auxiliary information to achieve reversibility. To achieve zero-bit error restoration, we utilize the differential evolution algorithm to optimize adversarial perturbations while minimizing distortion. Additionally, we introduce a dual-color space detection mechanism to localize perturbations, eliminating the need for extra auxiliary information. Ultimately, when combined with reversible data hiding, adversarial attacks can achieve reversibility. Experimental results demonstrate that the PSNR and SSIM between the restored images by the method and the original images are ∞ and 1, respectively. The PSNR and SSIM between the reversible adversarial examples and the original images are 48.32 dB and 0.9986, respectively. Compared to state-of-the-art methods, the method maintains high visual fidelity at a comparable attack success rate. Full article

Spherical tanks are widely utilized in process industries due to their substantial storage capacity. These industries’ inherent challenges necessitate using highly efficient controllers to manage various process parameters, especially given their nonlinear behavior. This paper proposes the Approximate Generalized Time Moments (AGTM) optimization technique for designing the parameters of multi-model fractional-order controllers for regulating the output (liquid level) of a real-time nonlinear spherical tank. System identification for different regions of the nonlinear process is here innovatively conducted using a black-box model, which is determined to be nonlinear and approximated as a First Order Plus Dead Time (FOPDT) system over each region. Both model identification and controller design are performed in simulation and real-time using a National Instruments NI DAQmx 6211 Data Acquisition (DAQ) card (NI SYSTEMS INDIA PVT. LTD., Bangalore Karnataka, India) and MATLAB/SIMULINK software (MATLAB R2021a). The performance of the overall algorithm is evaluated through simulation and experimental testing, with several setpoints and load changes, and is compared to the performance of other algorithms tuned within the same framework. While traditional approaches, such as integer-order controllers or linear approximations, often struggle to provide consistent performance across the operating range of spherical tanks, it is originally shown how the combination of multi-model fractional-order controller design—AGTM optimization method—GA for expansion point selection and index minimization has benefits in specifically controlling a (difficult to be controlled) nonlinear process. Full article

The TacTip is a cost-effective, 3D-printed optical tactile sensor commonly used in deep learning and reinforcement learning for robotic manipulation. However, its specialized structure, which combines soft materials of varying hardnesses, makes it challenging to simulate the distribution of numerous printed markers on pins. This paper aims to create an interpretable, AI-applicable simulation of the deformation of TacTip under varying pressures and interactions with different objects, addressing the black-box nature of learning and simulation in haptic manipulation. The research focuses on simulating the TacTip sensor’s shape using a fully tunable, chain-based mathematical model, refined through comparisons with real-world measurements. We integrated the WRS system with our theoretical model to evaluate its effectiveness in object pose estimation. The results demonstrated that the prediction accuracy for all markers across a variety of contact scenarios exceeded 92%. Full article

Industry 4.0, leveraging tools like AI and the massive generation of data, is driving a paradigm shift in maintenance management. Specifically, in the realm of Artificial Intelligence (AI), traditionally “black box” models are now being unveiled through explainable AI techniques, which provide insights into model decision-making processes. This study addresses the underutilization of these techniques alongside On-Board Diagnostics data by maintenance management teams in urban bus fleets for addressing key issues affecting vehicle reliability and maintenance needs. In the context of urban bus fleets, diesel particulate filter regeneration processes frequently operate under suboptimal conditions, accelerating engine oil degradation and increasing maintenance costs. Due to limited documentation on the control system of the filter, the maintenance team faces obstacles in proposing solutions based on a comprehensive understanding of the system’s behavior and control logic. The objective of this study is to analyze and predict the various states during the diesel particulate filter regeneration process using Machine Learning and explainable artificial intelligence techniques. The insights obtained aim to provide the maintenance team with a deeper understanding of the filter’s control logic, enabling them to develop proposals grounded in a comprehensive understanding of the system. This study employs a combination of traditional Machine Learning models, including XGBoost, LightGBM, Random Forest, and Support Vector Machine. The target variable, representing three possible regeneration states, was transformed using a one-vs-rest approach, resulting in three binary classification tasks where each target state was individually classified against all other states. Additionally, explainable AI techniques such as Shapley Additive Explanations, Partial Dependence Plots, and Individual Conditional Expectation were applied to interpret and visualize the conditions influencing each regeneration state. The results successfully associate two states with specific operating conditions and establish operational thresholds for key variables, offering practical guidelines for optimizing the regeneration process. Full article

