Artificial Intelligence Machine Learning Research Articles

Sustainable application of ML/AI methods in DCW understanding & screening

Different composition waterflooding (DCW) is an enhanced oil recovery method based on altering the content of the injected water. There are multiple parameters of rock, brine and oil that play a role in impacting the success of the DCW. In such complex processes AI can be a tool efficient in investigating the process for screening and decision making through classification and regression. In the oil and gas industry with the burgeoning amount of data AI/ML are becoming more the necessity due to its efficiency in handling such large data and processing it to convert data into information and into knowledge for efficient decision making. Through this work, a review of the sustainable modus operandi of Artificial Intelligence (AI) Tools and Machine Learning (ML) is performed. The comprehensive review covers the applications of AI in Oil & Gas and EOR and more particularly DCW. Also a look at the future challenges and way forward to make this AI/ML as sustainable practice for the success of projects and improved efficiency in the O&G industry is discussed.

Application of Artificial Intelligence in Oncology Nursing: A Scoping Review.

Artificial intelligence (AI) has been increasingly used in healthcare during the last decade, and recent applications in oncology nursing have shown great potential in improving care for patients with cancer. It is timely to comprehensively synthesize knowledge about the progress of AI technologies in oncology nursing. The aims of this study were to synthesize and evaluate the existing evidence of AI technologies applied in oncology nursing. A scoping review was conducted based on the methodological framework proposed by Arksey and O'Malley and later improved by the Joanna Briggs Institute. Six English databases and 3 Chinese databases were searched dating from January 2010 to November 2022. A total of 28 articles were included in this review-26 in English and 2 in Chinese. Half of the studies used a descriptive design (level VI). The most widely used AI technologies were hybrid AI methods (28.6%) and machine learning (25.0%), which were primarily used for risk identification/prediction (28.6%). Almost half of the studies (46.4%) explored developmental stages of AI technologies. Ethical concerns were rarely addressed. The applicability and prospect of AI in oncology nursing are promising, although there is a lack of evidence on the efficacy of these technologies in practice. More randomized controlled trials in real-life oncology nursing settings are still needed. This scoping review presents comprehensive findings for consideration of translation into practice and may provide guidance for future AI education, research, and clinical implementation in oncology nursing.

English

Drugs for medical purpose aims to save a person’s life and improve the quality of life. They enhance one’s mental or physical fitness and treat disease. However, drugs might cause adverse reactions or even unexpected effects on patients, known as Adverse Drug Reactions (ADRs). For preventing ADRs, drug trials were carried out to evaluate and analyze the potential danger in the procedure of drug development. Deep learning (DL) is a type of Machine Learning (ML) in Artificial intelligence (AI) that has developed as a highly effective and promising method that can interrogate and combine diverse biological data kinds to produce new hypotheses. DL is extensively used in drug repurposing and discovery, but its application in ADR prediction employing gene expression data is limited. Therefore, this study introduces a Deer Hunting Optimization Driven Deep Learning Model for Robust Adverse Drug Reaction Recognition and Classification (DHODL-ADRRC) technique. The DHODL-ADRRC technique involves diverse phases of data preprocessing to normalize the data. In addition, the Binary Fire Hawks Optimization (BFHO) based feature selection (FS) method is used to elect an optimum set of features. Moreover, the Attention-based Convolutional Bidirectional Long Short-Term Memory (ACBLSTM) algorithm is utilized for ADR identification. Furthermore, the DHO model is exploited for the tuning process adjusting of the ACBLSTM model, advancing the classification outputs. The simulation output of the DHODL-ADRRC algorithm was investigated on the ADR dataset, and the extensive outputs highlighted the advanced achievement of the DHODL-ADRRC method over other current methods in terms of different measures.

