State-of-the-art in artificial neural network applications: A survey

Research on a three-dimensional radiation field reconstruction algorithm based on an improved 3D CNN

This article investigates the application of an improved three-dimensional convolutional neural network (3D CNN) for sparse data-based reconstruction of radiation fields. Sparse radiation data points are consolidated into structured three-dimensional matrices and fed into a self-attention integrated CNN, enabling the network to interpolate and produce complete radiation distribution grids. The model's validity is assessed through experiments with randomly sourced radiation in scenarios both with and without shielding, as well as in refined grid configurations. Results indicate that in unshielded environments, a mere 5%(15 points) sampling yields an average relative error of 4%, while in shielded settings, a 7% (21 points) sampling maintains the error around 11%. In refined grid contexts, a 2% sampling rate suffices to limit the error to 6.58%. Thus, the improved 3D CNN is demonstrated to be highly effective for precise three-dimensional radiation field reconstruction in sparse data scenarios.

The development of honey recognition models with broad applicability based on the association of isotope and elemental content with ANNs

Honey adulteration represents a worldwide problem, driven by the illicit economic gain that producers, traders, or merchants pursue. Among the falsification methods that can unfairly influence the price is the incorrect declaration of the botanical origin and harvesting year. Therefore, the present study aimed to test the potential given by the application of Artificial Neural Networks (ANNs) for developing prediction models able to assess honey botanical origin and harvesting year based on isotope and elemental fingerprints. For each classification criterion, significant focus was dedicated to the data preprocessing phase to enhance the models' prediction capability. The obtained classification performances (accuracy scores >86% during the test phase) have highlighted the efficiency of ANNs for honey authentication as well as the feasibility of applying the developed classifiers for a large-scale application, supported by their ability to recognize the correct origin despite considerable variability in botanical source, geographical origin, and harvesting period.

Machine learning-aided hybrid technique for dynamics of rail transit stations classification: a case study

Accurate classification of rail transit stations is crucial for successful Transit-Oriented Development (TOD) and sustainable urban growth. This paper introduces a novel classification model integrating traditional methodologies with advanced machine learning algorithms. By employing mathematical models, clustering methods, and neural network techniques, the model enhances the precision of station classification, allowing for a refined evaluation of station attributes. A comprehensive case study on the Chengdu rail transit network validates the model's efficacy, highlighting its value in optimizing TOD strategies and guiding decision-making processes for urban planners and policymakers. The study employs several regression models trained on existing data to generate accurate ridership forecasts, and data clustering using mathematical algorithms reveals distinct categories of stations. Evaluation metrics confirm the rationality and accuracy of the results. Additionally, a neural network achieving high accuracy on labeled data enhances the model's predictive capabilities for unlabeled instances. The research demonstrates high accuracy, with the Mean Squared Error (MSE) for regression models (Multiple Linear Regression (MLR), Deep-Learning Neural Network (DNN), and K-Nearest Neighbor (KNN)) remaining below 0.012, while the neural networks used for station classification achieve 100% accuracy across seven time intervals and 98.15% accuracy for the eighth, ensuring reliable ridership forecasts and classification outcomes. Accuracy in rail transit station classification is critical, as it not only strengthens the model's predictive capabilities but also ensures more reliable data-driven decisions for transit planning and development, allowing for more precise ridership forecasts and evidence-based strategies for optimizing TOD. This classification model provides stakeholders with valuable insights into the dynamics and features of rail transit stations, supporting sustainable urban development planning.

A neural network paradigm for modeling psychometric data and estimating IRT model parameters: Cross estimation network

This paper presents a novel approach known as the cross estimation network (CEN) for fitting the datasets obtained from psychological or educational tests and estimating the parameters of item response theory (IRT) models. The CEN is comprised of two subnetworks: the person network (PN) and the item network (IN). The PN processes the response pattern of individual respondent and generates an estimate of the underlying ability, while the IN takes in the response pattern of individual item and outputs the estimates of the item parameters. Four simulation studies and an empirical study were comprehensively and rigorously conducted to investigate the performance of CEN on parameter estimation of the two-parameter logistic model under various testing scenarios. Results showed that CEN effectively fit the training data and produced accurate estimates of both person and item parameters. The trained PN and IN adhered to AI principles and acted as intelligent agents, delivering commendable evaluations for even unseen patterns of new respondents and items.

Smart solutions in hypertension diagnosis and management: a deep dive into artificial intelligence and modern wearables for blood pressure monitoring

Hypertension, a widespread cardiovascular issue, presents a major global health challenge. Traditional diagnosis and treatment methods involve periodic blood pressure monitoring and prescribing antihypertensive drugs. Smart technology integration in healthcare offers promising results in optimizing the diagnosis and treatment of various conditions. We investigate its role in improving hypertension diagnosis and treatment effectiveness using machine learning algorithms for early and accurate detection. Intelligent models trained on diverse datasets (encompassing physiological parameters, lifestyle factors, and genetic information) to detect subtle hypertension risk patterns. Adaptive algorithms analyze patient-specific data, optimizing treatment plans based on medication responses and lifestyle habits. This personalized approach ensures effective, minimally invasive interventions tailored to each patient. Wearables and smart sensors provide real-time health insights for proactive treatment adjustments and early complication detection.

Diagnostic accuracy of deep learning using speech samples in depression: a systematic review and meta-analysis

OBJECTIVE

This study aims to conduct a systematic review and meta-analysis of the diagnostic accuracy of deep learning (DL) using speech samples in depression.

MATERIALS AND METHODS

This review included studies reporting diagnostic results of DL algorithms in depression using speech data, published from inception to January 31, 2024, on PubMed, Medline, Embase, PsycINFO, Scopus, IEEE, and Web of Science databases. Pooled accuracy, sensitivity, and specificity were obtained by random-effect models. The diagnostic Precision Study Quality Assessment Tool (QUADAS-2) was used to assess the risk of bias.

RESULTS

A total of 25 studies met the inclusion criteria and 8 of them were used in the meta-analysis. The pooled estimates of accuracy, specificity, and sensitivity for depression detection models were 0.87 (95% CI, 0.81-0.93), 0.85 (95% CI, 0.78-0.91), and 0.82 (95% CI, 0.71-0.94), respectively. When stratified by model structure, the highest pooled diagnostic accuracy was 0.89 (95% CI, 0.81-0.97) in the handcrafted group.

DISCUSSION

To our knowledge, our study is the first meta-analysis on the diagnostic performance of DL for depression detection from speech samples. All studies included in the meta-analysis used convolutional neural network (CNN) models, posing problems in deciphering the performance of other DL algorithms. The handcrafted model performed better than the end-to-end model in speech depression detection.

CONCLUSIONS

The application of DL in speech provided a useful tool for depression detection. CNN models with handcrafted acoustic features could help to improve the diagnostic performance.

PROTOCOL REGISTRATION

The study protocol was registered on PROSPERO (CRD42023423603).

Role of Machine Learning Assisted Biosensors in Point-of-Care-Testing For Clinical Decisions

Point-of-Care-Testing (PoCT) has emerged as an essential component of modern healthcare, providing rapid, low-cost, and simple diagnostic options. The integration of Machine Learning (ML) into biosensors has ushered in a new era of innovation in the field of PoCT. This article investigates the numerous uses and transformational possibilities of ML in improving biosensors for PoCT. ML algorithms, which are capable of processing and interpreting complicated biological data, have transformed the accuracy, sensitivity, and speed of diagnostic procedures in a variety of healthcare contexts. This review explores the multifaceted applications of ML models, including classification and regression, displaying how they contribute to improving the diagnostic capabilities of biosensors. The roles of ML-assisted electrochemical sensors, lab-on-a-chip sensors, electrochemiluminescence/chemiluminescence sensors, colorimetric sensors, and wearable sensors in diagnosis are explained in detail. Given the increasingly important role of ML in biosensors for PoCT, this study serves as a valuable reference for researchers, clinicians, and policymakers interested in understanding the emerging landscape of ML in point-of-care diagnostics.