Bisphenol A (BPA) is a typical environmental estrogen that is distributed worldwide and has the potential to pose a hazard to the ecological environment and human health. The development of an efficient and sensitive sensing strategy for the monitoring of BPA residues is of paramount importance. A novel electrochemical sensor based on carbon black and carbon nanofibers composite (CB/f-CNF)-assisted signal amplification has been successfully constructed for the amperometric detection of BPA in foods. Herein, the hybrid CB/f-CNF was prepared using a simple one-step ultrasonication method, and exhibited good electron transfer capability and excellent catalytic properties, which can be attributed to the large surface area of carbon black and the strong enhancement of the conductivity and porosity of carbon nanofibers, which promote a faster electron transfer process on the electrode surface. Under the optimized conditions, the proposed CB/f-CNF/GCE sensor exhibited a wide linear response range (0.4–50.0 × 10⁻⁶ mol/L) with a low limit of detection of 5.9 × 10⁻⁸ mol/L for BPA quantification. Recovery tests were conducted on canned peaches and boxed milk, yielding satisfactory recoveries of 86.0–102.6%. Furthermore, the developed method was employed for the rapid and sensitive detection of BPA in canned meat and packaged milk, demonstrating comparable accuracy to the HPLC method. This work presents an efficient signal amplification strategy through the utilization of carbon/carbon nanocomposite sensitization technology. Full article

The convergence of Artificial Intelligence (AI) and neuroscience is redefining our understanding of the brain, unlocking new possibilities in research, diagnosis, and therapy. This review explores how AI’s cutting-edge algorithms—ranging from deep learning to neuromorphic computing—are revolutionizing neuroscience by enabling the analysis of complex neural datasets, from neuroimaging and electrophysiology to genomic profiling. These advancements are transforming the early detection of neurological disorders, enhancing brain–computer interfaces, and driving personalized medicine, paving the way for more precise and adaptive treatments. Beyond applications, neuroscience itself has inspired AI innovations, with neural architectures and brain-like processes shaping advances in learning algorithms and explainable models. This bidirectional exchange has fueled breakthroughs such as dynamic connectivity mapping, real-time neural decoding, and closed-loop brain–computer systems that adaptively respond to neural states. However, challenges persist, including issues of data integration, ethical considerations, and the “black-box” nature of many AI systems, underscoring the need for transparent, equitable, and interdisciplinary approaches. By synthesizing the latest breakthroughs and identifying future opportunities, this review charts a path forward for the integration of AI and neuroscience. From harnessing multimodal data to enabling cognitive augmentation, the fusion of these fields is not just transforming brain science, it is reimagining human potential. This partnership promises a future where the mysteries of the brain are unlocked, offering unprecedented advancements in healthcare, technology, and beyond. Full article

An important impediment to the incorporation of artificial intelligence-based tools into healthcare is their association with so-called black box medicine, a concept arising due to their complexity and the difficulties in understanding how they reach a decision. This situation may compromise the clinician’s trust in these tools, should any errors occur, and the inability to explain how decisions are reached may affect their relationship with patients. Explainable AI (XAI) aims to overcome this limitation by facilitating a better understanding of how AI models reach their conclusions for users, thereby enhancing trust in the decisions reached. This review first defined the concepts underlying XAI, establishing the tools available and how they can benefit digestive healthcare. Examples of the application of XAI in digestive healthcare were provided, and potential future uses were proposed. In addition, aspects of the regulatory frameworks that must be established and the ethical concerns that must be borne in mind during the development of these tools were discussed. Finally, we considered the challenges that this technology faces to ensure that optimal benefits are reaped, highlighting the need for more research into the use of XAI in this field. Full article

Pneumonia remains a global health issue, creating the need for accurate detection methods for effective treatment. Deep learning models like ResNet50 show promise in detecting pneumonia from chest X-rays; however, their black-box nature limits the transparency, which fails to meet that needed for clinical trust. This study aims to improve model interpretability by comparing four interpretability techniques, which are Layer-wise Relevance Propagation (LRP), Adversarial Training, Class Activation Maps (CAMs), and the Spatial Attention Mechanism, and determining which fits best the model, enhancing its transparency with minimal impact on its performance. Each technique was evaluated for its impact on the accuracy, sensitivity, specificity, AUC-ROC, Mean Relevance Score (MRS), and a calculated trade-off score that balances interpretability and performance. The results indicate that LRP was the most effective in enhancing interpretability, achieving high scores across all metrics without sacrificing diagnostic accuracy. The model achieved 0.91 accuracy and 0.85 interpretability (MRS), demonstrating its potential for clinical integration. In contrast, Adversarial Training, CAMs, and the Spatial Attention Mechanism showed trade-offs between interpretability and performance, each highlighting unique image features but with some impact on specificity and accuracy. Full article

