AI-based Prediction Research Articles

Abstract OT-39-01: Artificial intelligence supporting cancer patients across Europe - the ASCAPE project for breast cancer patients

Abstract Background Many breast cancer patients experience adverse effects of cancer or treatment, which can considerably decrease quality of life (QoL). The current strategy of supporting breast cancer patients does not meet their needs due to the limited personalized-based approach in rehabilitation plan and the lack of healthcare, financial and other resources. ASCAPE (Artificial intelligence Supporting CAncer Patients across Europe) is a collaborative research project involving 15 partners from 7 countries, including academic medical centers, small and medium-sized enterprises, research centers and universities, aiming to leverage the recent advances in Big Data and AI (Artificial Intelligence) to support cancer patients’ QoL and health status. Specifically, ASCAPE aims to provide personalized- and AI-based predictions for QoL issues in breast cancer patients as well as suggest potential interventions to their physicians. Trial design During the first part of the project, large-scale retrospective datasets with breast cancer patients will be analyzed to develop and train AI-based models for specific QoL issues. During the second part of the project, a multicenter prospective longitudinal study is planned. Eligible patients will be followed for one year with validated questionnaires regarding different QoL issues, and wearables that will collect active monitoring data on physical activity, sleep pattern, and heart rate. The collected data will be used to further train and optimize the AI-based models and personalized-based intervention suggestions.Based on the retrospective and prospective data, an ASCAPE-integrated prototype will be developed, enabling personalized- and AI-based predictions and intervention suggestions. This approach will be evaluated at the end of the prospective study regarding patients´ and physicians´ experience as well as health economics. Eligibility criteria Breast cancer patients planned for curative treatment with surgery with or without oncological therapy or breast cancer patients at least 1 year post-treatment (except endocrine therapy) will be eligible for the prospective study. Specific aims 1.To develop and optimize AI-based predictions for QoL issues in breast cancer patients as well as potential intervention suggestions.2.To evaluate the AI-based follow-up approach for breast cancer survivors in terms of patients´ experience, physicians´ experience, and health economics. Statistical methods For discrete QoL outcome variables, ASCAPE will examine the efficiency of classification-based machine learning models trained using decision tree learning algorithms, nearest-neighbors based algorithms, probabilistic learning algorithms, support vector machines and (deep) neural networks. Regressive counterparts of aforementioned methods will be analyzed for numeric QoL outcome variables including also regression specific methods (e.g., ridge regression, lasso regression and elastic net regression). The accuracy of trained models will be estimated relying on standard machine learning validation procedures such as the K-fold cross-validation and leave-one-out cross-validation. The ASCAPE platform will utilize state-of-the-art explainability techniques to make the machine learning models’ predictions transparent and comprehensible for the patient and the physician.Present accrual and target accrual Four retrospective datasets will be used for the first part of the project including approximately 18,000 breast cancer patients. For the prospective study, it is planned to be included about 30 patients monthly during a period of 12 months. Contact information for people with a specific interest in the trial https://ascape-project.eu/artificial-intelligence-supporting-cancer-patients-across-europe Citation Format: Antonios Valachis, Serge Autexier, Imma Grau, Lucian Itu, Dusan Jakotevic, Thanos Kosmidis, Montserrat Muñoz, Konstantinos Perakis, Johannes Rust, Milos Savic, Paris Kosmidis. Artificial intelligence supporting cancer patients across Europe - the ASCAPE project for breast cancer patients [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr OT-39-01.