Combining the probabilistic finite element model and artificial neural network to study nutrient levels in the human intervertebral discs

BACKGROUND

Diffusion distance and diffusivity are known to affect nutrient transport rates, but the probabilistic analysis of these two factors remains vacant. There is a lack of effective tools to evaluate disc nutrient levels.

METHODS

Five-hundred-disc samples with different combinations of morphological and water content parameters were generated, which were used to evaluate nutrient levels in unloaded and loaded states. Spearman correlation coefficients between inputs and responses were calculated. Artificial neural networks were trained to predict nutrient concentrations based on the dataset generated by the probabilistic finite element model.

FINDINGS

In unloaded and loaded states, the minimum oxygen concentration of nucleus pulposus was negatively correlated with disc height (r = -0.83, p < 0.01 and r = -0.76, p < 0.01, respectively), and the minimum glucose concentration of annulus fibrosus was positively correlated with its water content (r = 0.68, p < 0.01 and r = 0.73, p < 0.01, respectively). The maximum lactate concentration of cartilage endplate was affected by endplate thickness (r = 0.94, p < 0.01 and r = 0.95, p < 0.01, respectively). For trained neural networks, nutrient concentrations could be well predicted, with coefficients of determination greater than 0.95 and mean absolute percentage errors less than 5 %.

INTERPRETATION

This study underscores the importance of disc height, annulus fibrosus water content, and endplate thickness in regulating nutrient levels, and precise control of these parameters should be prioritized in the design of tissue-engineered discs. Moreover, artificial neural networks might be a promising tool for evaluating nutrient levels.

Evaluation of Rehabilitation Outcomes in Patients with Chronic Neurological Health Conditions Using a Machine Learning Approach

Background: Over one billion people worldwide suffer from neurological conditions that cause mobility impairments, often persisting despite rehabilitation. Chronic neurological disease (CND) patients who lack access to continuous rehabilitation face gradual functional decline. The International Classification of Functioning, Disability, and Health (ICF) provides a comprehensive framework for assessing these patients. Objective: This study aims to evaluate the outcomes of a non-hospitalized neuromotor rehabilitation project for CND patients in Italy using the Barthel Index (BI) as the primary outcome measure. The rehabilitation was administered through an Individual Rehabilitation Plan (IRP), tailored by a multidisciplinary team and coordinated by a physiatrist. The IRP involved an initial comprehensive assessment, individualized therapy administered five days a week, and continuous adjustments based on patient progress. The secondary objectives include assessing mental status and sensory and communication functions, and identifying predictive factors for BI improvement using an artificial neural network (ANN). Methods: A retrospective observational study of 128 CND patients undergoing a rehabilitation program between 2018 and 2023 was conducted. Variables included demographic data, clinical assessments (BI, SPMSQ, and SVaMAsc), and ICF codes. Data were analyzed using descriptive statistics, linear regressions, and ANN to identify predictors of BI improvement. Results: Significant improvements in the mean BI score were observed from admission (40.28 ± 29.08) to discharge (42.53 ± 30.02, p < 0.001). Patients with severe mobility issues showed the most difficulty in transfers and walking, as indicated by the ICF E codes. Females, especially older women, experienced more cognitive decline, affecting rehabilitation outcomes. ANN achieved 86.4% accuracy in predicting BI improvement, with key factors including ICF mobility codes and the number of past rehabilitation projects. Conclusions: The ICF mobility codes are strong predictors of BI improvement in CND patients. More rehabilitation sessions and targeted support, especially for elderly women and patients with lower initial BI scores, can enhance outcomes and reduce complications. Continuous rehabilitation is essential for maintaining progress in CND patients.

Flexible Wearable Chemoresistive Ethylene Gas-Monitoring Device Utilizing Pd/Ti3C2Tx Nanocomposites for In Situ Nondestructive Monitoring of Kiwifruit Ripeness

Kiwifruit, renowned for its antioxidant properties and nutritional richness, faces challenges in maintaining quality during transportation, often leading to suboptimal products reaching the market. To address this issue, a wireless transmission flexible ethylene monitoring device (WFEMD) was developed. This device comprises a flexible ethylene gas sensor and a signal transmission processing unit integrated with electronic components, enabling real-time monitoring capabilities. In this study, the catalytic activity of Pd and Pd/Ti heterojunctions was leveraged to enhance the ethylene gas sensing. The impact of Ti3C2Tx modified with varying masses of Pd nanoparticles on ethylene gas response levels was investigated. The signal transmission processing unit, fabricated by using the laser direct-writing method, was optimized to collect signals from the flexible ethylene gas sensor, convert them into corresponding ethylene concentrations, and transmit data via an antenna. By introducing a random forest (RF) classification algorithm, a remarkable 97.5% accuracy in predicting kiwifruit ripeness grades was achieved. The algorithm facilitated precise classification by collecting key parameters such as ethylene and CO2 during transportation. The WFEMD enables real-time acquisition of kiwifruit ethylene gas information, which is transmitted wirelessly for data visualization and traceability via mobile terminals. This empowers managers with timely insights into ethylene emissions and ripeness predictions, facilitating informed decision-making processes.

Screening depression among university students utilizing GHQ-12 and machine learning

The escalating incidence of depression has brought attention to the increasing concern regarding the mental well-being of university students in the current academic environment. Given the increasing mental health challenges faced by students, there is a critical need for efficient, scalable, and accurate screening methods. This study aims to address the issue by using the General Health Questionnaire-12 (GHQ-12), a well recognized tool for evaluating psychological discomfort, in combination with machine learning (ML) techniques. Firstly, for effective screening of depression, a comprehensive questionnaire has been created with the help of an expert psychiatrist. The questionnaire includes the GHQ-12, socio-demographic, and job and career-related inquiries. A total of 804 responses has been collected from various public and private universities across Bangladesh. The data has been then analyzed and preprocessed. It has been found that around 60% of the study population are suffering from depression. Lastly, 16 different ML models, including both traditional algorithms and ensemble methods has been applied to examine the data to identify trends and predictors of depression in this demographic. The models' performance has been rigorously evaluated in order to ascertain their effectiveness in precisely identifying individuals who are at risk. Among the ML models, Extremely Randomized Tree (ET) has achieved the highest accuracy of 90.26%, showcasing its classification effectiveness. A thorough investigation of the performance of the models compared, therefore clarifying their possible relevance in the early detection of depression among university students, has been presented in this paper. The findings shed light on the complex interplay among socio-demographic variables, stressors associated with one's profession, and mental well-being, which offer an original viewpoint on utilizing ML in psychological research.

Classification of Alzheimer disease using DenseNet-201 based on deep transfer learning technique

Alzheimer's disease (AD) is a brain illness that causes gradual memory loss. AD has no treatment and cannot be cured, so early detection is critical. Various AD diagnosis approaches are used in this regard, but Magnetic Resonance Imaging (MRI) provides the most helpful neuroimaging tool for detecting AD. In this paper, we employ a DenseNet-201 based transfer learning technique for diagnosing different Alzheimer's stages as Non-Demented (ND), Moderate Demented (MOD), Mild Demented (MD), Very Mild Demented (VMD), and Severe Demented (SD). The suggested method for a dataset of MRI scans for Alzheimer's disease is divided into five classes. Data augmentation methods were used to expand the size of the dataset and increase DenseNet-201's accuracy. It was found that the proposed strategy provides a very high classification accuracy. This practical and reliable model delivers a success rate of 98.24%. The findings of the experiments demonstrate that the suggested deep learning approach is more accurate and performs well compared to existing techniques and state-of-the-art methods.

