Proposed Machine Learning Model Research Articles

An experimental machine learning study investigating the decision-making process of students and qualified radiographers when interpreting radiographic images.

AI is becoming more prevalent in healthcare and is predicted to be further integrated into workflows to ease the pressure on an already stretched service. The National Health Service in the UK has prioritised AI and Digital health as part of its Long-Term Plan. Few studies have examined the human interaction with such systems in healthcare, despite reports of biases being present with the use of AI in other technologically advanced fields, such as finance and aviation. Understanding is needed of how certain user characteristics may impact how radiographers engage with AI systems in use in the clinical setting to mitigate against problems before they arise. The aim of this study is to determine correlations of skills, confidence in AI and perceived knowledge amongst student and qualified radiographers in the UK healthcare system. A machine learning based AI model was built to predict if the interpreter was either a student (n = 67) or a qualified radiographer (n = 39) in advance, using important variables from a feature selection technique named Boruta. A survey, which required the participant to interpret a series of plain radiographic examinations with and without AI assistance, was created on the Qualtrics survey platform and promoted via social media (Twitter/LinkedIn), therefore adopting convenience, snowball sampling This survey was open to all UK radiographers, including students and retired radiographers. Pearson's correlation analysis revealed that males who were proficient in their profession were more likely than females to trust AI. Trust in AI was negatively correlated with age and with level of experience. A machine learning model was built, the best model predicted the image interpreter to be qualified radiographers with 0.93 area under curve and a prediction accuracy of 93%. Further testing in prospective validation cohorts using a larger sample size is required to determine the clinical utility of the proposed machine learning model.

Fall-portent detection for construction sites based on computer vision and machine learning

PurposeFall accidents can cause casualties and economic losses in the construction industry. Fall portents, such as loss of balance (LOB) and sudden sways, can result in fatal, nonfatal or attempted fall accidents. All of them are worthy of studying to take measures to prevent future accidents. Detecting fall portents can proactively and comprehensively help managers assess the risk to workers as well as in the construction environment and further prevent fall accidents.Design/methodology/approachThis study focused on the postures of workers and aimed to directly detect fall portents using a computer vision (CV)-based noncontact approach. Firstly, a joint coordinate matrix generated from a three-dimensional pose estimation model is employed, and then the matrix is preprocessed by principal component analysis, K-means and pre-experiments. Finally, a modified fusion K-nearest neighbor-based machine learning model is built to fuse information from the x, y and z axes and output the worker's pose status into three stages.FindingsThe proposed model can output the worker's pose status into three stages (steady–unsteady–fallen) and provide corresponding confidence probabilities for each category. Experiments conducted to evaluate the approach show that the model accuracy reaches 85.02% with threshold-based postprocessing. The proposed fall-portent detection approach can extract the fall risk of workers in the both pre- and post-event phases based on noncontact approach.Research limitations/implicationsFirst, three-dimensional (3D) pose estimation needs sufficient information, which means it may not perform well when applied in complicated environments or when the shooting distance is extremely large. Second, solely focusing on fall-related factors may not be comprehensive enough. Future studies can incorporate the results of this research as an indicator into the risk assessment system to achieve a more comprehensive and accurate evaluation of worker and site risk.Practical implicationsThe proposed machine learning model determines whether the worker is in a status of steady, unsteady or fallen using a CV-based approach. From the perspective of construction management, when detecting fall-related actions on construction sites, the noncontact approach based on CV has irreplaceable advantages of no interruption to workers and low cost. It can make use of the surveillance cameras on construction sites to recognize both preceding events and happened accidents. The detection of fall portents can help worker risk assessment and safety management.Originality/valueExisting studies using sensor-based approaches are high-cost and invasive for construction workers, and others using CV-based approaches either oversimplify by binary classification of the non-entire fall process or indirectly achieve fall-portent detection. Instead, this study aims to detect fall portents directly by worker's posture and divide the entire fall process into three stages using a CV-based noncontact approach. It can help managers carry out more comprehensive risk assessment and develop preventive measures.

Short-term solar energy forecasting: Integrated computational intelligence of LSTMs and GRU.