This study explores the optimisation of rearing substrates for black soldier fly larvae (BSFL). First, the ideal dry matter content of substrates was determined, comparing the standard 30% dry matter (DM) with substrates hydrated to their maximum water holding capacity (WHC). Substrates at maximal WHC yielded significantly higher larval survival rates (p = 0.0006). Consequently, the WHC approach was adopted for further experiments. Using these hydrated artificial substrates, fractional factorial designs based on central composite and Box–Behnken designs were employed to assess the impact of macronutrient composition on bioconversion efficiency. The results demonstrated significant main, interaction, and quadratic effects on bioconversion efficiency. Validation with real-life substrates of varied protein content, including indigestible feather meal, affirmed the predictive model’s accuracy after accounting for protein source digestibility. This research underscores the importance of optimal hydration and macronutrient composition in enhancing BSFL growth and bioconversion efficiency. Full article

A future unmanned system needs the ability to perceive, decide and control in an open dynamic environment. In order to fulfill this requirement, it needs to construct a method with a universal environmental perception ability. Moreover, this perceptual process needs to be interpretable and understandable, so that future interactions between unmanned systems and humans can be unimpeded. However, current mainstream DNN (deep learning neural network)-based AI (artificial intelligence) is a ‘black box’. We cannot interpret or understand how the decision is made by these AIs. An SNN (spiking neural network), which is more similar to a biological brain than a DNN, has the potential to implement interpretable or understandable AI. In this work, we propose a neuron group-based structural learning method for an SNN to better capture the spatial and temporal information from the external environment, and propose a time-slicing scheme to better interpret the spatial and temporal information of responses generated by an SNN. Results show that our method indeed helps to enhance the environment perception ability of the SNN, and possesses a certain degree of robustness, enhancing the potential to build an interpretable or understandable AI in the future. Full article

Background: Artificial intelligence (AI) has recently made unprecedented contributions in every walk of life, but it has not been able to work its way into diagnostic medicine and standard clinical practice yet. Although data scientists, researchers, and medical experts have been working in the direction of designing and developing computer aided diagnosis (CAD) tools to serve as assistants to doctors, their large-scale adoption and integration into the healthcare system still seems far-fetched. Diagnostic radiology is no exception. Imagining techniques like magnetic resonance imaging (MRI), computed tomography (CT), and positron emission tomography (PET) scans have been widely and very effectively employed by radiologists and neurologists for the differential diagnoses of neurological disorders for decades, yet no AI-powered systems to analyze such scans have been incorporated into the standard operating procedures of healthcare systems. Why? It is absolutely understandable that in diagnostic medicine, precious human lives are on the line, and hence there is no room even for the tiniest of mistakes. Nevertheless, with the advent of explainable artificial intelligence (XAI), the old-school black boxes of deep learning (DL) systems have been unraveled. Would XAI be the turning point for medical experts to finally embrace AI in diagnostic radiology? This review is a humble endeavor to find the answers to these questions. Methods: In this review, we present the journey and contributions of AI in developing systems to recognize, preprocess, and analyze brain MRI scans for differential diagnoses of various neurological disorders, with special emphasis on CAD systems embedded with explainability. A comprehensive review of the literature from 2017 to 2024 was conducted using host databases. We also present medical domain experts’ opinions and summarize the challenges up ahead that need to be addressed in order to fully exploit the tremendous potential of XAI in its application to medical diagnostics and serve humanity. Results: Forty-seven studies were summarized and tabulated with information about the XAI technology and datasets employed, along with performance accuracies. The strengths and weaknesses of the studies have also been discussed. In addition, the opinions of seven medical experts from around the world have been presented to guide engineers and data scientists in developing such CAD tools. Conclusions: Current CAD research was observed to be focused on the enhancement of the performance accuracies of the DL regimens, with less attention being paid to the authenticity and usefulness of explanations. A shortage of ground truth data for explainability was also observed. Visual explanation methods were found to dominate; however, they might not be enough, and more thorough and human professor-like explanations would be required to build the trust of healthcare professionals. Special attention to these factors along with the legal, ethical, safety, and security issues can bridge the current gap between XAI and routine clinical practice. Full article