Identifying (electrophysiological) predictors of atrial fibrillation recurrences after catheter ablation: first results from the ISOLATION registry

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – EU funding. Main funding source(s): a.o. The European Union (ITN Network Personalize AF: Personalized Therapies for Atrial Fibrillation: a translational network, grant number 860974; MAESTRIA: Machine Learning Artificial Intelligence Early Detection Stroke Atrial Fibrillation, grant number 965286). Background/Introduction Noninvasive quantification of electrophysiological properties of the atria is increasingly used for outcome prediction in atrial fibrillation (AF) patients. Data on the added value of ECG parameters to predict AF recurrences in patients undergoing AF ablation are scarce. Purpose To investigate the predictive performance of clinical and echocardiographic characteristics as well as features extracted from extended surface electrocardiogram (extECG) recordings for AF recurrences in patients after AF ablation. Methods ISOLATION is a prospective, observational, multicenter registry of patients scheduled for AF catheter ablation, including cryoballoon and radiofrequency ablation. 178 patients (66.4%) patients used anti arrhythmic drugs at baseline. Clinical characteristics and results of routine blood tests were collected. In addition, an extECG was recorded (12 additional leads, 5 minutes recording) before ablation to analyze P-wave characteristics. The main study endpoint was ablation success after 12 months, defined as freedom from any episode of documented atrial arrhythmia after the blanking period recorded on holter or ECG. Results Currently, 627 of 650 scheduled patients have been included. In total, 268 patients completed the 1-year follow-up. 190 patients (70.9%) had paroxysmal AF and 75 patients (28.0%) had persistent AF. Median age was 65±9 years, 38.1% were female. 46 patients (17.2%) had a recurrence of AF after 12 months [figure 1]. Univariate analysis showed that higher age (64 years vs. 69 years, p=0.023, Mann-Whitney U-test), lower P-wave mean area (4.402 mV vs. 3.567 mV, p=0.034) and lower P-wave mean amplitude (0.087 mV vs. 0.081 mV, p=0.033), were associated with a higher recurrence rate of AF [figure 2]. However, multivariate analysis showed no significant independent prognostic factors so far. Conclusion Our first results show that patients with AF recurrences after catheter ablation are older, show lower P-wave mean area and lower P-wave mean amplitude. We need to wait for the completion of follow-up in more patients, before we can determine the predictive value of P-wave parameters derived from extECGs for the prediction of AF recurrence and to identify predictive factors with more subtle effects on risk for AF recurrences.

Validation of Artificial Intelligence Containing Products Across the Regulated Healthcare Industries.

The introduction of artificial intelligence / machine learning (AI/ML) products to the regulated fields of pharmaceutical research and development (R&D) and drug manufacture, and medical devices (MD) and in vitro diagnostics (IVD), poses new regulatory problems: a lack of a common terminology and understanding leads to confusion, delays and product failures. Validation as a key step in product development, common to each of these sectors including computerized systems and AI/ML development, offers an opportune point of comparison for aligning people and processes for cross-sectoral product development. A comparative approach, built upon workshops and a subsequent written sequence of exchanges, is summarized in a look-up table suitable for mixed-teams work. 1. A bottom-up, definitions led, approach which leads to a distinction between broad vs narrow validation, and their relationship to regulatory regimes. 2. Common basis introduction to the primary methodologies for software validation, including AI-containing software validation. 3. Pharmaceutical drug development and MD/IVD-specific perspectives on compliant AI software development, as a basis for collaboration. Alignment of the terms and methodologies used in validation of software products containing artificial intelligence/machine learning (AI/ML) components across the regulated industries of human health is a vital first step in streamlining processes and improving workflows.

Developmental milestones as ACNS turns one year old

Developmental milestones as ACNS turns one year old

An Inflection Point in Cancer Protein Biomarkers: What was and What's Next

Biomarkers remain the highest value proposition in cancer medicine today—especially protein biomarkers. Despite decades of evolving regulatory frameworks to facilitate the review of emerging technologies, biomarkers have been mostly about promise with very little to show for improvements in human health. Cancer is an emergent property of a complex system, and deconvoluting the integrative and dynamic nature of the overall system through biomarkers is a daunting proposition. The last 2 decades have seen an explosion of multiomics profiling and a range of advanced technologies for precision medicine, including the emergence of liquid biopsy, exciting advances in single-cell analysis, artificial intelligence (machine and deep learning) for data analysis, and many other advanced technologies that promise to transform biomarker discovery. Combining multiple omics modalities to acquire a more comprehensive landscape of the disease state, we are increasingly developing biomarkers to support therapy selection and patient monitoring. Furthering precision medicine, especially in oncology, necessitates moving away from the lens of reductionist thinking toward viewing and understanding that complex diseases are, in fact, complex adaptive systems. As such, we believe it is necessary to redefine biomarkers as representations of biological system states at different hierarchical levels of biological order. This definition could include traditional molecular, histologic, radiographic, or physiological characteristics, as well as emerging classes of digital markers and complex algorithms. To succeed in the future, we must move past purely observational individual studies and instead start building a mechanistic framework to enable integrative analysis of new studies within the context of prior studies. Identifying information in complex systems and applying theoretical constructs, such as information theory, to study cancer as a disease of dysregulated communication could prove to be “game changing” for the clinical outcome of cancer patients.