Problems with erroneous forecasts of electricity production from solar farms create serious operational, technological, and financial challenges to both Solar farm owners and electricity companies. Accurate prediction results are necessary for efficient spinning reserve planning as well as regulating inertia and power supply during contingency events. In this work, the impact of several climatic conditions on solar electricity generation in Amherst. Furthermore, three machine learning models using Lasso Regression, ridge Regression, ElasticNet regression, and Support Vector Regression, as well as deep learning models for time series analysis include long short-term memory, bidirectional LSTM, and gated recurrent unit along with their variants for estimating solar energy generation for every five-minute interval on Amherst weather power station. These models were evaluated using mean absolute error root means square error, mean square error, and mean absolute percentage error. It was observed that horizontal solar irradiance and water saturation deficiency had a highly proportional relationship with Solar PV electricity generation. All proposed machine learning models turned out to perform well in predicting electricity generation from the analyzed solar farm. Bi-LSTM has performed the best among all models with 0.0135, 0.0315, 0.0012, and 0.1205 values of MAE, RMSE, MSE, and MAPE, respectively. Comparison with the existing methods endorses the use of our proposed RNN variants for higher efficiency, accuracy, and robustness. Multistep-ahead solar energy prediction is also carried out by exploiting hybrids of LSTM, Bi-LSTM, and GRU.

GODDAG: Generating Origin-Destination Flow for New Cities Via Domain Adversarial Training

Origin-destination (OD) flow data, which reflects population mobility patterns in the city, is very important in many urban applications, such as urban planning and public resource allocation, etc. However, due to the high cost of money and time during device deployment and social surveys, it is challenging to obtain OD flow data, especially in developing cities and emerging cities where historical OD flow data is scarce. Therefore, it is necessary to investigate a method that can generate OD flow in cities where OD flow data are not available. The research on modeling population mobility in the city has a long history. Traditional gravity models, etc., are too simple to model the complex population mobility; recently proposed machine learning models and deep learning models are not applicable in cities where data are scarce because the parameters must be fitted with abundant data. To solve the problem of difficult access to OD flow data, we propose a method to learn mobility knowledge with ample data in the source city and generate OD flow data in new cities named <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GODDAG</b> ( <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><u>G</u></b> enerating <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><u>O</u></b> rigin- <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><u>D</u></b> estination Flow via <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><u>D</u></b> omain <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><u>A</u></b> dversarial Trainin <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><u>g</u></b> ). Our proposed method consists of two parts, one is a GNN (graph neural networks) based mobility model generating OD flow between every two regions based on regional attributes such as census and POI distribution, and the other is a domain adversarial training strategy to make the model have better transfer ability between different cities. Extensive experiments are conducted on two real-world datasets to prove the validity of our methods.

Machine learning based energy demand prediction

Energy demand prediction can benefit electricity consumers, distribution network service providers, and system operators. It heavily depends on auxiliary factors, such as weather parameters (e.g., ambient temperature), which makes this problem more complex. Moreover, for many industrial applications, instead of aiming for the highest prediction accuracy, a more easily understandable and interpretable model that can lead to higher accuracy against the baseline model is the priority. Therefore, this problem still requires more investigation, especially when there is a specified prediction baseline to be compared with. This paper proposes a machine learning (ML) based prediction framework that investigates how temperature combined with energy consumption and simple and interpretable ML methods can be used to provide more precise demand forecasts and thus baselines closer to actual load profiles. The proposed framework is tested on two different real-world energy demand datasets. The analysis shows that using a simple ML model, such as a polynomial regression model, results in a more accurate prediction than the current baselines used in the energy market. The proposed ML models are not black-box type models, and thus are easier to explain and interpret. The ML-based forecasted demand is used as a baseline for demand response (DR) and is compared with the existing baselines used in the demand response market of Australia’s national grid.

Machine learning strategy for identifying altered gut microbiomes for diagnostic screening in myasthenia gravis.

