Feedforward Neural Network Classifier Research Articles

Development of a long noncoding RNA-based machine learning model to predict COVID-19 in-hospital mortality

Tools for predicting COVID-19 outcomes enable personalized healthcare, potentially easing the disease burden. This collaborative study by 15 institutions across Europe aimed to develop a machine learning model for predicting the risk of in-hospital mortality post-SARS-CoV-2 infection. Blood samples and clinical data from 1286 COVID-19 patients collected from 2020 to 2023 across four cohorts in Europe and Canada were analyzed, with 2906 long non-coding RNAs profiled using targeted sequencing. From a discovery cohort combining three European cohorts and 804 patients, age and the long non-coding RNA LEF1-AS1 were identified as predictive features, yielding an AUC of 0.83 (95% CI 0.82–0.84) and a balanced accuracy of 0.78 (95% CI 0.77–0.79) with a feedforward neural network classifier. Validation in an independent Canadian cohort of 482 patients showed consistent performance. Cox regression analysis indicated that higher levels of LEF1-AS1 correlated with reduced mortality risk (age-adjusted hazard ratio 0.54, 95% CI 0.40–0.74). Quantitative PCR validated LEF1-AS1’s adaptability to be measured in hospital settings. Here, we demonstrate a promising predictive model for enhancing COVID-19 patient management.

Deep learning-based identification of esophageal cancer subtypes through analysis of high-resolution histopathology images.

Esophageal cancer (EC) remains a significant health challenge globally, with increasing incidence and high mortality rates. Despite advances in treatment, there remains a need for improved diagnostic methods and understanding of disease progression. This study addresses the significant challenges in the automatic classification of EC, particularly in distinguishing its primary subtypes: adenocarcinoma and squamous cell carcinoma, using histopathology images. Traditional histopathological diagnosis, while being the gold standard, is subject to subjectivity and human error and imposes a substantial burden on pathologists. This study proposes a binary class classification system for detecting EC subtypes in response to these challenges. The system leverages deep learning techniques and tissue-level labels for enhanced accuracy. We utilized 59 high-resolution histopathological images from The Cancer Genome Atlas (TCGA) Esophageal Carcinoma dataset (TCGA-ESCA). These images were preprocessed, segmented into patches, and analyzed using a pre-trained ResNet101 model for feature extraction. For classification, we employed five machine learning classifiers: Support Vector Classifier (SVC), Logistic Regression (LR), Decision Tree (DT), AdaBoost (AD), Random Forest (RF), and a Feed-Forward Neural Network (FFNN). The classifiers were evaluated based on their prediction accuracy on the test dataset, yielding results of 0.88 (SVC and LR), 0.64 (DT and AD), 0.82 (RF), and 0.94 (FFNN). Notably, the FFNN classifier achieved the highest Area Under the Curve (AUC) score of 0.92, indicating its superior performance, followed closely by SVC and LR, with a score of 0.87. This suggested approach holds promising potential as a decision-support tool for pathologists, particularly in regions with limited resources and expertise. The timely and precise detection of EC subtypes through this system can substantially enhance the likelihood of successful treatment, ultimately leading to reduced mortality rates in patients with this aggressive cancer.

Development and validation of a deep learning system for detection of small bowel pathologies in capsule endoscopy: a pilot study in a Singapore institution.

Deep learning models can assess the quality of images and discriminate among abnormalities in small bowel capsule endoscopy (CE), reducing fatigue and the time needed for diagnosis. They serve as a decision support system, partially automating the diagnosis process by providing probability predictions for abnormalities. We demonstrated the use of deep learning models in CE image analysis, specifically by piloting a bowel preparation model (BPM) and an abnormality detection model (ADM) to determine frame-level view quality and the presence of abnormal findings, respectively. We used convolutional neural network-based models pretrained on large-scale open-domain data to extract spatial features of CE images that were then used in a dense feed-forward neural network classifier. We then combined the open-source Kvasir-Capsule dataset (n = 43) and locally collected CE data (n = 29). Model performance was compared using averaged five-fold and two-fold cross-validation for BPMs and ADMs, respectively. The best BPM model based on a pre-trained ResNet50 architecture had an area under the receiver operating characteristic and precision-recall curves of 0.969±0.008 and 0.843±0.041, respectively. The best ADM model, also based on ResNet50, had top-1 and top-2 accuracies of 84.03±0.051 and 94.78±0.028, respectively. The models could process approximately 200-250 images per second and showed good discrimination on time-critical abnormalities such as bleeding. Our pilot models showed the potential to improve time to diagnosis in CE workflows. To our knowledge, our approach is unique to the Singapore context. The value of our work can be further evaluated in a pragmatic manner that is sensitive to existing clinician workflow and resource constraints.

