Deep learning using electrocardiographs

Artificial Intelligence: The Future of Maxillofacial Prognosis and Diagnosis?

Introduction: The diagnosis of myocardial infarction (MI) needs to be swift and accurate, but definitively diagnosing it based on the first test encountered in clinical practice, the electrocardiogram (ECG), is not an easy task. The purpose of the study was to develop a deep learning (DL) algorithm using multitask learning method to differentiate patients experiencing MI from those without coronary artery disease using image-based ECG data. Methods: A DL model was developed based on 11,227 ECG images. We developed a new ECG interpretation algorithm through signal-guided multitask learning, building on a previously published single-task algorithm. The utility of this model was evaluated by testing 51 physicians in interpreting ECGs with and without the assistance of the DL model. Results: The proposed model demonstrated superior performance, achieving 90.56% accuracy, 83.82% sensitivity, 93.02% specificity, 81.44% precision, and an F1 score of 82.61% in discriminating MI ECG. Overall, the median accuracy of ECG interpretation improved from 62% to 68% with the DL algorithm. Trainees and specialists in internal medicine experienced significant accuracy increases (60–66% for trainees, 72–80% for specialists). In the MI group, NSTEMI accuracy was notably lower than STEMI (33% vs. 80%, p < 0.001), but the DL algorithm improved interpretative capabilities in both NSTEMI and STEMI. Conclusions: Signal-guided multitask DL algorithm demonstrated superior performance compared with previous single-task algorithm. The DL algorithm supports the physicians’ decision discriminating MI ECGs from non-MI ECGs. The improvement was consistent in subgroups of STEMI and NSTEMI.

Background A universal screening strategy for atrial fibrillation (AF) is not established despite its rising incidence. Emerging consumer and medical-grade artificial intelligence (AI) electrocardiography (ECG) devices hold promise for AF screening. Latest research suggests that AI can extract clinically relevant features from sinus rhythm (SR)-ECG, potentially identifying patients nearing an AF diagnosis. This positions AI as a novel tool for proactive AF screening that could enhance early detection and intervention. Purpose To assess the effectiveness of AI algorithms in detecting patients approaching an AF diagnosis using SR-ECG and analyse the factors influencing their performance. Methods A systematic search was conducted in PubMed, EMBASE, Scopus, Web of Science, Google Scholar, ClinicalTrials.gov, European Union Clinical Trials Register, and International Clinical Trials Registry Platform. To enhance inclusivity, references of included studies were also included. Eligible studies were published from January 2014 to May 2024, used AI to detect AF patients from SR-ECGs in patients without prior AF history, tested the model on unseen data, and confirmed AF via traditional methods within one year maximum. Summarized performance metrics were compared using the Mann-Whitney U test with subgroup analyses by AI type and AF confirmation timeframe. Study quality was assessed with QUADAS-2. Results The search yielded 3568 articles, of which 14 studies describing 33 AI models were included. A total of 4,182,953 ECGs from 1,459,653 patients were analysed by the models. Median model performance for AF detection using SR-ECG was: accuracy 58.0%, sensitivity 62.0%, specificity 57.8%, precision 52.0%, and area under the curve (AUC) 0.740. Classical machine learning was used by 12 (36.4%) models and deep learning by 21 models (63.6%), of which 13 used a convolutional neural network. Deep learning outperformed traditional machine learning in accuracy (83.0% vs 57.0%; p=0.027), sensitivity (72.6% vs 54.5%; p=0.009), and AUC (0.830 vs 0.610; p&amp;lt;0.001). The confirmatory AF event was captured within a maximum of 31 days by 13 (29.4%) models, and within a 31-day to one-year period by 20 (60.6%) models. Models using a 31-day confirmation window showed higher accuracy (83.2% vs 56.0%; p=0.001) and AUC (0.851 vs 0.630; p&amp;lt;0.001) than those using a one-year timeframe. A formal meta-analysis was deemed unsuitable due to considerable methodological heterogeneity among studies. Conclusions Deep learning algorithms and shorter confirmation windows demonstrate improved AF identification performance, highlighting timely follow-up. Expanding AI usage through accessible platforms may enhance ECG screening in underserved areas, addressing health disparities. Further research is needed to assess the clinical benefits and cost-effectiveness of such widespread application.Performance by AI typePerformance by confirmation timeframe