Myasthenia gravis (MG) is a neuromuscular junction disease with a complex pathophysiology and clinical variation for which no clear biomarker has been discovered. We hypothesized that because changes in gut microbiome composition often occur in autoimmune diseases, the gut microbiome structures of patients with MG would differ from those without, and supervised machine learning (ML) analysis strategy could be trained using data from gut microbiota for diagnostic screening of MG. Genomic DNA from the stool samples of MG and those without were collected and established a sequencing library by constructing amplicon sequence variants (ASVs) and completing taxonomic classification of each representative DNA sequence. Four ML methods, namely least absolute shrinkage and selection operator, extreme gradient boosting (XGBoost), random forest, and classification and regression trees with nested leave-one-out cross-validation were trained using ASV taxon-based data and full ASV-based data to identify key ASVs in each data set. The results revealed XGBoost to have the best predicted performance. Overlapping key features extracted when XGBoost was trained using the full ASV-based and ASV taxon-based data were identified, and 31 high-importance ASVs (HIASVs) were obtained, assigned importance scores, and ranked. The most significant difference observed was in the abundance of bacteria in the Lachnospiraceae and Ruminococcaceae families. The 31 HIASVs were used to train the XGBoost algorithm to differentiate individuals with and without MG. The model had high diagnostic classification power and could accurately predict and identify patients with MG. In addition, the abundance of Lachnospiraceae was associated with limb weakness severity. In this study, we discovered that the composition of gut microbiomes differed between MG and non-MG subjects. In addition, the proposed XGBoost model trained using 31 HIASVs had the most favorable performance with respect to analyzing gut microbiomes. These HIASVs selected by the ML model may serve as biomarkers for clinical use and mechanistic study in the future. Our proposed ML model can identify several taxonomic markers and effectively discriminate patients with MG from those without with a high accuracy, the ML strategy can be applied as a benchmark to conduct noninvasive screening of MG.

Study of various machine learning approaches for Sentinel-2 derived bathymetry.

In recent years precise and up-to-date information regarding seabed depth has become more and more important for companies and institutions that operate on coastlines. While direct, in-situ measurements are performed regularly, they are expensive, time-consuming and impractical to be performed in short time intervals. At the same time, an ever-increasing amount of satellite imaging data becomes available. With these images, it became possible to develop bathymetry estimation algorithms that can predict seabed depth and utilize them systematically. Since there are a number of theoretical approaches, physical models, and empirical techniques to use satellite observations in order to estimate depth in the coastal zone, the presented article compares the performance and precision of the most common one to modern machine learning algorithms. More specifically, the models based on shallow neural networks, decision trees and Random Forest algorithms have been proposed, investigated and confronted with the performance of pure analytical models. The particular proposed machine learning models differ also in a set of satellite data bands used as an input as well as in applying or not geographical weighting in the learning process. The obtained results point towards the best performance of the regression tree algorithm that incorporated as inputs information about data localization, raw reflectance data from four satellite data bands and a quotient of logarithms of B2 and B3 bands. The study for the paper was performed in relatively optically difficult and spatially variant conditions of the south Baltic coastline starting at Szczecin, Poland on the west (53°26'17'' N, 14°32'32'' E) to Hel peninsula (54°43'04,3774'' N 18°37'56,9175'' E). The reference bathymetry data was acquired from Polish Marine Administration. It was obtained through profile probing with single-beam sonar or direct in-situ probing.

MRI Radiomics-Based Machine Learning Models for Ki67 Expression and Gleason Grade Group Prediction in Prostate Cancer.

The Ki67 index and the Gleason grade group (GGG) are vital prognostic indicators of prostate cancer (PCa). This study investigated the value of biparametric magnetic resonance imaging (bpMRI) radiomics feature-based machine learning (ML) models in predicting the Ki67 index and GGG of PCa. A total of 122 patients with pathologically proven PCa who had undergone preoperative MRI were retrospectively included. Radiomics features were extracted from T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) maps. Then, recursive feature elimination (RFE) was applied to remove redundant features. ML models for predicting Ki67 expression and GGG were constructed based on bpMRI and different algorithms, including logistic regression (LR), support vector machine (SVM), random forest (RF), and K-nearest neighbor (KNN). The performances of different models were evaluated with receiver operating characteristic (ROC) analysis. In addition, a joint analysis of Ki67 expression and GGG was performed by assessing their Spearman correlation and calculating the diagnostic accuracy for both indices. The ML model based on LR and ADC + T2 (LR_ADC + T2, AUC = 0.8882) performed best in predicting Ki67 expression, and ADC_wavelet-LHH_firstorder_Maximum had the highest feature weighting. The SVM_DWI + T2 (AUC = 0.9248) performed best in predicting GGG, and DWI_wavelet HLL_glcm_SumAverage had the highest feature weighting. The Ki67 and GGG exhibited a weak positive correlation (r = 0.382, p < 0.001), and LR_ADC + DWI had the highest diagnostic accuracy in predicting both (0.6230). The proposed ML models are suitable for predicting both Ki67 expression and GGG in PCa. This algorithm could be used to identify indolent or invasive PCa with a noninvasive, repeatable, and accurate diagnostic method.