Artificial Intelligence and the Future of Dispute Resolution: The Age of AI-DR

This chapter addresses the rise of artificial intelligence (AI) and its implications for the field of dispute resolution. In recent decades, digital technology has re-shaped alternative dispute resolution into online dispute resolution (ODR). In the last few years, ODR has spread to courts, further blurring boundaries between formal and informal dispute resolution, as well as between online and offline avenues of redress. More recently, AI has begun to infiltrate dispute resolution, mostly in the realm of AI-based prediction. With the maturing of ODR and the spread of AI, we can expect the overlap between the two to further evolve in both private and public dispute resolution, giving rise to a new form of dispute resolution, AI-DR. AI is increasingly being employed in a variety of legal and ADR settings and its role can be expected to grow further in the near future. The questions that remain open are the design choices that will be made with respect to AI and the means that will be employed to maximize its contribution and curb the challenges associated with AI-based decisions, recommendations, and predictions. Dispute resolution processes need not only to be efficient but to be fair, trustworthy, and accountable and must be perceived as such if they are to sustain their legitimacy and fulfill their societal role. In the sections that follow, we describe the relevance and current use of AI in dispute resolution, while drawing on particular examples to illustrate the promise and limitations of such use, which can be expected to expand dramatically in the next decade.

Robust and Interpretable Convolutional Neural Networks to Detect Glaucoma in Optical Coherence Tomography Images.

Recent studies suggest that deep learning systems can now achieve performance on par with medical experts in diagnosis of disease. A prime example is in the field of ophthalmology, where convolutional neural networks (CNNs) have been used to detect retinal and ocular diseases. However, this type of artificial intelligence (AI) has yet to be adopted clinically due to questions regarding robustness of the algorithms to datasets collected at new clinical sites and a lack of explainability of AI-based predictions, especially relative to those of human expert counterparts. In this work, we develop CNN architectures that demonstrate robust detection of glaucoma in optical coherence tomography (OCT) images and test with concept activation vectors (TCAVs) to infer what image concepts CNNs use to generate predictions. Furthermore, we compare TCAV results to eye fixations of clinicians, to identify common decision-making features used by both AI and human experts. We find that employing fine-tuned transfer learning and CNN ensemble learning create end-to-end deep learning models with superior robustness compared to previously reported hybrid deep-learning/machine-learning models, and TCAV/eye-fixation comparison suggests the importance of three OCT report sub-images that are consistent with areas of interest fixated upon by OCT experts to detect glaucoma. The pipeline described here for evaluating CNN robustness and validating interpretable image concepts used by CNNs with eye movements of experts has the potential to help standardize the acceptance of new AI tools for use in the clinic.

COVID‐19 and the New Technologies of Organizing: Digital Exhaust, Digital Footprints, and Artificial Intelligence in the Wake of Remote Work

COVID‐19 and the New Technologies of Organizing: Digital Exhaust, Digital Footprints, and Artificial Intelligence in the Wake of Remote Work