Optimized FFNN with multichannel CSP-ICA framework of EEG signal for BCI

The electroencephalogram (EEG) of the patient is used to identify their motor intention, which is then converted into a control signal through a brain-computer interface (BCI) based on motor imagery. Whenever gathering features from EEG signals, making a BCI is difficult in part because of the enormous dimensionality of the data. Three stages make up the suggested methodology: pre-processing, extraction of features, selection, and categorization. To remove unwanted artifacts, the EEG signals are filtered by a fifth-order Butterworth multichannel band-pass filter. This decreases execution time and memory use, both of which improve system performance. Then a novel multichannel optimized CSP-ICA feature extraction technique is used to separate and eliminate non-discriminative information from discriminative information in the EEG channels. Furthermore, CSP uses the concept of an Artificial Bee Colony (ABC) algorithm to automatically identify the simultaneous global ideal frequency band and time interval combination for the extraction and classification of common spatial pattern characteristics. Finally, a Tunable optimized feed-forward neural network (FFNN) classifier is utilized to extract and categorize the temporal and frequency domain features, which employs an FFNN classifier with Tunable-Q wavelet transform. The proposed framework, therefore optimizes signal processing, enabling enhanced EEG signal classification for BCI applications. The result shows that the models that use Tunable optimized FFNN produce higher classification accuracy of more than 20% when compared to the existing models.

Performance Evaluation of Deep Learning Techniques in The Detection of IOT Malware

Internet of Things (IoT) equipment is rapidly being used in a variety of businesses and for a variety of reasons (for example, sensing and collecting data from the environment in both public and military settings). Because of their expanding involvement in a wide range of applications and their rising computational and processing capabilities, they are a viable attack target for malware tailored to infect specific IoT devices. This study investigates the potential of detecting IoT malware using different deep learning techniques: the classic feedforward neural network (FNN), convolutional neural networks (CNN), long short-term memory (LSTM), and recurrent neural networks (RNN). The proposed method analyses the execution operation codes of IOT app sequences using modern NLP (natural language processing) methods. The current work utilized an IoT application dataset with 500 malware (collected from the IOTPOT dataset) and 500 goodware samples to train the proposed algorithms. The trained model is tested against 2971 fresh IoT malware and goodware samples. The samples were input into deep learning models, and performance metrics were obtained. The results demonstrate that the RNN model had the best accuracy (99.19%) in detecting fresh malware samples. On the other hand, the results were compared by the time required for training; the CNN model shows that it could achieve high accuracy (98.05%) with less training time. A comparison with various deep learning classifiers demonstrates that the RNN and CNN techniques produce the best results.

Development of surrogate-optimization models for a novel transcritical power cycle integrated with a small modular reactor

In recent years, various types of surrogate optimization models have been proposed to reduce the computational time and to improve the emulation accuracy. In this study, by leveraging an ANN surrogate model developed earlier, a comprehensive and efficient optimization algorithm is conceived for the global optimal design of an integrated regenerative methanol transcritical cycle. It combines a unique converging/diverging classifier model into the surrogate model to form a surrogate-based model, which significantly improves the prediction accuracy of the objective function. Six binary classifiers are explored and the multi-layer feed-forward (MLF) neural network classifier is selected. In addition, within the five global optimizers being explored, the basin-hopping (BH) and dual-annealing (DA) are selected. The optimal surrogate-based model and global optimizers are then combined to form a unique surrogate-optimizer model. The surrogate-optimizer model is slightly outperformed by the physics-based model in terms of the optimization results, the time consumption of the surrogate-optimizer model during the optimization searching process is 99% less than that of the physics-based model. As the results, the surrogate-optimizer model is slightly outperformed by the physics-based model in terms of the optimization results, where the Levelized Cost of Energy (LCOE) of the Surrogate-DA and Surrogate-BH models are 77.912 and 78.876 $/MWh, respectively, compared to the 77.190 $/MWh of the Baseline model with fairly close penalties between them. In the meantime, the time consumption of the surrogate-optimizer model during the optimization searching process is 99% less than that of the physics-based model.

A Deep Learning Approach for Network Intrusion Detection Using a Small Features Vector

With the growth in network usage, there has been a corresponding growth in the nefarious exploitation of this technology. A wide array of techniques is now available that can be used to deal with cyberattacks, and one of them is network intrusion detection. Artificial Intelligence (AI) and Machine Learning (ML) techniques have extensively been employed to identify network anomalies. This paper provides an effective technique to evaluate the classification performance of a deep-learning-based Feedforward Neural Network (FFNN) classifier. A small feature vector is used to detect network traffic anomalies in the UNSW-NB15 and NSL-KDD datasets. The results show that a large feature set can have redundant and unuseful features, and it requires high computation power. The proposed technique exploits a small feature vector and achieves better classification accuracy.