Cardiometabolic risk estimation using exposome data and machine learning

BackgroundThe human exposome encompasses all exposures that individuals encounter throughout their lifetime. It is now widely acknowledged that health outcomes are influenced not only by genetic factors but also by the interactions between these factors and various exposures. Consequently, the exposome has emerged as a significant contributor to the overall risk of developing major diseases, such as cardiovascular disease (CVD) and diabetes. Therefore, personalized early risk assessment based on exposome attributes might be a promising tool for identifying high-risk individuals and improving disease prevention. ObjectiveDevelop and evaluate a novel and fair machine learning (ML) model for CVD and type 2 diabetes (T2D) risk prediction based on a set of readily available exposome factors. We evaluated our model using internal and external validation groups from a multi-center cohort. To be considered fair, the model was required to demonstrate consistent performance across different sub-groups of the cohort. MethodsFrom the UK Biobank, we identified 5,348 and 1,534 participants who within 13 years from the baseline visit were diagnosed with CVD and T2D, respectively. An equal number of participants who did not develop these pathologies were randomly selected as the control group. 109 readily available exposure variables from six different categories (physical measures, environmental, lifestyle, mental health events, sociodemographics, and early-life factors) from the participant's baseline visit were considered. We adopted the XGBoost ensemble model to predict individuals at risk of developing the diseases. The model's performance was compared to that of an integrative ML model which is based on a set of biological, clinical, physical, and sociodemographic variables, and, additionally for CVD, to the Framingham risk score. Moreover, we assessed the proposed model for potential bias related to sex, ethnicity, and age. Lastly, we interpreted the model's results using SHAP, a state-of-the-art explainability method. ResultsThe proposed ML model presents a comparable performance to the integrative ML model despite using solely exposome information, achieving a ROC-AUC of 0.78±0.01 and 0.77±0.01 for CVD and T2D, respectively. Additionally, for CVD risk prediction, the exposome-based model presents an improved performance over the traditional Framingham risk score. No bias in terms of key sensitive variables was identified. ConclusionsWe identified exposome factors that play an important role in identifying patients at risk of CVD and T2D, such as naps during the day, age completed full-time education, past tobacco smoking, frequency of tiredness/unenthusiasm, and current work status. Overall, this work demonstrates the potential of exposome-based machine learning as a fair CVD and T2D risk assessment tool.

Dynamic slicing reconfiguration for virtualized 5G networks using ML forecasting of computing capacity

As 5G deployments continue to increase worldwide, new applications can fully leverage the exceptional features of the emerging mobile networks. Ultra-Reliable Low Latency Communications (URLLC) serve as an excellent example of applications highly sensitive to jitter and packet loss. To meet these demanding requirements, 5G relies on network slicing, network virtualization, and software-defined networks. This ecosystem enables the precise allocation of resources for each network slice. However, the applications’ resource demands may vary over time. In this challenging and overwhelming environment, traditional human decision-making for slice reconfiguration is not suitable anymore, due to the multitude of parameters and the need for extremely fast response times. Machine Learning (ML) comes as a tool that can enable better use of the available resources with faster and more intelligent management. This paper introduces an ML model that can predict slices’ traffic and dynamically reconfigure computational capacity. With these forecasting capabilities, the virtualized resources can be fine-tuned to suit the slices’ requirements, guaranteeing their Quality of Service (QoS). By doing so, Mobile Network Operators can make optimized use of the equipment, tailoring their needs to each service while complying with the QoS level. The results obtained demonstrate that the proposed ML model, in combination with a specific set of hysteresis rules, can accurately predict the saturation of virtualized capacity with up to 91% accuracy and proactively adapt it to the network slice requirements.