THE HIDDEN BLUEPRINT: A DISSECTION OF THE TECHNOLOGICAL FOOTPRINT ON LIVING SYSTEMS

The proliferation of technological systems which permeate, and often dominate, human sociological systems is ripe for critical exploration. The purpose of this panel is to investigate the myriad ways in which technology has encroached upon—and, arguably, engineered—human behaviors, once considered the domain of psychological and biological outcomes. Additionally, this panel will offer critical insight into how disproportionate and otherwise ethically skewed technological influence may be mitigated at the point of creation in order to restore balance between tech consumer and tech creator. Topics in this panel span human migration, algorithmic encoding, human-machine communication, datafication of the human subject, glitch technology, and AI-based prediction. This panel will investigate ways in which technology has converged with human systems, often unquestioned, and offer analysis into the potential consequences of this convergence, as well as potential solutions for alleviating the omnipresence of technical sovereignty on living systems. Specific panel presentations discuss: • The inorganic digital footprint that mitigates the tangible human footprint of human migration. • The pitfalls of cybernetic prediction and the corrosive nature of AI-based prediction. • The nature of human-machine communication, as documented by interactions with Amazon Go. • Normalization of gender, as evidenced by algorithmic determination in Spotify suggestions. • The emancipatory power of “breaking” the black box of participatory technology. Each panel topic identifies technology as the basis for living systems at the socio-cultural level. For example, multiple papers investigate how algorithms seek to manage, mitigate, and encourage human behavior, while others look at the physical technological infrastructure which guides the flow of human migration. The overarching goal of each paper encourages the end-user to take a deeper look at the symbiotic relationship between digital infrastructure and living systems in order to critically examine the consequences of convergence without critical oversight. Each paper in this panel also identifies a humanistic concept that interrogates the blurring between natural and artificial, human and machine, agency, and autonomy. Dissection of the hidden blueprint of techno-socio systems requires scholars to reconsider what it means to be human and forces a critical inquiry about why, how, and under what circumstances machines can or should engineer or augment human actors, and the extent to which machines can be made to act responsibly. For example, as displaced populations are forced from their nation-states, the ethical implications that those migrants of diaspora may tailor their migratory routes to current technological infrastructure must be considered. As we evolve towards engineering more scalable communication networks capable of harnessing the allegiance of a wide swath of the population based on geographic proximity, we mustn’t neglect to consider that these networks are driving human behavior as much, if not more, than they are supporting it. Similarly, when considering the hidden philosophical blueprint that prescribes predictive artificial intelligence, we mustn’t neglect to consider the assumptions which belie this philosophy, nor the impact of systems derived from such philosophy which claim to predict the patterns of our human behavior. These research topics identify the importance and relevance of scholarship in the area of human-machine communication and advocate for their inclusion in the conceptualization, prototyping, and creation of ethical robotic and AI technologies. Increasingly, humans find themselves socializing with intelligent agents and robots at home, in schools, and at work. Further, humans often do recognize the extent to which technological systems drive human behavior. This panel offers a glimpse into the necessary, provocative, and timely discussion about HMC and the role of critical scholarship in shaping technologies of the future. As technology progresses, we find ourselves on the precipice of social and cultural evolution in which the illusory real disguised as “user-friendliness” becomes more and more ubiquitous in design theory. In many respects, the trajectory of the modern-day Internet suggests that rather than heading towards emancipation (what user-friendliness promises)--we are headed closer to invisibility. Each of these panels reveals a mechanism by which we may peek into the constructs of our techno-social reality, as so that we may mitigate the dangers of such invisibility.

Detection of Disease-Specific Parent Cells Via Distinct Population of Nano-Vesicles by Machine Learning.

The diagnosis and prognosis of pathological conditions, such as age-related macular degeneration (AMD) and cancer still need improvement. AMD is primarily caused due to the dysfunction of retinal pigment epithelium (RPE), whereas endothelial cells (ECs) play one of the major roles in angiogenesis; an important process which occurs in malignant progression of cancer. Several reports suggested the augmented release of nano-vesicles under pathological conditions, including from RPE as well as cancer-associated ECs, which take part in various biological processes, including intercellular communication in disease progression. Importantly, these nano-vesicles are around 30-1000 nm and carry the fingerprint of their initiating parent cells (IPCs). Therefore, these nano-vesicles could be utilized as the diagnostic tool for AMD and cancer, respectively. However, the analysis of nano-vesicles for biomarker study is confounded by their extensive heterogeneous nature. To confront this challenge, we utilized artificial intelligence (AI) based machine learning (ML) algorithms such as support vector machine (SVM) and decision tree model on the dataset of nano-vesicles from RPE and ECs cell lines with low dimensionality. Overall, Gaussian SVM demonstrated the highest prediction accuracy of the IPCs of nano-vesicles, among all the chosen SVM classifiers. Additionally, the bagged tree showed the highest prediction among the chosen decision tree-based classifiers. Therefore, the overall bagged tree showed the best performance for the prediction of IPCs of nanovesicles, suggesting the applicability of AI-based prediction approach in diagnosis and prognosis of pathological conditions, including non-invasive liquid biopsy via various biofluids-derived nano-vesicles.