A Multiple Response Prediction Model for Dissimilar AA-5083 and AA-6061 Friction Stir Welding Using a Combination of AMIS and Machine Learning

This study presents a methodology that combines artificial multiple intelligence systems (AMISs) and machine learning to forecast the ultimate tensile strength (UTS), maximum hardness (MH), and heat input (HI) of AA-5083 and AA-6061 friction stir welding. The machine learning model integrates two machine learning methods, Gaussian process regression (GPR) and a support vector machine (SVM), into a single model, and then uses the AMIS as the decision fusion strategy to merge SVM and GPR. The generated model was utilized to anticipate three objectives based on seven controlled/input parameters. These parameters were: tool tilt angle, rotating speed, travel speed, shoulder diameter, pin geometry, type of reinforcing particles, and tool pin movement mechanism. The effectiveness of the model was evaluated using a two-experiment framework. In the first experiment, we used two newly produced datasets, (1) the 7PI-V1 dataset and (2) the 7PI-V2 dataset, and compared the results with state-of-the-art approaches. The second experiment used existing datasets from the literature with varying base materials and parameters. The computational results revealed that the proposed method produced more accurate prediction results than the previous methods. For all datasets, the proposed strategy outperformed existing methods and state-of-the-art processes by an average of 1.35% to 6.78%.

O127 Application of machine learning approaches in the diagnosis and management of colorectal cancer

Abstract Introduction Colorectal cancer (CRC) is one of the most common cancers worldwide. In recent years, Artificial Intelligence (AI) and Machine Learning (ML) algorithms have gained traction in CRC research for their ability to efficiently analyse large volumes of variables and offer objective data driven models to optimise clinical decision making. The present review provides a contemporary summary of the current application of these approaches in CRC. Methods A literature search was performed on AI models and ML learning in CRC and relevant articles were scrutinized. Results AI aided detection systems were successful in identifying CRC, and able to outperform endoscopists in detecting adenomas (detection rate 0.29 vs. 0.20, p < 0.001). ML models based on radiomics showed promise in identifying CRC metastasis and even outperforming national guideline-based models (AUC 0.83 vs. 0.73, p<0.01) in predicting risk of T1 CRC metastasis to lymph nodes. ML models were also able to identify novel genomic targets, predict response to oncological therapy, and predict 5-year disease free survival in stage II and III CRC (AUC of 0.698). In the operative context, AI can be trained to identify surgical steps and operative actions with accuracy scores of over 80% as well as. Conclusion AI modelling takes advantage of increasing digital medical data keeping to allow rapid high-volume data analysis which has found application in the diagnosis, staging, treatment, and prognosis of CRC and may be useful in the development of intra-operative guidance systems to optimize surgical approaches and training.

Crop yield prediction via explainable AI and interpretable machine learning: Dangers of black box models for evaluating climate change impacts on crop yield

Statistical crop modeling is pivotal for understanding climate impacts on crop yields. Choices of models matter: Linear regression is interpretable but limited in predictive power; machine learning predicts well but often remains a black box. To develop explainable artificial intelligence (AI) for exploring historical crop yield data and predicting crop yield, here we reported a Bayesian ensemble model (BM) that is interpretable with great explanatory and predictive power. BM embraces many competitive models via Bayesian model averaging, fits complex functions, and quantifies model uncertainty. Long-term crop yields are driven by both climate and technology; the common practice of first detrending and then analyzing the detrended data has an incorrigible bias. Therefore, BM was also aimed at decomposing historical yield data to jointly estimate technological trends and climate effects on crop yield. We compared BM with ElasticNet, Neural Network, MARS, SVM, Random Forests, and XGBoost. BM excelled at both predicting and explaining. When tested on synthetic data, BM was the only method unveiling the true relationships: BM has stronger interpretability; other methods predicted well but for wrong reasons. When tested on maize yield data in Ohio, BM detected two technological shifts, attributable to hybrid corn adoption in the 1940′s and the technological slowing-down in the 1970′s: No other methods detected such changepoints. BM derived nonlinear asymmetric crop responses to climate and non-negligible temperature-precipitation interacting effects, with patterns consistent with theoretical or experimental evidence. Extrapolation of all the models for future yield prediction was highly uncertain, but BM provided more reliable predictions under novel climate whereas Random Forests and XGBoost proved unsuitable for extrapolation. Overall, BM provided new insights unattainable by the existing black-box methods. We caution against blind use of black-box machine learning for statistical crop modeling and call for more efforts to apply interpretable machine learning for mechanistic understandings of crop-climate interactions.