Urban traffic volume estimation using intelligent transportation system crowdsourced data

Traffic volume is a crucial information for many different fields, such as city planner, logistic planning and more. However, installing sensors on each road to collect traffic volume data for the whole traffic network is impractical due to high cost and human labour. Most recent studies implement machine learning, mathematical and statistical methods to learn the behaviour of traffic volume. However, the randomness of traffic volume can hardly be defined by equations or statistical models which leads to the proposed machine learning model. This paper proposed a novel spatial prediction to fill up the traffic volume of a whole network with an estimated 10% of ground truth data. To make up for the lack of data, a spatial-temporal weightage is assigned to each road before fitting the training sample into a tree ensemble model to perform a prediction of the connecting roads. The weightage is first computed using the 10% ground truth data and then the weightage is spread to connecting roads via an innovative repetitive breadth-first search (BFS) method that capture the spatial correlation of a traffic network. Various experiments were conducted to assess the significance of spatial weighting and it was observed that incorporating the weighting resulted in a 1.69% improvement in the Mean Absolute Percentage Error (MAPE). The temporal relationship can be learnt from the trend of hourly traffic data for every day of the week. The proposed model achieved an average percentage error of 2.63% with reduced average percentage error by 95% compared to existing methods.

Evolving Well Stimulation Optimization Tool with OliFANT

In Indonesia, for a half decade, the decrease of oil and gas production from 2016 is 4.23% and 3.53% respectively (ESDM, 2021). This production decrease has a domino effect on the investment loss. According to the International Trade Administration, investment in Indonesia’s oil and gas industry in 2019 reached around US$ 12 billion, which was decreasing from around US$ 16 billion in 2016. Such loss is a serious disaster, thus applying digital transformation such as machine learning to the most-used method, well stimulation, is immediately needed. Unfortunately, the implemented well stimulations nowadays are prone to short-lived effects due to the unreliable selection methods, as they do not have any integrated database. This research, as the pilot project, focuses on field data collected in West Indonesia from sandstone and carbonate lithologies, and the type of stimulation used is acidizing. This tool, OliFANT, defines the success of stimulation based on the productivity index before and after stimulation. The method uses geostatistical approaches and optimizing decline curve analysis for analysing and modelling spatially correlated data. The accuracy of the model is validated at a minimum of 75%, which shows its high reliability. It can also forecast the duration effect of the stimulation, additionally it provides the estimation of profit scenarios. The proposed machine learning model adopts an empirical working principle by utilizing reservoir parameters and test data of stimulation, which are inputted into a user-friendly interface after filling in a comprehensive database. In conclusion, the main benefits of using this tool are cutting evaluation time and achieving higher cost-efficiency. This software can be continuously improved by adding more data to widen the variety of the methods. Considering that each field has different types of properties, this tool is built to be adaptable to every reservoir condition. Over and above that, this tool can be implemented for other stimulated wells and be modified for other methods and operations, such as drilling and workover. In the future, it can be a one-stop solution for stimulation plan validation, where data-driven solutions pave the way for success.

Deep learning for anomaly detection in wire-arc additive manufacturing

Wire-arc additive manufacturing (WAAM) is becoming the most important metal additive manufacturing process in many industries. In this paper, one of the common problems of irregularity in the metal deposition in WAAM has been addressed and solved using machine learning (ML). A deep learning-based convolutional neural network (CNN) was used to classify the two classes of deposited beads, i.e. ‘regular bead’ and ‘irregular bead’. A digital camera was installed with a WAAM setup to obtain the images of beads after deposition. A single layer of deposition was conducted on a substrate using aluminium 5356 alloy filler wire using robotic-controlled gas-metal arc welding (GMAW) setup. The performance of the ML model was validated using classification accuracy and processing time. The developed CNN model was checked with three types of proposed datasets. The dataset containing the training and testing ratio of 60:40 achieved an accuracy of 86.53% and 88.08% with 30 and 60 epochs respectively for testing. The proposed ML model was successful in anomaly detection in the deposited bead of WAAM and hence it helps in improving the quality of deposited layers and mechanical properties of fabricated parts.

Machine Learning Algorithms in Identifying Balanced Diet Plan for Healthy Life style

The present generation of all ages is terribly facing the challenges of obesity in recent times. The people suffering from this disorder practice different diet plans for weight reduction without considering the balanced proportion of nutrients in their diet. This paper aims in highlighting the ill effects of unbalanced diet plans and proposes a machine learning (ML) model based on support vector machine to make decisions on the balanced nature of the diet. The efficiency of the proposed ML model is compared with other ML algorithms. The accuracy results of the proposed model are more convincing in comparison with other ML algorithms. The proposed ML model is applied to deterministic type of secondary data sets and this shall be extended by applying to fuzzy data sets. This research work applies the algorithms of machine learning to health-based decision-making systems

SOLAR RADIATION FORECASTING USING ADAPTIVE NEURO FUZZY INFERENCE SYSTEM (ANFIS)