Graph Embedded Intuitionistic Fuzzy Random Vector Functional Link Neural Network for Class Imbalance Learning

The domain of machine learning is confronted with a crucial research area known as class imbalance (CI) learning, which presents considerable hurdles in the precise classification of minority classes. This issue can result in biased models where the majority class takes precedence in the training process, leading to the underrepresentation of the minority class. The random vector functional link (RVFL) network is a widely used and effective learning model for classification due to its good generalization performance and efficiency. However, it suffers when dealing with imbalanced datasets. To overcome this limitation, we propose a novel graph-embedded intuitionistic fuzzy RVFL for CI learning (GE-IFRVFL-CIL) model incorporating a weighting mechanism to handle imbalanced datasets. The proposed GE-IFRVFL-CIL model offers a plethora of benefits: 1) leveraging graph embedding (GE) to preserve the inherent topological structure of the datasets; 2) employing intuitionistic fuzzy (IF) theory to handle uncertainty and imprecision in the data; and 3) the most important, it tackles CI learning. The amalgamation of a weighting scheme, GE, and IF sets leads to the superior performance of the proposed models on KEEL benchmark imbalanced datasets with and without Gaussian noise. Furthermore, we implemented the proposed GE-IFRVFL-CIL on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset and achieved promising results, demonstrating the model's effectiveness in real-world applications. The proposed GE-IFRVFL-CIL model offers a promising solution to address the CI issue, mitigates the detrimental effect of noise and outliers, and preserves the inherent geometrical structures of the dataset.

Optimized deep CNN for detection and classification of diabetic retinopathy and diabetic macular edema

Diabetic Retinopathy (DR) and Diabetic Macular Edema (DME) are vision related complications prominently found in diabetic patients. The early identification of DR/DME grades facilitates the devising of an appropriate treatment plan, which ultimately prevents the probability of visual impairment in more than 90% of diabetic patients. Thereby, an automatic DR/DME grade detection approach is proposed in this work by utilizing image processing. In this work, the retinal fundus image provided as input is pre-processed using Discrete Wavelet Transform (DWT) with the aim of enhancing its visual quality. The precise detection of DR/DME is supported further with the application of suitable Artificial Neural Network (ANN) based segmentation technique. The segmented images are subsequently subjected to feature extraction using Adaptive Gabor Filter (AGF) and the feature selection using Random Forest (RF) technique. The former has excellent retinal vein recognition capability, while the latter has exceptional generalization capability. The RF approach also assists with the improvement of classification accuracy of Deep Convolutional Neural Network (CNN) classifier. Moreover, Chicken Swarm Algorithm (CSA) is used for further enhancing the classifier performance by optimizing the weights of both convolution and fully connected layer. The entire approach is validated for its accuracy in determination of grades of DR/DME using MATLAB software. The proposed DR/DME grade detection approach displays an excellent accuracy of 97.91%.

Novel authentication of African geographical coffee types (bean, roasted, powdered) by handheld NIR spectroscopic method

African coffee is among the best traded coffee types worldwide, and rapid identification of its geographical origin is very important when trading the commodity. The study was important because it used NIR techniques to geographically differentiate between various types of coffee and provide a supply chain traceability method to avoid fraud. In this study, geographic differentiation of African coffee types (bean, roasted, and powder) was achieved using handheld near-infrared spectroscopy and multivariant data processing. Five African countries were used as the origins for the collection of Robusta coffee. The samples were individually scanned at a wavelength of 740-1070 nm, and their spectra profiles were preprocessed with mean centering (MC), multiplicative scatter correction (MSC), and standard normal variate (SNV). Support vector machines (SVM), linear discriminant analysis (LDA), neural networks (NN), random forests (RF), and partial least square discriminate analysis (PLS-DA) were then used to develop a prediction model for African coffee types. The performance of the model was assessed using accuracy and F1-score. Proximate chemical composition was also conducted on the raw and roasted coffee types. The best classification algorithms were developed for the following coffee types: raw bean coffee, SD-PLSDA, and MC + SD-PLSDA. These models had an accuracy of 0.87 and an F1-score of 0.88. SNV + SD-SVM and MSC + SD-NN both had accuracy and F1 scores of 0.97 for roasted coffee beans and 0.96 for roasted coffee powder, respectively. The results revealed that efficient quality assurance may be achieved by using handheld NIR spectroscopy combined with chemometrics to differentiate between different African coffee types according to their geographical origins.

The Framework of Quantifying Biomarkers of OCT and OCTA Images in Retinal Diseases

Despite the significant advancements facilitated by previous research in introducing a plethora of retinal biomarkers, there is a lack of research addressing the clinical need for quantifying different biomarkers and prioritizing their importance for guiding clinical decision making in the context of retinal diseases. To address this issue, our study introduces a novel framework for quantifying biomarkers derived from optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA) images in retinal diseases. We extract 452 feature parameters from five feature types, including local binary patterns (LBP) features of OCT and OCTA, capillary and large vessel features, and the foveal avascular zone (FAZ) feature. Leveraging this extensive feature set, we construct a classification model using a statistically relevant p value for feature selection to predict retinal diseases. We obtain a high accuracy of 0.912 and F1-score of 0.906 in the task of disease classification using this framework. We find that OCT and OCTA's LBP features provide a significant contribution of 77.12% to the significance of biomarkers in predicting retinal diseases, suggesting their potential as latent indicators for clinical diagnosis. This study employs a quantitative analysis framework to identify potential biomarkers for retinal diseases in OCT and OCTA images. Our findings suggest that LBP parameters, skewness and kurtosis values of capillary, the maximum, mean, median, and standard deviation of large vessel, as well as the eccentricity, compactness, flatness, and anisotropy index of FAZ, may serve as significant indicators of retinal conditions.

Prediction and analysis of dominant factors influencing moisture content during vacuum screening based on machine learning

The study of the dominant factors influencing moisture content is essential for investigating vacuum filtration mechanisms. In view of the present situation where there is insufficient experimental data and the dominant factors influencing the moisture content of a filter cake have not been identified, in this study a vacuum filtration apparatus was designed and constructed. Quartz sand particles were used as the filtration material. 300 datasets of moisture contents of a filter cake were obtained under different experimental conditions. Multiple Linear Regression, artificial neural network, decision tree, random forest, and extreme gradient boosting were used to establish a prediction model for moisture content during vacuum screening. By comprehensively analyzing the feature importance rankings and the effects of positive and negative correlations, the dominant factors influencing the moisture content of the filter cake during vacuum screening were the particle ratio, screen mesh, and airflow rate. This finding not only provides a scientific basis for the optimization of vacuum screening technology but also points the way for improving screening efficiency in practical applications. It is of significant importance for deepening the understanding of the vacuum screening mechanism and promoting its extensive application.

Machine learning application in otology

This review presents a comprehensive history of Artificial Intelligence (AI) in the context of the revolutionary application of machine learning (ML) to medical research and clinical utilization, particularly for the benefit of researchers interested in the application of ML in otology. To this end, we discuss the key components of ML-input, output, and algorithms. In particular, some representation algorithms commonly used in medical research are discussed. Subsequently, we review ML applications in otology research, including diagnosis, influential identification, and surgical outcome prediction. In the context of surgical outcome prediction, specific surgical treatments, including cochlear implantation, active middle ear implantation, tympanoplasty, and vestibular schwannoma resection, are considered. Finally, we highlight the obstacles and challenges that need to be overcome in future research.

Artificial neural network to predict post-operative hypocalcemia following total thyroidectomy

The primary objective of this study was to use artificial neural network (ANN) to predict the post operative hypocalcemia and severity of hypocalcemia following total thyroidectomy. The secondary objective was to determine the weightage for the factors predicting the hypocalcemia with the ANN. A single center, retrospective case series included treatment-naive patients undergoing total thyroidectomy for benign or malignant thyroid nodules from January 2020 to December 2022. Artificial neural network (ANN) - Multilayer Perceptron (MLP) used to predict post-operative hypocalcemia in ANN. Multivariate analysis was used construct validity. The data of 196 total thyroidectomy cases was used for training and testing. The mean incorrect prediction during training and testing was 3.18% (± σ = 0.65%) and 3.66% (± σ = 1.88%) for hypocalcemia. The cumulative Root-Mean-Square-Error (RMSE) for MLP model was 0.29 (± σ = 0.02) and 0.32 (± σ = 0.04) for training and testing, respectively. Area under ROC was 0.98 for predicting hypocalcemia 0.942 for predicting the severity of hypocalcemia. Multivariate analysis showed lower levels of post operative parathormone levels to be predictor of hypocalcemia (p < 0.01). The maximum weightage given to iPTH (100%) > Need for sternotomy (28.55%). Our MLP NN model predicted the post-operative hypocalcemia in 96.8% of training samples and 96.3% of testing samples, and severity in 92.8% of testing sample in 10 generations. however, it must be used with caution and always in conjunction with the expertise of the surgical team. Level of Evidence - 3b.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12070-024-04608-9.