As lithium-ion batteries become increasingly popular worldwide, accurately determining their capacity is crucial for various devices that rely on them. Numerous data-driven methods have been applied to evaluate battery-related parameters. In the application of these methods, input features play a critical role. Most researchers often use the same input features to compare the performance of various neural network models. However, because most models are regarded as black-box models, different methods may show different dependencies on specific features given the inherent differences in their internal structures. And the corresponding optimal inputs of different neural network models should be different. Therefore, comparing the differences in optimized input features for different neural networks is essential. This paper extracts 11 types of lithium battery-related health features, and experiments are conducted on two traditional machine learning networks and three advanced deep learning networks in three aspects of input differences. The experiment aims to systematically evaluate how changes in health feature types, dimensions, and data volume affect the performance of different methods and find the optimal input for each method. The results demonstrate that each network has its own optimal input in the aspects of health feature types, dimensions, and data volume. Moreover, under the premise of obtaining more accurate prediction accuracy, different networks have different requirements for input data. Therefore, in the process of using different types of neural networks for battery capacity prediction, it is very important to determine the type, dimension, and number of input health features according to the structure, category, and actual application requirements of the network. Different inputs will lead to larger differences in results. The optimization degree of mean absolute error (MAE) can be improved by 10–50%, and other indicators can also be optimized to varying degrees. Therefore, it is very important to optimize the network in a targeted manner. Full article

This study investigated the performance optimization of nickel-cobalt (Ni-Co)-coated absorber panels in solar flat plate collectors (SFPCs) using response surface methodology for sustainable operation and optimized performance. Ni-Co coatings, applied through an electroplating process, represent a novel approach by offering superior thermal conductivity, durability, and environmental benefits compared to conventional black chrome coatings, addressing critical concerns related to ecological impact and long-term reliability. Experiments were conducted to evaluate the thermal efficiency of Ni-Co-coated panels with and without reflectors under varying flow rates, collector angles, and reflector angles. The thermal efficiency was calculated based on the inlet and outlet water temperatures, solar radiation intensity, and panel area. The results showed that the SFPC achieved average efficiencies of 50.9% without reflectors and 59.0% with reflectors, demonstrating the effectiveness of the coatings in enhancing solar energy absorption and heat transfer. A validated quadratic regression model (R² = 0.9941) predicted efficiency based on the process variables, revealing significant individual and interaction effects. Optimization using the Box–Behnken design identified the optimal parameter settings for maximum efficiency: a flow rate of 1.32 L/min, collector angle of 46.91°, and reflector angle of 42.34°, yielding a predicted efficiency of 79.2%. These findings highlight the potential of Ni-Co coatings and reflectors for enhancing SFPC performance and provide valuable insights into the sustainable operation of solar thermal systems. Furthermore, the introduction of Ni-Co coatings offers a sustainable alternative to black chrome, reducing environmental risks while enhancing efficiency, thereby contributing to the advancement of renewable energy technologies. Full article

To address the challenges of black-box video adversarial attacks, such as excessive query times and suboptimal attack performance due to the lack of result feedback during the attack process, we propose a reinforcement learning-based sparse adversarial attack method called RLVS. This approach leverages reinforcement learning to identify key frames for efficient gradient estimation, significantly reducing the number of queries. First, a self-attention network is integrated into the agent policy network to enable more precise selection of key frames. Second, designed reward functions allow the agent to continuously adapt to the sparse key frames by querying the black-box threat model and receiving feedback on attack outcomes. Lastly, gradient estimation is applied solely to the selected key frames, estimating only the gradient sign rather than the full gradient, further enhancing attack efficiency. We conducted experiments on two video recognition models using three popular action datasets. The experimental results demonstrate that our method outperforms other black-box video attack methods in terms of attack efficiency and effectiveness, achieving higher fooling rates with fewer queries and minimal perturbations. Full article

In this paper, we address the issues of the explainability of reinforcement learning-based machine learning agents trained with Proximal Policy Optimization (PPO) that utilizes visual sensor data. We propose an algorithm that allows an effective and intuitive approximation of the PPO-trained neural network (NN). We conduct several experiments to confirm our method’s effectiveness. Our proposed method works well for scenarios where semantic clustering of the scene is possible. Our approach is based on the solid theoretical foundation of Gradient-weighted Class Activation Mapping (GradCAM) and Classification and Regression Tree with additional proxy geometry heuristics. It excels in the explanation process in a virtual simulation system based on a video system with relatively low resolution. Depending on the convolutional feature extractor of the PPO-trained neural network, our method obtains 0.945 to 0.968 accuracy of approximation of the black-box model. The proposed method has important application aspects. Through its use, it is possible to estimate the causes of specific decisions made by the neural network due to the current state of the observed environment. This estimation makes it possible to determine whether the network makes decisions as expected (decision-making is related to the model’s observation of objects belonging to different semantic classes in the environment) and to detect unexpected, seemingly chaotic behavior that might be, for example, the result of data bias, bad design of the reward function or insufficient generalization abilities of the model. We publish all source codes so our experiments can be reproduced. Full article