Hybrid intelligent systems have previously been centered on forecasting solar energy using meteorological data parameters; nevertheless, such forecasting approaches have yielded unreliable results. The goal of this research is to design and develop ANFIS model for forecasting solar energy based on widely shifting environmental factors utilizing experimental data collected in Abuja. This is accomplished by developing a system that predicts solar energy generation using an Artificial Neural Network (ANN) and fuzzy logic. This research examines four Adaptive Neuro Fuzzy Inference System (ANFIS) models that were developed and tested in Abuja, Nigeria, for horizontal sun radiation prediction. These models were generated by varying the number of inputs for each model with the output being solar radiation. Data set of 30 years were collected from the National Space Research Development Agency (NASDRA) and procured for the presented simulation study. Simulation using ANFIS methodology was carried out using MATLAB Tool. The obtained data was divided into two categories: training and testing with training having 70% and testing 30% of the data sets. The simulation results were checked against the data and confirmed to be within allowable limits. The coefficients of determination (R2), correlation coefficient(R), Mean Square Error (MSE), Root Mean Square Error (RMSE) were calculated to demonstrate the effectiveness of the proposed machine learning models. It can be safely concluded that model 4 gave an accurate result with high efficiency and less error.

Predicting visual recovery in pituitary adenoma patients post-endoscopic endonasal transsphenoidal surgery: Harnessing delta-radiomics of the optic chiasm from MRI

ObjectivesTo investigate whether morphological changes after surgery and delta-radiomics of the optic chiasm obtained from routine MRI could help predict postoperative visual recovery of pituitary adenoma patients.MethodsA total of 130 pituitary adenoma patients were retrospectively enrolled and divided into the recovery group (n = 87) and non-recovery group (n = 43) according to visual outcome 1 year after endoscopic endonasal transsphenoidal surgery. Morphological parameters of the optic chiasm were measured preoperatively and postoperatively, including chiasmal thickness, deformed angle, and suprasellar extension. Delta-radiomics of the optic chiasm were calculated based on features extracted from preoperative and postoperative coronal T2-weighted images, followed by machine learning modeling using least absolute shrinkage and selection operator wrapped with support vector machine through fivefold cross-validation in the development set. The delta-radiomic model was independently evaluated in the test set, and compared with the combined model that incorporated delta-radiomics, significant clinical and morphological parameters.ResultsPostoperative morphological changes of the optic chiasm could not significantly be used as predictors for the visual outcome. In contrast, the delta-radiomics model represented good performances in predicting visual recovery, with an AUC of 0.821 in the development set and 0.811 in the independent test set. Moreover, the combined model that incorporated age and delta-radiomics features of the optic chiasm achieved the highest AUC of 0.841 and 0.840 in the development set and independent test set, respectively.ConclusionsOur proposed machine learning models based on delta-radiomics of the optic chiasm can be used to predict postoperative visual recovery of pituitary adenoma patients.Clinical relevance statementOur delta-radiomics-based models from MRI enable accurate visual recovery predictions in pituitary adenoma patients who underwent endoscopic endonasal transsphenoidal surgery, facilitating better clinical decision-making and ultimately improving patient outcomes.Key Points• Prediction of the postoperative visual outcome for pituitary adenoma patients is important but challenging.• Delta-radiomics of the optic chiasm after surgical decompression represented better prognostic performances compared with its morphological changes.• The proposed machine learning models can serve as novel approaches to predict visual recovery for pituitary adenoma patients in clinical practice.

Artificial Neural Network prediction of forming limit diagram for directionally-rolled, size scaled copper strips

Estimating the forming limit diagram (FLD) is tedious and cost-intensive. Methods driven by data and artificial intelligence are used to determine the relationship between scaled thickness and the forming rates of various cups drawn out of ETP copper sheets. Machine learning (ML) techniques have a good chance of predicting the FLD of copper alloys, and they are being used increasingly in sensitive electronic and structural applications. The current research aims to create ML-based artificial neural network (ANN) tools to model the relationship between scaled thickness and forming rates as a function of formability. The FLD is measured for copper strips of initial dimensions of 1500 mm long, 750 mm wide, and 6 mm thick, whose thickness was reduced successively by 50% in nine incremental steps. Thus, 3, 1.5, 0.75, 0.38, and 0.19 mm sheets were obtained and used to determine FLD through the Nakajima approach. An FEA model of the drawing was made in Altair Inspire Form, and the simulation results were used to train a two-step ML. A Bayesian regularization (BR) and Levenberg-Marquardt algorithm (LM) were used in the first step to predict strains’ maximum and minimum points. In the second step, the minor strains predicted in the first step are used as inputs. Using the same feature set, the BR and LM algorithms predict the major strain, showing a linear trend until the middle and then a nonlinear trend. The trained ML model was used to predict unknown intermediate values for estimating the over-learning and over-fitting problems here for 2 and 0.25 mm thick sheets and are validated experimentally. The variation between the FLDs of predicted and experimentally verified data falls between 2% and 5%. Such small changes in the FLD values show that the proposed ML model could be used to predict the FLDs of copper strips.