Improving the performance of machine learning algorithms to overcome the curse of dimensionality while maintaining interpretability is still a challenging issue for researchers in artificial intelligence. Patch learning (PL), based on the improved adaptive network-based fuzzy inference system (ANFIS) and continuous local optimization for the input domain, is characterized by high accuracy. However, PL can only handle low-dimensional data set regression. Based on the parallel and serial ensembles, two deep patch learning algorithms with embedded adaptive fuzzy systems (DPLFSs) are proposed in this paper. First, using the maximum information coefficient (MIC) and Pearson's correlation coefficients for feature selection, the variables with the least relationship (linear or nonlinear) are excluded. Second, principal component analysis is used to reduce the complexity further of DPLFSs. Meanwhile, fuzzy C-means clustering is used to enhance the interpretability of DPLFSs. Then, an improved PL method is put forward for the training of each sub-fuzzy system in a fashion of bottom-up layer-by-layer, and finally, the structure optimization is performed to significantly improve the interpretability of DPLFSs. Experiments on several benchmark data sets show the advantages of a DPLFS: (1) it can handle medium-scale data sets; (2) it can overcome the curse of dimensionality faced by PL; (3) its precision and generalization are greatly improved; and (4) it can overcome the poor interpretability of deep learning networks. Compared with shallow and deep learning algorithms, DPLFSs have the advantages of interpretability, self-learning, and high precision. DPLFS1 is superior for medium-scale data; DPLFS2 is more efficient and effective for high-dimensional problems, has a faster convergence, and is more interpretable.

Real-World Surveillance of FDA-Cleared Artificial Intelligence Models: Rationale and Logistics.

The present study reports the development of a deep learning artificial intelligence (AI) model for predicting the thermal performance of evaporative cooling systems, which are widely used for thermal comfort in different applications. The existing, conventional methods for the analysis of evaporation-assisted cooling systems rely on experimental, mathematical, and empirical approaches in order to determine their thermal performance, which limits their applications in diverse and ambient spatiotemporal conditions. The objective of this research was to predict the thermal performance of three evaporation-assisted air-conditioning systems—direct, indirect, and Maisotsenko evaporative cooling systems—by using an AI approach. For this purpose, a deep learning algorithm was developed and lumped hyperparameters were initially chosen. A correlation analysis was performed prior to the development of the AI model in order to identify the input features that could be the most influential for the prediction efficiency. The deep learning algorithm was then optimized to increase the learning rate and predictive accuracy with respect to experimental data by tuning the hyperparameters, such as by manipulating the activation functions, the number of hidden layers, and the neurons in each layer by incorporating optimizers, including Adam and RMsprop. The results confirmed the applicability of the method with an overall value of R2 = 0.987 between the input data and ground-truth data, showing that the most competent model could predict the designated output features (Toutdb, wout, and Eoutair). The suggested method is straightforward and was found to be practical in the evaluation of the thermal performance of deployed air conditioning systems under different conditions. The results supported the hypothesis that the proposed deep learning AI algorithm has the potential to explore the feasibility of the three evaporative cooling systems in dynamic ambient conditions for various agricultural and livestock applications.

Electrocardiography (ECG) has gained popularity in recognizing abnormally quick and slow heart rates. Previously, clinical data was used for heart disease diagnosis. Advance technology led to improvement in deep learning (DL) and optimization algorithms that use image data to classify and predict heart diseases. DL eliminates data pre-processing as in machine learning (ML), to scan and search features that correlate and combine them to enable faster learning. DL requires huge data to perform accurately. DL is expensive concerning hardware requirements and training which increases the overall cost. Hence, grey wolf optimization (GWO) can provide competitive results of improved local optima. This paper presents an advanced approach of index matching to perform analysis in two categories by merging clinical data with ECG for classifying and diagnosing heart diseases. Best features are selected by GWO on merging 12-lead Physikalisch-Technische Bundesanstalt (PTB-XL) ECG data and clinical data. Evaluations using deep neural networks on the merged data set are performed. In comparison, the results show that convolutional neural networks (CNNs) using Visual Geometry Group (VGG-16) outperform all other algorithms. Performance is evaluated with various parameters like accuracy, recall, precision, and area under the receiver operator characteristic curve (AUC). Deep learning-based algorithms have a promising future in ECG analysis regarding quantitative exactness and quality.