Identification of important factors in an inpatient fall risk prediction model to improve the quality of care using EHR and electronic administrative data: A machine-learning approach

BackgroundInpatient falls, many resulting in injury or death, are a serious problem in hospital settings. Existing falls risk assessment tools, such as the Morse Fall Scale, give a risk score based on a set of factors, but don’t necessarily signal which factors are most important for predicting falls. Artificial intelligence (AI) methods provide an opportunity to improve predictive performance while also identifying the most important risk factors associated with hospital-acquired falls. We can glean insight into these risk factors by applying classification tree, bagging, random forest, and adaptive boosting methods applied to Electronic Health Record (EHR) data. ObjectiveThe purpose of this study was to use tree-based machine learning methods to determine the most important predictors of inpatient falls, while also validating each via cross-validation. Materials and methodsA case-control study was designed using EHR and electronic administrative data collected between January 1, 2013 to October 31, 2013 in 14 medical surgical units. The data contained 38 predictor variables which comprised of patient characteristics, admission information, assessment information, clinical data, and organizational characteristics. Classification tree, bagging, random forest, and adaptive boosting methods were used to identify the most important factors of inpatient fall-risk through variable importance measures. Sensitivity, specificity, and area under the ROC curve were computed via ten-fold cross validation and compared via pairwise t-tests. These methods were also compared to a univariate logistic regression of the Morse Fall Scale total score. ResultsIn terms of AUROC, bagging (0.89), random forest (0.90), and boosting (0.89) all outperformed the Morse Fall Scale (0.86) and the classification tree (0.85), but no differences were measured between bagging, random forest, and adaptive boosting, at a p-value of 0.05. History of Falls, Age, Morse Fall Scale total score, quality of gait, unit type, mental status, and number of high fall risk increasing drugs (FRIDs) were considered the most important features for predicting inpatient fall risk. ConclusionsMachine learning methods have the potential to identify the most relevant and novel factors for the detection of hospitalized patients at risk of falling, which would improve the quality of patient care, and to more fully support healthcare provider and organizational leadership decision-making. Nurses would be able to enhance their judgement to caring for patients at risk for falls. Our study may also serve as a reference for the development of AI-based prediction models of other iatrogenic conditions. To our knowledge, this is the first study to report the importance of patient, clinical, and organizational features based on the use of AI approaches.

High throughput assessment of biomarkers in tissue microarrays using artificial intelligence: PTEN loss as a proof-of-principle in multi-center prostate cancer cohorts.

Phosphatase and tensin homolog (PTEN) loss is associated with adverse outcomes in prostate cancer and has clinical potential as a prognostic biomarker. The objective of this work was to develop an artificial intelligence (AI) system for automated detection and localization of PTEN loss on immunohistochemically (IHC) stained sections. PTEN loss was assessed using IHC in two prostate tissue microarrays (TMA) (internal cohort, n = 272 and external cohort, n = 129 patients). TMA cores were visually scored for PTEN loss by pathologists and, if present, spatially annotated. Cores from each patient within the internal TMA cohort were split into 90% cross-validation (N = 2048) and 10% hold-out testing (N = 224) sets. ResNet-101 architecture was used to train core-based classification using a multi-resolution ensemble approach (×5, ×10, and ×20). For spatial annotations, single resolution pixel-based classification was trained from patches extracted at ×20 resolution, interpolated to ×40 resolution, and applied in a sliding-window fashion. A final AI-based prediction model was created from combining multi-resolution and pixel-based models. Performance was evaluated in 428 cores of external cohort. From both cohorts, a total of 2700 cores were studied, with a frequency of PTEN loss of 14.5% in internal (180/1239) and external 13.5% (43/319) cancer cores. The final AI-based prediction of PTEN status demonstrated 98.1% accuracy (95.0% sensitivity, 98.4% specificity; median dice score = 0.811) in internal cohort cross-validation set and 99.1% accuracy (100% sensitivity, 99.0% specificity; median dice score = 0.804) in internal cohort test set. Overall core-based classification in the external cohort was significantly improved in the external cohort (area under the curve = 0.964, 90.6% sensitivity, 95.7% specificity) when further trained (fine-tuned) using 15% of cohort data (19/124 patients). These results demonstrate a robust and fully automated method for detection and localization of PTEN loss in prostate cancer tissue samples. AI-based algorithms have potential to streamline sample assessment in research and clinical laboratories.