Accurate prediction and intelligent control of COD and other parameters removal from pharmaceutical wastewater using electrocoagulation coupled with catalytic ozonation process

In this study, we employed the response surface method (RSM) and the long short-term memory (LSTM) model to optimize operational parameters and predict chemical oxygen demand (COD) removal in the electrocoagulation-catalytic ozonation process (ECOP) for pharmaceutical wastewater treatment. Through RSM simulation, we quantified the effects of reaction time, ozone dose, current density, and catalyst packed rate on COD removal. Then, the optimal conditions for achieving a COD removal efficiency exceeding 50% were identified. After evaluating ECOP performance under optimized conditions, LSTM predicted COD removal (56.4%), close to real results (54.6%) with a 0.2% error. LSTM outperformed RSM in predictive capacity for COD removal. In response to the initial COD concentration and effluent discharge standards, intelligent adjustment of operating parameters becomes feasible, facilitating precise control of the ECOP performance based on this LSTM model. This intelligent control strategy holds promise for enhancing the efficiency of ECOP in real pharmaceutical wastewater treatment scenarios. PRACTITIONER POINTS: This study utilized the response surface method (RSM) and the long short-term memory (LSTM) model for pharmaceutical wastewater treatment optimization. LSTM predicted COD removal (56.4%) closely matched experimental results (54.6%), with a minimal error of 0.2%. LSTM demonstrated superior predictive capacity, enabling intelligent parameter adjustments for enhanced process control. Intelligent control strategy based on LSTM holds promise for improving electrocoagulation-catalytic ozonation process efficiency in pharmaceutical wastewater treatment.

Taha M, Zhang Y. DeepDate: A deep fusion model based on whale optimization and artificial neural network for Arabian date classification

PURPOSE

As agricultural technology continues to develop, the scale of planting and production of date fruit is increasing, which brings higher yields. However, the increasing yields also put a lot of pressure on the classification step afterward. Image recognition based on deep learning algorithms can help to identify and classify the date fruit species, even in natural light.

METHOD

In this paper, a deep fusion model based on whale optimization and an artificial neural network for Arabian date classification is proposed. The dataset used in this study includes five classes of date fruit images (Barhi, Khalas, Meneifi, Naboot Saif, Sullaj). The process of designing each model can be divided into three phases. The first phase is feature extraction. The second phase is feature selection. The third phase is the training and testing phase. Finally, the best-performing model was selected and compared with the currently established models (Alexnet, Squeezenet, Googlenet, Resnet50).

RESULTS

The experimental results show that, after trying different combinations of optimization algorithms and classifiers, the highest test accuracy achieved by DeepDate was 95.9%. It takes less time to achieve a balance between classification accuracy and time consumption. In addition, the performance of DeepDate is better than that of many deep transfer learning models such as Alexnet, Squeezenet, Googlenet, VGG-19, NasNet, and Inception-V3.

CONCLUSION

The proposed DeepDate improves the accuracy and efficiency of classifying date fruits and achieves better results in classification metrics such as accuracy and F1. DeepDate provides a promising classification solution for date fruit classification with higher accuracy. To further advance the industry, it is recommended that stakeholders invest in technology transfer programs to bring advanced image recognition and AI tools to smaller producers, enhancing sustainability and productivity across the sector. Collaborations between agricultural technologists and growers could also foster more tailored solutions that address specific regional challenges in date fruit production.

COVIDHealth: A novel labeled dataset and machine learning-based web application for classifying COVID-19 discourses on Twitter

The COVID-19 pandemic has sparked widespread health-related discussions on social media platforms like Twitter (now named 'X'). However, the lack of labeled Twitter data poses significant challenges for theme-based classification and tweet aggregation. To address this gap, we developed a machine learning-based web application that automatically classifies COVID-19 discourses into five categories: health risks, prevention, symptoms, transmission, and treatment. We collected and labeled 6,667 COVID-19-related tweets using the Twitter API, and applied various feature extraction methods to extract relevant features. We then compared the performance of seven classical machine learning algorithms (Decision Tree, Random Forest, Stochastic Gradient Descent, Adaboost, K-Nearest Neighbor, Logistic Regression, and Linear SVC) and four deep learning techniques (LSTM, CNN, RNN, and BERT) for classification. Our results show that the CNN achieved the highest precision (90.41%), recall (90.4%), F1 score (90.4%), and accuracy (90.4%). The Linear SVC algorithm exhibited the highest precision (85.71%), recall (86.94%), and F1 score (86.13%) among classical machine learning approaches. Our study advances the field of health-related data analysis and classification, and offers a publicly accessible web-based tool for public health researchers and practitioners. This tool has the potential to support addressing public health challenges and enhancing awareness during pandemics. The dataset and application are accessible at https://github.com/Bishal16/COVID19-Health-Related-Data-Classification-Website.

SERS microscopy as a tool for comprehensive biochemical characterization in complex samples

Surface enhanced Raman scattering (SERS) spectra of biomaterials such as cells or tissues can be used to obtain biochemical information from nanoscopic volumes in these heterogeneous samples. This tutorial review discusses the factors that determine the outcome of a SERS experiment in complex bioorganic samples. They are related to the SERS process itself, the possibility to selectively probe certain regions or constituents of a sample, and the retrieval of the vibrational information in order to identify molecules and their interaction. After introducing basic aspects of SERS experiments in the context of biocompatible environments, spectroscopy in typical microscopic settings is exemplified, including the possibilities to combine SERS with other linear and non-linear microscopic tools, and to exploit approaches that improve lateral and temporal resolution. In particular the great variation of data in a SERS experiment calls for robust data analysis tools. Approaches will be introduced that have been originally developed in the field of bioinformatics for the application to omics data and that show specific potential in the analysis of SERS data. They include the use of simulated data and machine learning tools that can yield chemical information beyond achieving spectral classification.

Paired competing neurons improving STDP supervised local learning in Spiking Neural Networks

Direct training of Spiking Neural Networks (SNNs) on neuromorphic hardware has the potential to significantly reduce the energy consumption of artificial neural network training. SNNs trained with Spike Timing-Dependent Plasticity (STDP) benefit from gradient-free and unsupervised local learning, which can be easily implemented on ultra-low-power neuromorphic hardware. However, classification tasks cannot be performed solely with unsupervised STDP. In this paper, we propose Stabilized Supervised STDP (S2-STDP), a supervised STDP learning rule to train the classification layer of an SNN equipped with unsupervised STDP for feature extraction. S2-STDP integrates error-modulated weight updates that align neuron spikes with desired timestamps derived from the average firing time within the layer. Then, we introduce a training architecture called Paired Competing Neurons (PCN) to further enhance the learning capabilities of our classification layer trained with S2-STDP. PCN associates each class with paired neurons and encourages neuron specialization toward target or non-target samples through intra-class competition. We evaluate our methods on image recognition datasets, including MNIST, Fashion-MNIST, and CIFAR-10. Results show that our methods outperform state-of-the-art supervised STDP learning rules, for comparable architectures and numbers of neurons. Further analysis demonstrates that the use of PCN enhances the performance of S2-STDP, regardless of the hyperparameter set and without introducing any additional hyperparameters.