Pancreatic neuroendocrine tumors (pNETs) are relatively rare and consist of 2% of the all pancreatic tumors. Although some of pNETs have a benign, nonprogressive course, some may be progressive and result with metastasis. We aimed to estimate the grade of pNETs by using artificial intelligence (AI) via deep learning (DL) algorithms as indexing to cyto/histopathological classification according to the World Health Organization 2017. A total of 803 EUS images were collected from 44 patients who had a cyto/histo-pathologically confirmed diagnosis with EUS fine-needle aspiration or biopsy (FNA/B). First, raw EUS images were prepared for processing by AI via DL algorithms, and convolutional neural networks were utilized to train the machine to predict the grades from EUS images. IBM SPSS 25.0 program was used for statistical analyses. Thirty of the 44 patients (68%) were female, with a median age of 61 (range, 16-80) years. pNETs were mostly located in the pancreatic head: 24 cases (55%). Location was the neck in 3 (7%), body in 10 (22%), and tail in 7 (16%) patients. According to EUS-FNA/B results, 27 patients were grade 1 (G1) (61%); 12, grade 2 (G2) (27%); and 5, grade 3 (G3) (12%). In reference to the performance of AI for predicting the pathological grade, sensitivity was 94.29%; specificity, 97.14%; and accuracy, 96.19%. When the patient groups were subanalyzed as G1, G2, and G3 by the AI model to predict the pathological grade, the accuracy was as follows: for G1, 93.15%; for G2, 91.61%; and for G3, 98.05%. This pilot study suggests that pNET grade prediction can be reliably done on EUS images using AI-based technology.

Existing research shows that cooperative multi-agent deep reinforcement learning (c-MADRL) is vulnerable to adversarial attacks, and c-MADRL is increasingly being applied to safety-critical domains. However, the robustness of c-MADRL against adversarial attacks has not been fully studied. In the setting of c-MADRL, unlike the single-agent scenario, an adversary can attack multiple agents or all agents at each time step, but the attacker needs more computation to generate adversarial examples and will be more easily detected. Therefore, how the attacker chooses one or several agents instead of all agents to attack is a significant issue in the setting of c-MADRL. Aiming to address this issue, this paper proposes a novel adversarial attack approach, which dynamically groups the agents according to relevant features and selects a group to attack based on the group’s contribution to the overall reward, thus effectively reducing the cost and number of attacks, as well as improving attack efficiency and decreasing the chance of attackers being detected. Moreover, we exploit the transferability of adversarial examples to greatly reduce the computational cost of generating adversarial examples. Our method is tested in multi-agent particle environments (MPE) and in StarCraft II. Experimental results demonstrate that our proposed method can effectively degrade the performance of multi-agent deep reinforcement learning algorithms with fewer attacks and lower computational costs.

With intentional feature perturbations to a deep learning model, the adversary generates an adversarial example to deceive the deep learning model. As an adversarial example has recently been considered in the most severe problem of deep learning technology, its defense methods have been actively studied. Such effective defense methods against adversarial examples are categorized into one of the three architectures: (1) model retraining architecture; (2) input transformation architecture; and (3) adversarial example detection architecture. Especially, defense methods using adversarial example detection architecture have been actively studied. This is because defense methods using adversarial example detection architecture do not make wrong decisions for the legitimate input data while others do. In this paper, we note that current defense methods using adversarial example detection architecture can classify the input data into only either a legitimate one or an adversarial one. That is, the current defense methods using adversarial example detection architecture can only detect the adversarial examples and cannot classify the input data into multiple classes of data, i.e., legitimate input data and various types of adversarial examples. To classify the input data into multiple classes of data while increasing the accuracy of the clustering model, we propose an advanced defense method using adversarial example detection architecture, which extracts the key features from the input data and feeds the extracted features into a clustering model. From the experimental results under various application datasets, we show that the proposed method can detect the adversarial examples while classifying the types of adversarial examples. We also show that the accuracy of the proposed method outperforms the accuracy of recent defense methods using adversarial example detection architecture.