Artificial intelligence analysis of electroencephalogram and evoked potential in patients with depression based on machine learning

AbstractWith the continuous improvement of people's mental pressure and life pace, people's study and life pressure would increase, leading to the increase of people's depression. Depression is a mental illness, a chronic mental illness that is inconsistent with the patient's physical condition. In recent years, as people know more and more about depression, and they have more and more research on depression, many research scholars have provided new ideas for the treatment of depression, and this paper takes this as the research direction and research basis. This paper introduces the background of EEG (electroencephalogram, EEG) and evoked potential and artificial intelligence (Artificial intelligence, AI) methods, and then analyzes the patients with depression based on AI, and summarizes the application of electronics. The concept analysis of depression, EEG and evoked potential is put forward. At the end of the article, the application of machine learning in depression is studied. At the same time, with the continuous development of machine learning in artificial intelligence, the EEG and evoked potential related work in patients with depression are also facing new opportunities and challenges.

Biosensors for point-of-care testing and personalized monitoring of gastrointestinal microbiota.

The gastrointestinal (GI) microbiota is essential in maintaining human health. Alteration of the GI microbiota or gut microbiota (GM) from homeostasis (i.e., dysbiosis) is associated with several communicable and non-communicable diseases. Thus, it is crucial to constantly monitor the GM composition and host-microbe interactions in the GI tract since they could provide vital health information and indicate possible predispositions to various diseases. Pathogens in the GI tract must be detected early to prevent dysbiosis and related diseases. Similarly, the consumed beneficial microbial strains (i.e., probiotics) also require real-time monitoring to quantify the actual number of their colony-forming units within the GI tract. Unfortunately, due to the inherent limitations associated with the conventional methods, routine monitoring of one's GM health is not attainable till date. In this context, miniaturized diagnostic devices such as biosensors could provide alternative and rapid detection methods by offering robust, affordable, portable, convenient, and reliable technology. Though biosensors for GM are still at a relatively preliminary stage, they can potentially transform clinical diagnosis in the near future. In this mini-review, we have discussed the significance and recent advancements of biosensors in monitoring GM. Finally, the progresses on future biosensing techniques such as lab-on-chip, smart materials, ingestible capsules, wearable devices, and fusion of machine learning/artificial intelligence (ML/AI) have also been highlighted.

The Role of Radiomics in Fibrosis Crohn's Disease: A Review.

Inflammatory bowel disease (IBD) is a global health concern that has been on the rise in recent years. In addition, imaging is the established method of care for detecting, diagnosing, planning treatment, and monitoring the progression of IBD. While conventional imaging techniques are limited in their ability to provide comprehensive information, cross-sectional imaging plays a crucial role in the clinical management of IBD. However, accurately characterizing, detecting, and monitoring fibrosis in Crohn's disease remains a challenging task for clinicians. Recent advances in artificial intelligence technology, machine learning, computational power, and radiomic emergence have enabled the automated evaluation of medical images to generate prognostic biomarkers and quantitative diagnostics. Radiomics analysis can be achieved via deep learning algorithms or by extracting handcrafted radiomics features. As radiomic features capture pathophysiological and biological data, these quantitative radiomic features have been shown to offer accurate and rapid non-invasive tools for IBD diagnostics, treatment response monitoring, and prognosis. For these reasons, the present review aims to provide a comprehensive review of the emerging radiomics methods in intestinal fibrosis research that are highlighted and discussed in terms of challenges and advantages.