Machine learning-based prediction of cadmium pollution in topsoil and identification of critical driving factors in a mining area

Mining activities have resulted in a substantial accumulation of cadmium (Cd) in agricultural soils, particularly in southern China. Long-term Cd exposure can cause plant growth inhibition and various diseases. Rapid identification of the extent of soil Cd pollution and its driving factors are essential for soil management and risk assessment. However, traditional geostatistical methods are difficult to simulate the complex nonlinear relationships between soil Cd and potential features. In this study, sequential extraction and hotspot analyses indicated that Cd accumulation increased significantly near mining sites and exhibited high mobility. The concentration of Cd was estimated using three machine learning models based on 3169 topsoil samples, seven quantitative variables (soil pH, Fe, Ca, Mn, TOC, Al/Si and ba value) and three quantitative variables (soil parent rock, terrain and soil type). The random forest model achieved marginally better performance than the other models, with an R2 of 0.78. Importance analysis revealed that soil pH and Ca and Mn contents were the most significant factors affecting Cd accumulation and migration. Conversely, due to the essence of controlling Cd migration being soil property, soil type, terrain, and soil parent materials had little impact on the spatial distribution of soil Cd under the influence of mining activities. Our results provide a better understanding of the geochemical behavior of soil Cd in mining areas, which could be helpful for environmental management departments in controlling the diffusion of Cd pollution and capturing key targets for soil remediation.

Automatic classification of acute lymphoblastic leukemia cells and lymphocyte subtypes based on a novel convolutional neural network

Acute lymphoblastic leukemia (ALL) is a life-threatening disease that commonly affects children and is classified into three subtypes: L1, L2, and L3. Traditionally, ALL is diagnosed through morphological analysis, involving the examination of blood and bone marrow smears by pathologists. However, this manual process is time-consuming, laborious, and prone to errors. Moreover, the significant morphological similarity between ALL and various lymphocyte subtypes, such as normal, atypic, and reactive lymphocytes, further complicates the feature extraction and detection process. The aim of this study is to develop an accurate and efficient automatic system to distinguish ALL cells from these similar lymphocyte subtypes without the need for direct feature extraction. First, the contrast of microscopic images is enhanced using histogram equalization, which improves the visibility of important features. Next, a fuzzy C-means clustering algorithm is employed to segment cell nuclei, as they play a crucial role in ALL diagnosis. Finally, a novel convolutional neural network (CNN) with three convolutional layers is utilized to classify the segmented nuclei into six distinct classes. The CNN is trained on a labeled dataset, allowing it to learn the distinguishing features of each class. To evaluate the performance of the proposed model, quantitative metrics are employed, and a comparison is made with three well-known deep networks: VGG-16, DenseNet, and Xception. The results demonstrate that the proposed model outperforms these networks, achieving an approximate accuracy of 97%. Moreover, the model's performance surpasses that of other studies focused on 6-class classification in the context of ALL diagnosis. RESEARCH HIGHLIGHTS: Deep neural networks eliminate the requirement for feature extraction in ALL classification The proposed convolutional neural network achieves an impressive accuracy of approximately 97% in classifying six ALL and lymphocyte subtypes.

Graph Embedded Ensemble Deep Randomized Network for Diagnosis of Alzheimer's Disease

Randomized shallow/deep neural networks with closed form solution avoid the shortcomings that exist in the back propagation (BP) based trained neural networks. Ensemble deep random vector functional link (edRVFL) network utilize the strength of two growing fields, i.e., deep learning and ensemble learning. However, edRVFL model doesn't consider the geometrical relationship of the data while calculating the final output parameters corresponding to each layer considered as base model. In the literature, graph embedded frameworks have been successfully used to describe the geometrical relationship within data. In this paper, we propose an extended graph embedded RVFL (EGERVFL) model that, unlike standard RVFL, employs both intrinsic and penalty subspace learning (SL) criteria under the graph embedded framework in its optimization process to calculate the model's output parameters. The proposed shallow EGERVFL model has only single hidden layer and hence, has less representation learning. Therefore, we further develop an ensemble deep EGERVFL (edEGERVFL) model that can be considered a variant of edRVFL model. Unlike edRVFL, the proposed edEGERVFL model solves graph embedded based optimization problem in each layer and hence, has better generalization performance than edRVFL model. We evaluated the proposed approaches for the diagnosis of Alzheimer's disease and furthermore on UCI datasets. The experimental results demonstrate that the proposed models perform better than baseline models. The source code of the proposed models is available at https://github.com/mtanveer1/.

Imaging at the nexus: how state of the art imaging techniques can enhance our understanding of cancer and fibrosis

Both cancer and fibrosis are diseases involving dysregulation of cell signaling pathways resulting in an altered cellular microenvironment which ultimately leads to progression of the condition. The two disease entities share common molecular pathophysiology and recent research has illuminated the how each promotes the other. Multiple imaging techniques have been developed to aid in the early and accurate diagnosis of each disease, and given the commonalities between the pathophysiology of the conditions, advances in imaging one disease have opened new avenues to study the other. Here, we detail the most up-to-date advances in imaging techniques for each disease and how they have crossed over to improve detection and monitoring of the other. We explore techniques in positron emission tomography (PET), magnetic resonance imaging (MRI), second generation harmonic Imaging (SGHI), ultrasound (US), radiomics, and artificial intelligence (AI). A new diagnostic imaging tool in PET/computed tomography (CT) is the use of radiolabeled fibroblast activation protein inhibitor (FAPI). SGHI uses high-frequency sound waves to penetrate deeper into the tissue, providing a more detailed view of the tumor microenvironment. Artificial intelligence with the aid of advanced deep learning (DL) algorithms has been highly effective in training computer systems to diagnose and classify neoplastic lesions in multiple organs. Ultimately, advancing imaging techniques in cancer and fibrosis can lead to significantly more timely and accurate diagnoses of both diseases resulting in better patient outcomes.

Evaluation of a Voltametric E-Tongue Combined with Data Preprocessing for Fast and Effective Machine Learning-Based Classification of Tomato Purées by Cultivar

The potential of a voltametric E-tongue coupled with a custom data pre-processing stage to improve the performance of machine learning techniques for rapid discrimination of tomato purées between cultivars of different economic value has been investigated. To this aim, a sensor array with screen-printed carbon electrodes modified with gold nanoparticles (GNP), copper nanoparticles (CNP) and bulk gold subsequently modified with poly(3,4-ethylenedioxythiophene) (PEDOT), was developed to acquire data to be transformed by a custom pre-processing pipeline and then processed by a set of commonly used classifiers. The GNP and CNP-modified electrodes, selected based on their sensitivity to soluble monosaccharides, demonstrated good ability in discriminating samples of different cultivars. Among the different data analysis methods tested, Linear Discriminant Analysis (LDA) proved to be particularly suitable, obtaining an average F1 score of 99.26%. The pre-processing stage was beneficial in reducing the number of input features, decreasing the computational cost, i.e., the number of computing operations to be performed, of the entire method and aiding future cost-efficient hardware implementation. These findings proved that coupling the multi-sensing platform featuring properly modified sensors with the custom pre-processing method developed and LDA provided an optimal tradeoff between analytical problem solving and reliable chemical information, as well as accuracy and computational complexity. These results can be preliminary to the design of hardware solutions that could be embedded into low-cost portable devices.

We got nuts! use deep neural networks to classify images of common edible nuts

BACKGROUND

Nuts are nutrient-dense foods that contribute to healthier eating. Food image datasets enable artificial intelligence (AI) powered diet-tracking apps to help people monitor daily eating patterns.

AIM

This study aimed to create an image dataset of commonly consumed nut types and use it to build an AI computer vision model to automate nut type classification tasks.

METHODS

iPhone 11 was used to take photos of 11 nut types-almond, brazil nut, cashew, chestnut, hazelnut, macadamia, peanut, pecan, pine nut, pistachio, and walnut. The dataset contains 2200 images, 200 per nut type. The dataset was randomly split into the training (60% or 1320 images), validation (20% or 440 images), and test sets (20% or 440 images). A neural network model was constructed and trained using transfer learning and other computer vision techniques-data augmentation, mixup, normalization, label smoothing, and learning rate optimization.