AI-based prediction for the risk of coronary heart disease among patients with type 2 diabetes mellitus

Type 2 diabetes mellitus (T2DM) is one common chronic disease caused by insulin secretion disorder that often leads to severe outcomes and even death due to complications, among which coronary heart disease (CHD) represents the most common and severe one. Given a huge number of T2DM patients, it is thus increasingly important to identify the ones with high risks of CHD complication but the quantitative method is still not available. Here, we first curated a dataset of 1,273 T2DM patients including 304 and 969 ones with or without CHD, respectively. We then trained an artificial intelligence (AI) model using randomly selected 4/5 of the dataset and use the rest data to validate the performance of the model. The result showed that the model achieved an AUC of 0.77 (fivefold cross-validation) on the training dataset and 0.80 on the testing dataset. To further confirm the performance of the presented model, we recruited 1,253 new T2DM patients as totally independent testing dataset including 200 and 1,053 ones with or without CHD. And the model achieved an AUC of 0.71. In addition, we implemented a model to quantitatively evaluate the risk contribution of each feature, which is thus able to present personalized guidance for specific individuals. Finally, an online web server for the model was built. This study presented an AI model to determine the risk of T2DM patients to develop to CHD, which has potential value in providing early warning personalized guidance of CHD risk for both T2DM patients and clinicians.

AI-based prediction of independent construction safety outcomes from universal attributes

This paper significantly improves on, and finishes to validate, an approach proposed in previous research in which safety outcomes were predicted from attributes with machine learning. Like in the original study, we use Natural Language Processing (NLP) to extract fundamental attributes from raw incident reports and machine learning models are trained to predict safety outcomes. The outcomes predicted here are injury severity, injury type, body part impacted, and incident type. However, unlike in the original study, safety outcomes were not extracted via NLP but were provided by independent human annotations, eliminating any potential source of artificial correlation between predictors and predictands. Results show that attributes are still highly predictive, confirming the validity of the original approach. Other improvements brought by the current study include the use of (1) a much larger dataset featuring more than 90,000 reports, (2) two new models, XGBoost and linear SVM (Support Vector Machines), (3) model stacking, (4) a more straightforward experimental setup with more appropriate performance metrics, and (5) an analysis of per-category attribute importance scores. Finally, the injury severity outcome is well predicted, which was not the case in the original study. This is a significant advancement.

Artificial intelligence, machine learning, and deep learning in women's health nursing.

Artificial intelligence (AI), which includes machine learning and deep learning has been introduced to nursing care in recent years. The present study reviews the following topics: the concepts of AI, machine learning, and deep learning; examples of AI-based nursing research; the necessity of education on AI in nursing schools; and the areas of nursing care where AI is useful. AI refers to an intelligent system consisting not of a human, but a machine. Machine learning refers to computers’ ability to learn without being explicitly programmed. Deep learning is a subset of machine learning that uses artificial neural networks consisting of multiple hidden layers. It is suggested that the educational curriculum should include big data, the concept of AI, algorithms and models of machine learning, the model of deep learning, and coding practice. The standard curriculum should be organized by the nursing society. An example of an area of nursing care where AI is useful is prenatal nursing interventions based on pregnant women’s nursing records and AI-based prediction of the risk of delivery according to pregnant women’s age. Nurses should be able to cope with the rapidly developing environment of nursing care influenced by AI and should understand how to apply AI in their field. It is time for Korean nurses to take steps to become familiar with AI in their research, education, and practice.

Phase I study of a new concept cancer vaccine composed artificial intelligence (AI)-designed shared-antigen peptides plus combined synergistically activating antigen-specific CTL reaction (CYT001) in patients with advanced hepatocellular carcinoma (CRESCENT 1).