Automatic diagnosis based on clinical notes has become a popular research field recently, and many proposed deep learning models have achieved competitive performance in diseases inference. However, previous research reveals that deep learning models are susceptible to negligibly perturbed inputs named adversarial examples, which contradicts with the safety and reliability requirements of the medical domain. To analyze the vulnerability and robustness of current automatic diagnosis models, we investigate in the generation of adversarial text examples. The main challenges for generating adversarial text examples are divided into three parts. First, the word embedding space is discrete, which makes it hard to perturb as small as adversarial image examples generation. Second, previous adversarial example generation methods focus mainly on multi-class classification models, while automatic diagnosis is a multi-label classification task. Third, the semantic and medical meaning of clinical notes are vital in disease inference, and even small perturbations can change them to a large extent. In this paper, we address the three main challenges and propose Clinical-Attacker, a general framework for both white-box and black-box adversarial text examples generation against automatic diagnosis models. Experimental results on MIMIC-III dataset demonstrate that our framework can easily alter the predictions of automatic diagnosis models with the semantic and medical meaning preserved.

Atrial fibrillation (AF) is a common cardiac arrhythmia linked to an increased risk of stroke, heart failure, and mortality, yet its diagnosis remains challenging due to its intermittent and often asymptomatic nature. Traditional methods, such as standard electrocardiography (ECG) and prolonged cardiac monitoring, have limitations in terms of cost, accessibility, and diagnostic yield. Artificial intelligence (AI), particularly machine learning (ML) and deep learning, has emerged as a promising tool for AF detection and prediction by analyzing ECG data with high accuracy. AI models can identify subtle patterns in ECG signals that may indicate AF, even when the arrhythmia is not actively present, improving early diagnosis and risk stratification. Additionally, AI-powered ECG analysis has been integrated into wearable and mobile health devices, expanding screening capabilities beyond clinical settings. While studies have demonstrated AI's effectiveness, challenges such as data bias, model reliability across diverse populations, and regulatory considerations must be addressed before widespread clinical adoption. If these obstacles are overcome, AI has the potential to revolutionize AF management by enabling earlier detection, reducing the need for resource-intensive monitoring, and improving patient outcomes.

Deep learning has gained immense popularity in scientific computing, and its algorithms are widely used in complex problem-solving industries. Every deep learning algorithm use different types of neural networks to perform indented tasks. Deep learning (DL) algorithms have emerged from different machine learning and soft computing methodologies. Since then, a number of deep learning (DL) algorithms have been recently introduced in the scientific community and applied in various application fields. Today, the use of DLs has become indispensable due to their intelligence, effective learning, accuracy and reliability in model creation. However, a comprehensive list of DL algorithms has not yet been presented in the scientific literature. This article lists the most popular DL algorithms and their application areas. Deep learning uses ANN artificial neural networks to perform convoluted calculations on huge amounts of data. It is a type of machine learning based on the structure and function of the human brain. Deep learning algorithms train machines by learning from examples. Industries such as healthcare, e-commerce, entertainment and advertising often use deep learning.

Electrocardiography (ECG) data acquisition, data preprocessing, feature extraction and emotion recognition based on ECG feature classification were effectively implemented. Joy and sad movies were selected and presented to 154 subjects whose ECG data were recorded at the movie presentation time. The automatic location of QRS complex, which is of critical importance for ECG feature extraction by the computer, was performed by non-linear transformation on the first derivative of the ECG. By means of tabu search, the best combination of the ECG features was selected for classification. The ECG data were classified into certain cluster by Fisher projection.