RESULTS

The trained neural network model correctly predicted 338 out of 440 images (40 per nut type) in the validation set, achieving 99.55% accuracy. Moreover, the model classified the 440 images in the test set with 100% accuracy.

CONCLUSION

This study built a nut image dataset and used it to train a neural network model to classify images by nut type. The model achieved near-perfect accuracy on the validation and test sets, demonstrating the feasibility of automating nut type classification using smartphone photos. Being made open-source, the dataset and model can assist the development of diet-tracking apps that facilitate users' adoption and adherence to a healthy diet.

Detection of honey adulteration using machine learning

Honey adulteration is a growing concern due to its health benefits and high nutritional content. Traditional methods like Melissopalynology are ineffective in detecting adulterated honey. This research presents a comparative study of machine learning algorithms for detecting adulteration in honey. The study uses hyperspectral imaging, a promising tool for food quality assurance, to classify and predict adulteration in honey. The proposed model relies on hyper-spectrum images and improves the accuracy of existing models using hyperparameter tuning. The dataset used includes segmented and pre-processed hyperspectral images of adulterated honey samples. The study found that machine learning and hyperspectral imaging can accurately identify if honey has been adulterated, with over 98% classification accuracy. The results showed that between 5% and 10% of adulterated honey samples are misclassified, with C1 Clover honey being the most frequently misclassified. This study aims to develop an efficient and accurate honey counterfeit detection technology using machine learning technologies such as Artificial Neural Networks (ANN), Support-vector machines (SVM), K Nearest Neighbors, Random Forests, and Decision trees. The proposed model relies on hyper-spectrum images and overcomes generalization to unknown honey types of problems. The dataset used includes segmented and pre-processed hyperspectral images of adulterated honey samples from seven different brands with 12 different botanical origin labels. Feature reduction techniques, such as feature ranking-based feature selection, and autoencoder techniques are employed to classify the botanical origins of honey. The model parameters are enhanced or tuned by the training process, and hyperparameters are adjusted by running the whole training data. The researchers used Python, and well-known algorithms like ANN, SVM, KNN, random forests, and decision trees. The results show that machine learning and hyperspectral imaging can accurately identify if honey has been adulterated, with over 98% classification accuracy.

Numerical Optimization of Variable Blank Holder Force Trajectories in Stamping Process for Multi-Defect Reduction

An intelligent optimization technology was proposed to mitigate prevalent multi-defects, particularly failure, wrinkling, and springback in sheet metal forming. This method combined deep neural networks (DNNs), genetic algorithms (GAs), and Monte Carlo simulation (MCS), collectively as DNN-GA-MCS. Our primary aim was to determine intricate process parameters while elucidating the intricate relationship between processing methodologies and material properties. To achieve this goal, variable blank holder force (VBHF) trajectories were implemented into five sub-stroke steps, facilitating adjustments to the blank holder force via numerical simulations with an oil pan model. The Forming Limit Diagram (FLD) predicted by machine learning algorithms based on the Generalized Incremental Stress State Dependent Damage (GISSMO) model provided a robust framework for evaluating sheet failure dynamics during the stamping process. Numerical results confirmed significant improvements in formed quality: compared with the average value of training sets, the improvements of 18.89%, 13.59%, and 14.26% are achieved in failure, wrinkling, and springback; in the purposed two-segmented mode VBHF case application, the average value of three defects is improved by 12.62%, and the total summation of VBHF is reduced by 14.07%. Statistical methodologies grounded in material flow analysis were applied, accompanied by the proposal of distinctive optimization strategies for the die structure aimed at enhancing material flow efficiency. In conclusion, our advanced methodology exhibits considerable potential to improve sheet metal forming processes, highlighting its significant effect on defect reduction.

Assessing Viscosity in Sustainable Deep Eutectic Solvents and Cosolvent Mixtures: An Artificial Neural Network-Based Molecular Approach

Deep eutectic solvents (DESs) are gaining recognition as environmentally friendly solvent alternatives for diverse chemical processes. Yet, designing DESs tailored to specific applications is a resource-intensive task, which requires an accurate estimation of their physicochemical properties. Among them, viscosity is crucial, as it often dictates a DES's suitability as a solvent. In this study, an artificial neural network (ANN) is introduced to accurately describe the viscosity of DESs and their mixtures with cosolvents. The ANN utilizes molecular parameters derived from σ-profiles, computed using the conductor-like screening model for the real solvent segment activity coefficient (COSMO-SAC). The data set comprises 1891 experimental viscosity measurements for 48 DESs based on choline chloride, encompassing 279 different compositions, along with 1618 data points of DES mixtures with cosolvents as water, methanol, isopropanol, and dimethyl sulfoxide, covering a wide range of viscosity measurements from 0.3862 to 4722 mPa s. The optimal ANN structure for describing the logarithmic viscosity of DESs is configured as 9-19-16-1, achieving an overall average absolute relative deviation of 1.6031%. More importantly, the ANN shows a remarkable extrapolation capacity, as it is capable of predicting the viscosity of systems including solvents (ethanol) and hydrogen bond donors (2,3-butanediol) not considered in the training. The ANN model also demonstrates an extensive applicability domain, covering 94.17% of the entire database. These achievements represent a significant step forward in developing robust, open source, and highly accurate models for DESs using molecular descriptors.

Real-time monitoring and predictive analysis of VOC flux variations in soil vapor: Integrating PID sensing with machine learning for enhanced vapor intrusion forecasts

In the rapidly evolving domain of vapor intrusion (VI) assessments, traditional methodologies encompass detailed groundwater and soil vapor sampling coupled with comprehensive laboratory measurements. These models, blending empirical data, theoretical equations, and site-specific parameters, evaluate VI risks by considering a spectrum of influential factors, from volatile organic compounds (VOC) concentrations in groundwater to nuanced soil attributes. However, the challenge of variability, influenced by dynamic ambient conditions and intricate soil properties, remains. Our study presents an advanced on-site gas sensing station geared towards real-time VOC flux monitoring, enriched with an array of ambient sensors, and spearheaded by the reliable PID sensor for VOC detection. Integrating this dynamic system with machine learning, we developed predictive models, notably the random forest regression, which boasts an R-squared value exceeding 79 % and mean relative error near 0.25, affirming its capability to predict trichloroethylene (TCE) concentrations in soil vapor accurately. By synergizing real-time monitoring and predictive insights, our methodology refines VI risk assessments, equipping communities with proactive, informed decision-making tools and bolstering environmental safety. Implementing these predictive models can simplify monitoring for residents, reducing dependence on specialized systems. Once proven effective, there's potential to repurpose monitoring stations to other VI-prone regions, expanding their reach and benefit. The developed model can leverage weather forecasting data to predict and provide alerts for future VOC events.

Unveiling the potential of specific growth rate control in fed-batch fermentation: bridging the gap between product quantity and quality

During the epoch of sustainable development, leveraging cellular systems for production of diverse chemicals via fermentation has garnered attention. Industrial fermentation, extending beyond strain efficiency and optimal conditions, necessitates a profound understanding of microorganism growth characteristics. Specific growth rate (SGR) is designated as a key variable due to its influence on cellular physiology, product synthesis rates and end-product quality. Despite its significance, the lack of real-time measurements and robust control systems hampers SGR control strategy implementation. The narrative in this contribution delves into the challenges associated with the SGR control and presents perspectives on various control strategies, integration of soft-sensors for real-time measurement and control of SGR. The discussion highlights practical and simple SGR control schemes, suggesting their seamless integration into industrial fermenters. Recommendations provided aim to propose new algorithms accommodating mechanistic and data-driven modelling for enhanced progress in industrial fermentation in the context of sustainable bioprocessing.