TPS595 Background: CYT001 (CYTLIMIC Inc.) is a novel cancer vaccine involving artificial intelligence (AI)-designed shared-antigen peptides and optimal combined adjuvants that boost the cancer-immunity cycle. The two multi-HLA reactive peptides heat shock protein 70 (HSP70) and glypican 3 (GPC3) were screened by an AI-based prediction system according to the proteome, mRNA, and histopathology data fromhuman samples. These immunogenic peptides were confirmed to show cross-reactivity to HLA-A 24:02, 02:01, and 02:06. Poly-ICLC (Oncovir Inc.) binds to Toll-like receptor 3 (TLR3) and melanoma differentiation antigen 5 (MDA5) on antigen-presenting cells (APCs)and activates APCs. LAG-3Ig (Immutep Inc.) binds to the major histocompatibility complex (MHC) class II molecules of APCs and activates APCs. Both poly-ICLC and LAG-3Ig synergistically activate antigen-specific CTL reactions as effective combination adjuvants. The present study aims to evaluate the safety and tolerability of CYT001 (mixture of HSP70 peptide [2.0 mg], GPC3 peptide [2.0 mg], poly-ICLC [1.0 mg], and LAG-3Ig [1.4 mg]) in patients with advanced hepatocellular carcinoma (HCC). Methods: This is a single-center, phase 1, open-label, single-arm, investigator-initiated clinical trial of CYT001 for advanced HCC patients with no eligible standard systemic therapy, Child–Pugh A liver disease, and HLA-A 24:02, 02:01, or 02:06. Enrolled patients will receive CYT001 as a subcutaneous injection on days 1, 8, 15, and 21 in the 1st and 2nd cycles, days 1 and 15 in the 3rd and 4th cycles, and day 1 in the 5th cycle or later of 28-day cycles. The primary endpoint is dose-limiting toxicity, and the secondary endpoints are safety and the response rate. The transition of the CTL reactions of both the HSP70 and GPC3 peptides will be evaluated using blood samples of the subjects. Exploratory analyses include investigation of candidate biomarkers for treatment efficacy using liver tumor biopsy samples (baseline and after the 1st cycle) and blood samples (baseline and every days of administration). Clinical trial information: jRCT2031190072.

Adapting ANFIS to Improve Field Rebound Hammer Test for Concrete Compressive Strength Estimation

Rebound hammer tests are one of the most popular non-destructive testing methods to examine the concrete compressive strength in the field. Rebound hammer tests are relatively easy to conduct and low cost. More importantly, it will not cause damage to the existing structure and can obtain the results in a short time. However, concrete compressive strength estimations provided by rebound hammer tests have an average of around 20% mean absolute percentage error (MAPE) when comparing to the results from destructive tests. This research proposes an alternative approach to estimate the concrete compressive strengths using the rebound hammer test data. The alternative approach is to adopt the Artificial Neural Fuzzy Inference Systems, ANFIS, to develop an AI-based prediction model for the rebound hammer tests. A total of 100 rebound hammer tests are conducted in a 24-story residential building. Core samples are carefully taken to obtain the actual compressive tests. The data collected are used to train and validate the ANFIS prediction model. The results show that the proposed ANFIS model has successfully reduced the MAPE to 10.01%.

Development of AI-based Prediction and Assessment Program for Tunnelling Impact

Development of AI-based Prediction and Assessment Program for Tunnelling Impact

Characterizing Artificial Intelligence Applications in Cancer Research: A Latent Dirichlet Allocation Analysis.

BackgroundArtificial intelligence (AI)–based therapeutics, devices, and systems are vital innovations in cancer control; particularly, they allow for diagnosis, screening, precise estimation of survival, informing therapy selection, and scaling up treatment services in a timely manner.ObjectiveThe aim of this study was to analyze the global trends, patterns, and development of interdisciplinary landscapes in AI and cancer research.MethodsAn exploratory factor analysis was conducted to identify research domains emerging from abstract contents. The Jaccard similarity index was utilized to identify the most frequently co-occurring terms. Latent Dirichlet Allocation was used for classifying papers into corresponding topics.ResultsFrom 1991 to 2018, the number of studies examining the application of AI in cancer care has grown to 3555 papers covering therapeutics, capacities, and factors associated with outcomes. Topics with the highest volume of publications include (1) machine learning, (2) comparative effectiveness evaluation of AI-assisted medical therapies, and (3) AI-based prediction. Noticeably, this classification has revealed topics examining the incremental effectiveness of AI applications, the quality of life, and functioning of patients receiving these innovations. The growing research productivity and expansion of multidisciplinary approaches are largely driven by machine learning, artificial neural networks, and AI in various clinical practices.ConclusionsThe research landscapes show that the development of AI in cancer care is focused on not only improving prediction in cancer screening and AI-assisted therapeutics but also on improving other corresponding areas such as precision and personalized medicine and patient-reported outcomes.