The Design and Optimization of an Acoustic and Ambient Sensing AIoT Platform for Agricultural Applications.

The use of technology in agriculture has been gaining significant attention recently. By employing advanced tools and automation and leveraging the latest advancements in the Internet of Things (IoT) and artificial intelligence (AI), the agricultural sector is witnessing improvements in its crop yields and overall efficiency. This paper presents the design and performance analysis of a machine learning (ML) model for agricultural applications involving acoustic sensing. This model is integrated into an efficient Artificial Intelligence of Things (AIoT) platform tailored for agriculture. The model is then used in the design of a communication network architecture and for determining the distribution of the computing load between edge devices and the cloud. The study focuses on the design, analysis, and optimization of AI deployment for reliable classification models in agricultural applications. Both the architectural level and hardware implementation are taken into consideration when designing the radio module and computing unit. Additionally, the study encompasses the design and performance analysis of the hardware used to implement the sensor node specifically developed for sound classification in agricultural applications. The novelty of this work lies in the optimization of the integrated sensor node, which combines the proposed ML model and wireless network, resulting in an agricultural-specific AIoT platform. This co-design enables significant improvements in the performance and efficiency for acoustic and ambient sensing applications.

Multievent localization for loop-based Sagnac sensing system using machine learning.

In optical sensing applications such as pipeline monitoring and intrusion detection systems, accurate localization of the event is crucial for timely and effective response. This paper experimentally demonstrates multievent localization for long perimeter monitoring using a Sagnac interferometer loop sensor and machine learning techniques. The proposed method considers the multievent localization problem as a multilabel multiclassification problem by dividing the optical fiber into 250 segments. A deep neural network (DNN) model is used to predict the likelihood of event occurrence in each segment and accurately locate the events. The sensing loop comprises 106.245 km of single-mode fiber, equivalent to ∼50 km of effective sensing distance. The training dataset is constructed in simulation using VPItransmissionMaker, and the proposed machine learning model's complexity is reduced by using discrete cosine transform (DCT). The designed DNN is tested for event localization in both simulation and experiment. The simulation results show that the proposed model achieves an accuracy of 99% in predicting the location of one event within one segment error, an accuracy of 95% in predicting the location of one event out of the two within one segment error, and an accuracy of 78% in predicting the location of the two events within one segment error. The experimental results validate the simulation ones, demonstrating the proposed model's effectiveness in accurately localizing events with high precision. In addition, the paper includes a discussion on extending the proposed model to sense more than two events simultaneously.

Evaluating the accuracy and effectiveness of machine learning methods for rapidly determining the safety factor of road embankments

PurposeSlope stability analysis is essential for ensuring the safe design of road embankments. While various conventional methods, such as the finite element approach, are used to determine the safety factor of road embankments, there is ongoing interest in exploring the potential of machine learning techniques for this purpose.Design/methodology/approachWithin the study context, the outcomes of the ensemble machine learning models will be compared and benchmarked against the conventional techniques used to predict this parameter.FindingsGenerally, the study results have shown that the proposed machine learning models provide rapid and accurate estimates of the safety factor of road embankments and are, therefore, promising alternatives to traditional methods.Originality/valueAlthough machine learning algorithms hold promise for rapidly and accurately estimating the safety factor of road embankments, few studies have systematically compared their performance with traditional methods. To address this gap, this study introduces a novel approach using advanced ensemble machine learning techniques for efficient and precise estimation of the road embankment safety factor. Besides, the study comprehensively assesses the performance of these ensemble techniques, in contrast with established methods such as the finite element approach and empirical models, demonstrating their potential as robust and reliable alternatives in the realm of slope stability assessment.