Tunable-bias based optical neural network for reinforcement learning in path planning

Owing to the high integration, reconfiguration and strong robustness, Mach-Zehnder interferometers (MZIs) based optical neural networks (ONNs) have been widely considered. However, there are few works adding bias, which is important for neural networks, into the ONNs and systematically studying its effect. In this article, we propose a tunable-bias based optical neural network (TBONN) with one unitary matrix layer, which can improve the utilization rate of the MZIs, increase the trainable weights of the network and has more powerful representational capacity than traditional ONNs. By systematically studying its underlying mechanism and characteristics, we demonstrate that TBONN can achieve higher performance by adding more optical biases to the same side beside the inputted signals. For the two-dimensional dataset, the average prediction accuracy of TBONN with 2 biases (97.1%) is 5% higher than that of TBONN with 0 biases (92.1%). Additionally, utilizing TBONN, we propose a novel optical deep Q network (ODQN) algorithm to complete path planning tasks. By implementing simulated experiments, our ODQN shows competitive performance compared with the conventional deep Q network, but accelerates the computation speed by 2.5 times and 4.5 times for 2D and 3D grid worlds, respectively. Further, a more noticeable acceleration will be obtained when applying TBONN to more complex tasks. Also, we demonstrate the strong robustness of TBONN and the imprecision elimination method by using on-chip training.

Build Deep Neural Network Models to Detect Common Edible Nuts from Photos and Estimate Nutrient Portfolio

Nuts are nutrient-dense foods and can be incorporated into a healthy diet. Artificial intelligence-powered diet-tracking apps may promote nut consumption by providing real-time, accurate nutrition information but depend on data and model availability. Our team developed a dataset comprising 1380 photographs, each in RGB color format and with a resolution of 4032 × 3024 pixels. These images feature 11 types of nuts that are commonly consumed. Each photo includes three nut types; each type consists of 2-4 nuts, so 6-9 nuts are in each image. Rectangular bounding boxes were drawn using a visual geometry group (VGG) image annotator to facilitate the identification of each nut, delineating their locations within the images. This approach renders the dataset an excellent resource for training models capable of multi-label classification and object detection, as it was meticulously divided into training, validation, and test subsets. Utilizing transfer learning in Python with the IceVision framework, deep neural network models were adeptly trained to recognize and pinpoint the nuts depicted in the photographs. The ultimate model exhibited a mean average precision of 0.7596 in identifying various nut types within the validation subset and demonstrated a 97.9% accuracy rate in determining the number and kinds of nuts present in the test subset. By integrating specific nutritional data for each type of nut, the model can precisely (with error margins ranging from 0.8 to 2.6%) calculate the combined nutritional content-encompassing total energy, proteins, carbohydrates, fats (total and saturated), fiber, vitamin E, and essential minerals like magnesium, phosphorus, copper, manganese, and selenium-of the nuts shown in a photograph. Both the dataset and the model have been made publicly available to foster data exchange and the spread of knowledge. Our research underscores the potential of leveraging photographs for automated nut calorie and nutritional content estimation, paving the way for the creation of dietary tracking applications that offer real-time, precise nutritional insights to encourage nut consumption.

Estimation of soil erodibility in Peninsular Malaysia: A case study using multiple linear regression and artificial neural networks

Soil erodibility (K) is an essential component in estimating soil loss indicating the soil's susceptibility to detach and transport. Data Computing and processing methods, such as artificial neural networks (ANNs) and multiple linear regression (MLR), have proven to be helpful in the development of predictive models for natural hazards. The present case study aims to assess the efficiency of MLR and ANN models to forecast soil erodibility in Peninsular Malaysia. A total of 103 samples were collected from various sites and K values were calculated using the Tew equation developed for Malaysian soil. From several extracted parameters, the outcomes of correlation and principal component analysis (PCA) revealed the influencing factors to be used in the development of ANN and MLR models. Based on the correlation and PCA results, two sets of influencing factors were employed to develop predictive models. Two MLR (MLR-1 and MLR-2) models and four neural networks (NN-1, NN-2, NN-3, and NN-4) optimized using Levenberg-Marquardt (LM) and scaled conjugate gradient (SCG) were developed and evaluated. The model performance validation was conducted using the coefficient of determination (R2), mean squared error (MSE), root mean squared error (RMSE), and Nash-Sutcliffe efficiency coefficient (NSE). The analysis showed that ANN models outperformed MLR models. The R2 values of 0.446 (MLR-1), 0.430 (MLR-2), 0.894 (NN-1), 0.855 (NN-2), 0.940 (NN-3), and 0.826 (NN-4); MSE values of 0.0000306 (MLR-1), 0.0000315 (MLR-2), 0.0000158 (NN-1), 0.0000261 (NN-2), 0.0000318 (NN-3), and 0.0000216 (NN-4) suggested the higher accuracy and lower modelling error of ANN models as compared with MLR. This study could provide an empirical basis and methodological support for K factor estimation in the region.

Artificial Intelligence and Agency: Tie-breaking in AI Decision-Making

Determining the agency-status of machines and AI has never been more pressing. As we progress into a future where humans and machines more closely co-exist, understanding hallmark features of agency affords us the ability to develop policy and narratives which cater to both humans and machines. This paper maintains that decision-making processes largely underpin agential action, and that in most instances, these processes yield good results in terms of making good choices. However, in some instances, when faced with two (or more) choices, an agent may find themselves with equal reasons to choose either - thus being presented with a tie. This paper argues that in the event of a tie, the ability to create a voluntarist reason is a hallmark feature of agency, and second, that AI, through current tie-breaking mechanisms does not have this ability, and thus fails at this particular feature of agency.

The immunomodulatory effect of oral NaHCO3 is mediated by the splenic nerve: multivariate impact revealed by artificial neural networks

Stimulation of the inflammatory reflex (IR) is a promising strategy for treating systemic inflammatory disorders. Recent studies suggest oral sodium bicarbonate (NaHCO3) as a potential activator of the IR, offering a safe and cost-effective treatment approach. However, the mechanisms underlying NaHCO3-induced anti-inflammatory effects remain unclear. We investigated whether oral NaHCO3's immunomodulatory effects are mediated by the splenic nerve. Female rats received NaHCO3 or water (H2O) for four days, and splenic immune markers were assessed using flow cytometry. NaHCO3 led to a significant increase (p < 0.05, and/or partial eta squared > 0.06) in anti-inflammatory markers, including CD11bc + CD206 + (M2-like) macrophages, CD3 + CD4 + FoxP3 + cells (Tregs), and Tregs/M1-like ratio. Conversely, proinflammatory markers, such as CD11bc + CD38 + TNFα + (M1-like) macrophages, M1-like/M2-like ratio, and SSChigh/SSClow ratio of FSChighCD11bc + cells, decreased in the spleen following NaHCO3 administration. These effects were abolished in spleen-denervated rats, suggesting the necessity of the splenic nerve in mediating NaHCO3-induced immunomodulation. Artificial neural networks accurately classified NaHCO3 and H2O treatment in sham rats but failed in spleen-denervated rats, highlighting the splenic nerve's critical role. Additionally, spleen denervation independently influenced Tregs, M2-like macrophages, Tregs/M1-like ratio, and CD11bc + CD38 + cells, indicating distinct effects from both surgery and treatment. Principal component analysis (PCA) further supported the separate effects. Our findings suggest that the splenic nerve transmits oral NaHCO3-induced immunomodulatory changes to the spleen, emphasizing NaHCO3's potential as an IR activator with therapeutic implications for a wide spectrum of systemic inflammatory conditions.

Deep learning in structural bioinformatics: current applications and future perspectives

In this review article, we explore the transformative impact of deep learning (DL) on structural bioinformatics, emphasizing its pivotal role in a scientific revolution driven by extensive data, accessible toolkits and robust computing resources. As big data continue to advance, DL is poised to become an integral component in healthcare and biology, revolutionizing analytical processes. Our comprehensive review provides detailed insights into DL, featuring specific demonstrations of its notable applications in bioinformatics. We address challenges tailored for DL, spotlight recent successes in structural bioinformatics and present a clear exposition of DL-from basic shallow neural networks to advanced models such as convolution, recurrent, artificial and transformer neural networks. This paper discusses the emerging use of DL for understanding biomolecular structures, anticipating ongoing developments and applications in the realm of structural bioinformatics.