Automated classification of dry eye type analyzing interference fringe color images of tear film using machine learning techniques

The unstable balance in secretions of lipids and aqueous fluid to tear film is a significant cause of dry eye disease (DED). Arita et al. demonstrated a simple but very effective method that classifies dry eye types to the aqueous deficient dry eye (ADDE) and the evaporative dry eye (EDE) by focusing on the dry eye type-unique appearances of interference fringe colors and patterns of tear films. We thought this simple classification is very helpful for diagnoses and treatments. However, diagnostic bias by unskilled observers remains an issue to be solved.
 The artificial intelligence (AI)-based support for diagnosis is one of the hottest topics in the field of ophthalmology research. We expected that the AI-based model would reduce bias in DED-type diagnoses. Many studies have been reported targeting retinal diseases like age-related macular degeneration and/or diabetic retinopathy. Most of the works established AI-based predicting models using images taken by fundus cameras and/or optical coherence tomography (OCT) devices to capture disease-related structural disorders. In contrast, the interference fringes dynamically change the colors and patterns spatiotemporally. To the best of our knowledge, there is no AI-based model studied for distinguishing ADDE and EDE using interference fringe images. However, an AI-based study classifying the condition of the tear lipid layer by analyzing the textures of interference fringes compared to the device-unique grades has been reported. This suggested the possibility of using the unstructured characteristics, such as colors and/or complexities of interference fringes, as the numerical image features when building AI-based prediction models. In this study, we first examined several types of image characteristics extracted from the colors and patterns of fringes to obtain effective image features for the DED-type classification. We then evaluated whether the AI-based models would have sufficient abilities for this type of prediction by comparing their diagnoses with those made by an ophthalmologist skilled in this classification (the founder of this type classification).

Application of Artificial Intelligence in Predicting Machining Surface Quality

Achieving high-quality machining surface finishes is crucial in numerous manufacturing industries, as it directly impacts the performance and reliability of machined components. Traditionally, predicting machining surface quality involves extensive trial-and-error experiments, which are time-consuming, costly, and often impractical for complex machining processes. In recent years, the application of artificial intelligence (AI) techniques, particularly machine learning algorithms, has emerged as a promising approach for predicting machining surface quality accurately and efficiently. This abstract provides an overview of the application of AI in predicting machining surface quality, highlighting its benefits, challenges, and future prospects. The adoption of AI in predicting machining surface quality involves various stages. Firstly, a comprehensive dataset is collected, comprising machining parameters, tooling characteristics, and corresponding surface quality measurements. Next, pre-processing techniques are applied to clean and normalize the dataset, ensuring its suitability for training AI models. Subsequently, machine learning algorithms, such as support vector machines, neural networks, and random forests, are trained using the pre-processed data to develop predictive models. These models can capture complex relationships between machining parameters and surface quality, enabling accurate predictions.
 The performance of the trained models is assessed using appropriate evaluation metrics, such as root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R-squared). Comparative analyses are conducted to identify the most effective AI model for predicting machining surface quality. Additionally, sensitivity analyses and feature selection techniques can be applied to identify the critical machining parameters that significantly impact surface quality. The application of AI in predicting machining surface quality offers several practical implications. It enables manufacturers to optimize machining processes, reduce scrap rates, and enhance product quality. By accurately predicting surface quality, manufacturers can make informed decisions regarding machining parameters, tooling selection, and process optimization, resulting in improved efficiency and cost-effectiveness. Furthermore, AI-based prediction models can be integrated into real-time monitoring systems, enabling continuous quality control and immediate adjustments to machining processes.
 Despite the benefits, challenges exist in the application of AI for predicting machining surface quality. These include the availability and quality of training data, selection of appropriate features, and the interpretability of AI models. Overcoming these challenges requires continuous research and development efforts, such as the collection of large and diverse datasets, advanced feature engineering techniques, and the exploration of explainable AI methodologies.