Explainable AI for SE: Challenges and Future Directions

In recent years, artificial intelligence/machine learning (AI/ML) have been widely used in software engineering (SE) to improve developer productivity, software quality, and decision making. This includes well-known tools for code completion (for example, GitHub’s Copilot) but also code search; automated task recommendation; automated developer recommendation; automated defect/vulnerability/malware prediction, detection, localization, and repair; and many other purposes.

Cooperative Spectrum Sensing in Cognitive Radio Networks: A Systematic Review

Background:Spectrum is the backbone for wireless communications including internet services. Now days, the business of industries providing wired communication is constant while the business of industries dealing with wireless communications is growing very fast. There is large demand of radio spectrum for new wireless multimedia services. Although the present fixed spectrum allotment schemes do not cause any interference between users, but this fixed scheme of spectrum allocation do not allow accommodating the spectrum required for new wireless services. Cognitive radio (CR) relies on spectrum sensing to discover available frequency bands so that the spectrum can be used to its full potential, thus avoiding interference to the primary users (PU).Objectives:The purpose of this work is to present an in-depth overview of traditional as well as advanced artificial intelligence and machine learning based cooperative spectrum sensing (CSS) in cognitive radio networks.Method:Using the principles of artificial intelligence (AI), systems are able to solve issues by mimicking the function of human brains. Moreover, since its inception, machine learning has demonstrated that it is capable of solving a wide range of computational issues. Recent advancements in artificial intelligence techniques and machine learning (ML) have made it an emergent technology in spectrum sensing.Result:The result shows that more than 80% papers are on traditional spectrum sensing while less than 20% deals with artificial intelligence and machine learning approaches. More than 75% papers address the limitation of local spectrum sensing. The study presents the various methods implemented in the spectrum sensing along with merits and challenges.Conclusion:Spectrum sensing techniques are hampered by a variety of issues, including fading, shadowing, and receiver unpredictability. Challenges, benefits, drawbacks, and scope of cooperative sensing are examined and summarized. With this survey article, academics may clearly know the numerous conventional artificial intelligence and machine learning methodologies used and can connect sharp audiences to contemporary research done in cognitive radio networks, which is now underway.

Toward fairness in artificial intelligence for medical image analysis: identification and mitigation of potential biases in the roadmap from data collection to model deployment.

To recognize and address various sources of bias essential for algorithmic fairness and trustworthiness and to contribute to a just and equitable deployment of AI in medical imaging, there is an increasing interest in developing medical imaging-based machine learning methods, also known as medical imaging artificial intelligence (AI), for the detection, diagnosis, prognosis, and risk assessment of disease with the goal of clinical implementation. These tools are intended to help improve traditional human decision-making in medical imaging. However, biases introduced in the steps toward clinical deployment may impede their intended function, potentially exacerbating inequities. Specifically, medical imaging AI can propagate or amplify biases introduced in the many steps from model inception to deployment, resulting in a systematic difference in the treatment of different groups. Our multi-institutional team included medical physicists, medical imaging artificial intelligence/machine learning (AI/ML) researchers, experts in AI/ML bias, statisticians, physicians, and scientists from regulatory bodies. We identified sources of bias in AI/ML, mitigation strategies for these biases, and developed recommendations for best practices in medical imaging AI/ML development. Five main steps along the roadmap of medical imaging AI/ML were identified: (1)data collection, (2)data preparation and annotation, (3)model development, (4)model evaluation, and (5)model deployment. Within these steps, or bias categories, we identified 29 sources of potential bias, many of which can impact multiple steps, as well as mitigation strategies. Our findings provide a valuable resource to researchers, clinicians, and the public at large.

Augmenting Community Diagnosis of Safe Ear Disease Through Tele-Myringoscopy with Borescope Using AIML Tecniques.

This study utilized AIML (artificial intelligence & machine learning) techniques to analyze 115 images of central perforation of tympanic membrane obtained from Telemyringoscopy through Borescope in order to establish a facilitation-model for the community ear diagnosis. The Modified VGG19 with batch normalization revealed the highest training accuracy of 85 as compared to other CNN techniques. The training accuracy started to saturate around mid-70% and the Test accuracy was around 50%. Although AIML did not reveal a high predictive value, its potential based on our observations cannot be underestimated considering many limitations (sample size, image-quality, associated pathologies, illumination-factor) in this study. Such limitations if resolved may revolutionize community ear care through a better cost effective tele-myringoscopy with innovations in AIML/ telemedicine.