Generation and Storage of Random Voltage Values via Ring Oscillators Comprising Feedback Field-Effect Transistors

In this study, we demonstrate the generation and storage of random voltage values using a ring oscillator consisting of feedback field-effect transistors (FBFETs). This innovative approach utilizes the logic-in-memory function of FBFETs to extract continuous output voltages from oscillatory cycles. The ring oscillator exhibited uniform probability distributions of 51.6% for logic 0 and 48.4% for logic 1. The generation of analog voltages provides binary random variables that are stored for over 5000 s. This demonstrates the potential of the ring oscillator in advanced physical functions and true random number generator technologies.

A novel optimized neural network model for cyber attack detection using enhanced whale optimization algorithm

Cybersecurity is critical in today's digitally linked and networked society. There is no way to overestimate the importance of cyber security as technology develops and becomes more pervasive in our daily lives. Cybersecurity is essential to people's protection. One type of cyberattack known as "credential stuffing" involves using previously acquired usernames and passwords by attackers to access user accounts on several websites without authorization. This is feasible as a lot of people use the same passwords and usernames on several different websites. Maintaining the security of online accounts requires defence against credential-stuffing attacks. The problems of credential stuffing attacks, failure detection, and prediction can be handled by the suggested EWOA-ANN model. Here, a novel optimization approach known as Enhanced Whale Optimization Algorithm (EWOA) is put on to train the neural network. The effectiveness of the suggested attack identification model has been demonstrated, and an empirical comparison will be carried out with respect to specific security analysis.

Synthetic biological neural networks: From current implementations to future perspectives

Artificial neural networks, inspired by the biological networks of the human brain, have become game-changing computing models in modern computer science. Inspired by their wide scope of applications, synthetic biology strives to create their biological counterparts, which we denote synthetic biological neural networks (SYNBIONNs). Their use in the fields of medicine, biosensors, biotechnology, and many more shows great potential and presents exciting possibilities. So far, many different synthetic biological networks have been successfully constructed, however, SYNBIONN implementations have been sparse. The latter are mostly based on neural networks pretrained in silico and being heavily dependent on extensive human input. In this paper, we review current implementations and models of SYNBIONNs. We briefly present the biological platforms that show potential for designing and constructing perceptrons and/or multilayer SYNBIONNs. We explore their future possibilities along with the challenges that must be overcome to successfully implement a scalable in vivo biological neural network capable of online learning.

Predicting the distribution coefficient of cesium in solid phase groups using machine learning

The migration and retention of radioactive contaminants such as 137Cesium (137Cs) in various environmental media pose significant long-term storage challenges for nuclear waste. The distribution coefficient (Kd) is a critical parameter for assessing the mobility of radioactive contaminants and is influenced by various environmental conditions. This study presents machine-learning models based on the Japan Atomic Energy Agency Sorption Database (JAEA-SDB) to predict the Kd values for Cs in solid phase groups. We used three different machine learning models: random forest (RF), artificial neural network (ANN), and convolutional neural network (CNN). The models were trained on 14 input variables from the JAEA-SDB, including factors such as the Cs concentration, solid-phase properties, and solution conditions, which were preprocessed by normalization and log-transformation. The performances of the models were evaluated using the coefficient of determination (R2) and root mean squared error (RMSE). The RF, ANN, and CNN models achieved R2 values greater than 0.97, 0.86, and 0.88, respectively. We also analyzed the variable importance of RF using an out-of-bag (OOB) and a CNN with an attention module. Our results showed that the environmental media, initial radionuclide concentration, solid phase properties, and solution conditions were significant variables for Kd prediction. Our models accurately predict Kd values for different environmental conditions and can assess the environmental risk by analyzing the behavior of radionuclides in solid phase groups. The results of this study can improve safety analyses and long-term risk assessments related to waste disposal and prevent potential hazards and sources of contamination in the surrounding environment.

Municipal solid waste management for low-carbon transition: A systematic review of artificial neural network applications for trend prediction

Improper municipal solid waste (MSW) management contributes to greenhouse gas emissions, necessitating emissions reduction strategies such as waste reduction, recycling, and composting to move towards a more sustainable, low-carbon future. Machine learning models are applied for MSW-related trend prediction to provide insights on future waste generation or carbon emissions trends and assist the formulation of effective low-carbon policies. Yet, the existing machine learning models are diverse and scattered. This inconsistency poses challenges for researchers in the MSW domain who seek to identify and optimize the machine learning techniques and configurations for their applications. This systematic review focuses on MSW-related trend prediction using the most frequently applied machine learning model, artificial neural network (ANN), while addressing potential methodological improvements for reducing prediction uncertainty. Thirty-two papers published from 2013 to 2023 are included in this review, all applying ANN for MSW-related trend prediction. Observing a decrease in the size of data samples used in studies from daily to annual timescales, the summarized statistics suggest that well-performing ANN models can still be developed with approximately 33 annual data samples. This indicates promising opportunities for modeling macroscale greenhouse gas emissions in future works. Existing literature commonly used the grid search (manual) technique for hyperparameter (e.g., learning rate, number of neurons) optimization and should explore more time-efficient automated optimization techniques. Since there are no one-size-fits-all performance indicators, it is crucial to report the model's predictive performance based on more than one performance indicator and examine its uncertainty. The predictive performance of newly-developed integrated models should also be benchmarked to show performance improvement clearly and promote similar applications in future works. The review analyzed the shortcomings, best practices, and prospects of ANNs for MSW-related trend predictions, supporting the realization of practical applications of ANNs to enhance waste management practices and reduce carbon emissions.

Thin-film transistor for temporal self-adaptive reservoir computing with closed-loop architecture

Reservoir computing is a powerful neural network-based computing paradigm for spatiotemporal signal processing. Recently, physical reservoirs have been explored based on various electronic devices with outstanding efficiency. However, the inflexible temporal dynamics of these reservoirs have posed fundamental restrictions in processing spatiotemporal signals with various timescales. Here, we fabricated thin-film transistors with controllable temporal dynamics, which can be easily tuned with electrical operation signals and showed excellent cycle-to-cycle uniformity. Based on this, we constructed a temporal adaptive reservoir capable of extracting temporal information of multiple timescales, thereby achieving improved accuracy in the human-activity-recognition task. Moreover, by leveraging the former computing output to modify the hyperparameters, we constructed a closed-loop architecture that equips the reservoir computing system with temporal self-adaptability according to the current input. The adaptability is demonstrated by accurate real-time recognition of objects moving at diverse speed levels. This work provides an approach for reservoir computing systems to achieve real-time processing of spatiotemporal signals with compound temporal characteristics.

Modeling and Optimization of Herb-Fortified Fresh Kombucha Cheese: An Artificial Neural Network Approach for Enhancing Quality Characteristics

In this study, an Artificial Neural Network (ANN) model is used to solve the complex task of producing fresh cheese with the desired quality parameters. The study focuses on kombucha fresh cheese samples fortified with ground wild thyme, supercritical fluid extract of wild thyme, ground sage and supercritical fluid extract of sage and optimizes the parameters of chemical composition, antioxidant potential and microbiological profile. The ANN models demonstrate robust generalization capabilities and accurately predict the observed results based on the input parameters. The optimal neural network model (MLP 6-10-16) with 10 neurons provides high r2 values (0.993 for training, 0.992 for testing, and 0.992 for validation cycles). The ANN model identified the optimal sample, a supercritical fluid extract of sage, on the 20th day of storage, showcasing specific favorable process parameters. These parameters encompass dry matter, fat, ash, proteins, water activity, pH, antioxidant potential (TP, DPPH, ABTS, FRAP), and microbiological profile. These findings offer valuable insights into producing fresh cheese efficiently with the desired quality attributes. Moreover, they highlight the effectiveness of the ANN model in optimizing diverse parameters for enhanced product development in the dairy industry.