Editorial: "Cyber Security Prevention, Defenses Driven by AI and Mathematical Modelling and Simulation Tools". The rise of the Smart Computation driven by the Artificial Intelligence with the ubiquitous access to Data Computer Communication Technologies and Internet by anyone, from anywhere at any time, has become a working, learning and socializing platform for people worldwide and if often known as Cyberspace. The past century technological advancements in the Data Communications Technologies , Semiconductor Industry (SI) and a Very Large Scale Integration (VLSI) contributed to significant increase of computer systems performance and computational capacity of small portable computers and cell phones that have become available at low cost for people anywhere in the world. Due to relatively low cost, anyone and anywhere in the world may purchase a PC, Laptop, Smart Phone, or Tablet. These smart-computational devices are capable of powerful computation and connecting almost any computer system in the world via Internet 24/7. Given the current unstable geopolitical ecosystem in conjunction with the dynamic development in the artificial intelligence (AI), smart computation and fast Internet with the ubiquitous access to Internet 24/7, the reliable and secure provision of confidentiality, integrity and accessibility of data available via Internet, has become essential for government, business, the industry and individual worldwide. The fast developments in the field of AI gave rise to a new era of smart computation that can easily be utilized by the governments, special groups or very well trained and skilled individual with malicious intend due to their political , economic or personal motivation. Given the rise of next generation of smart cyberattacks, the use of AI and Mathematical Modeling and Simulations Tools has become essential part of cybersecurity tool kit, to make sure that contra cyber measures are most effective and reliable in real-time 24/7. The Research Topic was received with great interest in the scholastic and professional community worldwide. The paper selection was very competitive based on rigorous reviews presenting seven papers. The Research Topic presents a collection of selected peer reviewed papers presenting most recent advances and the state-of-the-art in the field of. Applied AI and Mathematical Modelling in Cyber Security. The papers show evidence that cyberattacks are becoming much more sophisticated, computationally complex, and challenging for the current cyber security mechanisms, which drives the research, innovation, and development of the next generation of AI assisted Cyber Security imbedded with the Mathematical Modelling and Simulations tools. The first paper presents practical application of eXplainable AI (XAI) in cybersecurity while enabling a machine-human interaction. The second paper discusses effective use of AI applied in the cyberattacks prevention utilizing Intrusion Detection and Prevention Systems (IDPS). The third paper presents use of deep learning and machine learning algorithms to identify the malicious malware. The fourth paper discusses the use of encryption and cryptographic algorithms in creating an effective cyber defenses in the Internet of Thing Cyber Infrastructures. The fifth paper presents effective methods to identify possible cyberattacks utilizing the noisy audio evidence. The sixth paper illustrates utilization of simulation based on the dropout mechanisms to identify obfuscated malicious traffic. The seventh paper presents effective applications of Generative AI Influenced Spread, Extended Models and Awareness Spread Models to analyze fake news propagation and enable the intervention strategy assessment. The Research Topic brings to light most recent developments in the field of applied AI and Mathematical Modeling and Simulations in Cyber Security, while presenting current and future challenges presented by ongoing challenges in the field of studies. Given the current dynamic advancements in the field of AI, Smart Computation and AI, we are yet at the very beginning of new era of new generation of ultra-smart and most sophisticated cyberattacks that will require a new ultra-smart cyber defenses to make sure that the future cyberspace will be safe, reliable and beneficial to the betterment of mankind for future generations for many years to come. The Research Topic promotes creation of the AI Inspired Cyber Security Center, bringing together multidisciplinary research teams working together while seeking best solutions and mechanism to make the current and future cyberspace safe, secure and reliable 27/7 in the nation and worldwide. The Topic Editors would like to use this opportunity to express most sincere gratitude to contributors, reviewers, to Dr. Enrique Morillas, Dr. Nicola Zannone, Mr Sean O'Reilly Ms Alice Gooch, and Colleagues at Frontiers in Computer Science for their strong support, kind guidance and good counsel provided during the course of the project.