Tear nanoDSF Denaturation Profile Is Predictive of Glaucoma.

Primary open-angle glaucoma (POAG) is a frequent blindness-causing neurodegenerative disorder characterized by optic nerve and retinal ganglion cell damage most commonly due to a chronic increase in intraocular pressure. The preservation of visual function in patients critically depends on the timeliness of detection and treatment of the disease, which is challenging due to its asymptomatic course at early stages and lack of objective diagnostic approaches. Recent studies revealed that the pathophysiology of glaucoma includes complex metabolomic and proteomic alterations in the eye liquids, including tear fluid (TF). Although TF can be collected by a non-invasive procedure and may serve as a source of the appropriate biomarkers, its multi-omics analysis is technically sophisticated and unsuitable for clinical practice. In this study, we tested a novel concept of glaucoma diagnostics based on the rapid high-performance analysis of the TF proteome by differential scanning fluorimetry (nanoDSF). An examination of the thermal denaturation of TF proteins in a cohort of 311 ophthalmic patients revealed typical profiles, with two peaks exhibiting characteristic shifts in POAG. Clustering of the profiles according to peaks maxima allowed us to identify glaucoma in 70% of cases, while the employment of artificial intelligence (machine learning) algorithms reduced the amount of false-positive diagnoses to 13.5%. The POAG-associated alterations in the core TF proteins included an increase in the concentration of serum albumin, accompanied by a decrease in lysozyme C, lipocalin-1, and lactotransferrin contents. Unexpectedly, these changes were not the only factor affecting the observed denaturation profile shifts, which considerably depended on the presence of low-molecular-weight ligands of tear proteins, such as fatty acids and iron. Overall, we recognized the TF denaturation profile as a novel biomarker of glaucoma, which integrates proteomic, lipidomic, and metallomic alterations in tears, and monitoring of which could be adapted for rapid non-invasive screening of the disease in a clinical setting.

Machine Learning and Deep Learning for the Built Heritage Analysis: Laser Scanning and UAV-Based Surveying Applications on a Complex Spatial Grid Structure

The reconstruction of 3D geometries starting from reality-based data is challenging and time-consuming due to the difficulties involved in modeling existing structures and the complex nature of built heritage. This paper presents a methodological approach for the automated segmentation and classification of surveying outputs to improve the interpretation and building information modeling from laser scanning and photogrammetric data. The research focused on the surveying of reticular, space grid structures of the late 19th–20th–21st centuries, as part of our architectural heritage, which might require monitoring maintenance activities, and relied on artificial intelligence (machine learning and deep learning) for: (i) the classification of 3D architectural components at multiple levels of detail and (ii) automated masking in standard photogrammetric processing. Focusing on the case study of the grid structure in steel named La Vela in Bologna, the work raises many critical issues in space grid structures in terms of data accuracy, geometric and spatial complexity, semantic classification, and component recognition.

A Systematic Survey of Approaches Used in Computer Music Generation

Music generation using Artificial Intelligence/ Machine learning (AI/ML) is a rapidly growing research area that aims to create algorithms capable of producing musical compositions autonomously. This paper provides an overview of the latest developments in the area of AI/ML-based music generation. The deep learning models have demonstrated promise in music generation, including Recurrent Neural Networks (RNN), long short-term memory (LSTM), WaveNet, Generative Pre-trained Transformer (GPT), and Genetic Algorithms (GA). A review of the existing research work on the creation of different music genres such as classical, jazz, and pop music using these models is covered. The challenges in music generation are covered, including the difficulty in evaluating the quality and creativity of the generated music along with the social issues surrounding the ownership and copyright of the created music. The development of AI/ML-based music generation systems has made significant advances in recent years. The ability to create an infinite number of original musical compositions is one of the main benefits of employing AI/ML algorithms for music production. Content producers can utilize these technologies to rapidly and easily create original soundtracks for their projects, which can be of special utility to them. Moreover, the advancement in the music generating systems could make it possible for more people to contribute to the creation of music, regardless of their level of technical or musical proficiency. As technology develops, it might also create previously unimagined opportunities for new forms of musical expression. The applications include helping musicians with their work and creating music for video games and movies