Artificial intelligence techniques for prediction of solar radiation data: a review

Artificial intelligence (AI) techniques are becoming useful as alternate approaches to conventional techniques. They have been used to solve complicated practical problems in various areas and are able to deal with non-linear problems. AI techniques have been applied for modelling, identification, optimisation, prediction, forecasting, and control of complex systems. Artificial neural networks have been used by the author in the field of solar energy for the estimation of solar radiation data. The main objective of this paper is to present an overview of the AI-techniques for prediction of solar radiation data. Published literature presented in this paper show the potential of AI as a design tool for the simulation of solar radiation data and also forecasting and modelling others meteorological data. The advantage of using an AI-based prediction of solar radiation is that it provides good estimation of data, especially in isolated areas, where the weather data are not available.

ARTIFICIAL INTELLIGENCE-BASED PREDICTION MODELS FOR ENVIRONMENTAL ENGINEERING

A literature survey was conducted to appraise the recent applications of artiflcal intelligence (AI)-based modeling studies in the environmental engineer- ing fleld. A number of studies on artiflcial neural networks (ANN), fuzzy logic and adaptive neuro-fuzzy systems (ANFIS) were reviewed and important aspects of these models were highlighted. The results of the extensive literature survey showed that most AI-based prediction models were implemented for the solution of water/wastewater (55.7%) and air pollution (30.8%) related environmental prob- lems compared to solid waste (13.5%) management studies. The present literature review indicated that among the many types of ANNs, the three-layer feed-forward and back-propagation (FFBP) networks were considered as one of the simplest and the most widely used network type. In general, the Levenberg-Marquardt algo- rithm (LMA) was found as the best-suited training algorithm for several complex and nonlinear real-life problems of environmental engineering. The literature sur- vey showed that for water and wastewater treatment processes, most of AI-based prediction models were introduced to estimate the performance of various biolog- ical and chemical treatment processes, and to control e†uent pollutant loads and ∞owrates from a speciflc system. In air polution related environmental problems, forecasting of ozone (O3) and nitrogen dioxide (NO2) levels, daily and/or hourly particulate matter (PM2:5 and PM10) emissions, and sulfur dioxide (SO2) and car- bon monoxide (CO) concentrations were found to be widely modeled. For solid waste management applications, reseachers conducted studies to model weight of waste generation, solid waste composition, and total rate of waste generation.

Predicting the length of hospital stay of burn patients: Comparisons of prediction accuracy among different clinical stages

A burn injury is a disastrous trauma and can have wide-ranging impacts on burn patients, their families, and society. Burn patients generally experience long hospital stays, and the accurate prediction of the length of those stays has strong implications for healthcare resource management and service delivery. In addition to prediction accuracy, the timing of length of hospital stay (LOS) predictions is also relevant, because LOS predictions during earlier clinical stages (e.g., admission) can provide an important component for service and resource planning as well as patient and family counseling, whereas LOS predictions at later clinical stages (e.g., post-treatment) can support resource utilization reviews and cost controls. This study evaluates the effectiveness of LOS predictions for burn patients during three different clinical stages: admission, acute, and post-treatment. In addition, we compare the prediction effectiveness of two artificial intelligence (AI)-based prediction techniques (i.e., model-tree-based regression and support vector machine regression), using linear regression analysis as our performance benchmark. On the basis of 1080 burn cases collected in Taiwan, the empirical evaluation suggests that the accuracy of LOS predictions at the acute stage does not improve compared with those during the admission stage, but LOS predictions at the post-treatment stage are significantly more accurate. Moreover, the AI-based prediction techniques, especially support vector machine regression, appear more effective than the regression technique for LOS predictions for burn patients across stages.