Automated COVID-19 detection with convolutional neural networks

This paper focuses on addressing the urgent need for efficient and accurate automated screening tools for COVID-19 detection. Inspired by existing research efforts, we propose two framework models to tackle this challenge. The first model combines a conventional CNN architecture as a feature extractor with XGBoost as the classifier. The second model utilizes a classical CNN architecture with a Feedforward Neural Network for classification. The key distinction between the two models lies in their classification layers. Bayesian optimization techniques are employed to optimize the hyperparameters of both models, enabling a “cheat-start” to the training process with optimal configurations. To mitigate overfitting, transfer learning techniques such as Dropout and Batch normalization are incorporated. The CovidxCT-2A dataset is used for training, validation, and testing purposes. To establish a benchmark, we compare the performance of our models with state-of-the-art methods reported in the literature. Evaluation metrics including Precision, Recall, Specificity, Accuracy, and F1-score are employed to assess the efficacy of the models. The hybrid model demonstrates impressive results, achieving high precision (98.43%), recall (98.41%), specificity (99.26%), accuracy (99.04%), and F1-score (98.42%). The standalone CNN model exhibits slightly lower but still commendable performance, with precision (98.25%), recall (98.44%), specificity (99.27%), accuracy (98.97%), and F1-score (98.34%). Importantly, both models outperform five other state-of-the-art models in terms of classification accuracy, as demonstrated by the results of this study.

Construction and approximation for a class of feedforward neural networks with sigmoidal function

As we know, feedforward neural networks (FNNs) with sigmoidal activation function are universal approximators. Theoretically, for any continuous function defined on a compact set, there exists an FNN such that the FNN can approximate the function with arbitrary accuracy, which constitutes a theoretical guarantee that FNN can be used as an efficient learning machine. This paper addresses the construction and approximation for FNNs. We construct an FNN with sigmoidal activation function and estimate its approximation error. In particular, an inverse theorem of the approximation is established, which implies the equivalence characterization theorem of the approximation and reveals the relationship between the topological structure of the FNN and its approximation ability. As keys in this study, the concepts of modulus of continuity of function, K-functional, and their relationship are utilized, and two Bernstein-type inequalities are established.

The utility of machine learning for predicting donor discard in abdominal transplantation.

Increasing access and better allocation of organs in the field of transplantation is a critical problem in clinical care. Limitations exist in accurately predicting allograft discard. Potential exists for machine learning to provide a balanced assessment of the potential for an organ to be used in a transplantation procedure. We accessed and utilized all available deceased donor United Network for Organ Sharing data from 1987 to 2020. With these data, we evaluated the performance of multiple machine learning methods for predicting organ use. The machine learning methods trialed included XGBoost, random forest, Naïve Bayes (NB), logistic regression, and fully connected feedforward neural network classifier methods. The top two methods, XGBoost and random forest, were fully developed using 10-fold cross-validation and Bayesian optimization of hyperparameters. The top performing model at predicting liver organ use was an XGBoost model which achieved an AUC-ROC of .925, an AUC-PR of .868, and an F1 statistic of .756. The top performing model for predicting kidney organ use classification was an XGBoost model which achieved an AUC-ROC of .952, and AUC-PR of .883, and an F1 statistic of .786. The XGBoost method demonstrated a significant improvement in predicting donor allograft discard for both kidney and livers in solid organ transplantation procedures. Machine learning methods are well suited to be incorporated into the clinical workflow; they can provide robust quantitative predictions and meaningful data insights for clinician consideration and transplantation decision-making.

Gastric Disorder Analysis Using Hybrid Optimization with Machine Learning

The stomach and all of its appendages, which include the oesophagus, duodenum, small intestine, and large intestine, amongst others, all play a crucial function within this system. Stomach dysrhythmias, which are linked to problems with the movement of gastrointestinal contents, affect a significant number of individuals all over the globe. These problems include inappropriate digestion (dyspepsia), nausea (vomiting sensation) for no apparent reason, vomiting, abdominal pain, stomach ulcers, gastroesophageal reflux disease, and other disorders. During the process of finding the anomalies, it is possible that a number of techniques, including as imaging, endoscopy, electrogastrogram, and clinical analysis, will be used. Electrogastrography signals, also known as electrogastrograms (EGG), were captured using surface Ag/AgCl electrodes that were put over the stomach in 20 healthy persons before the data was gathered and pre-processed. The datasets were produced from these signals (8 Females and 12 Males). In addition to this, the datasets were obtained from 10 individuals who were suffering from various stomach illnesses (3 Females and 8 Males). In the stage known as “pre-processing,” which needs the obtained dataset to be treated in advance, any noise that was present in the signal is removed. In order to rid the data of any noise and increase the overall quality of the input data, a technique that is known as the Wiener filter is used. A technique known as Hybrid Grey Wolf Optimization with Particle Swarm Optimization is utilized in the process of feature selection. This algorithm is responsible for removing any extraneous data from the features that have been collected from the signal. The procedure is sped up as a result of this. The classifiers get the qualities that have been chosen as their input in order to carry out an analysis of the many stomach disorders, such as primary gastric lymphoma, gastrointestinal stromal tumour (GIST), and neuroendocrine tumor. This enables the classifiers to do the analysis (carcinoid). The Multi-class Feed Forward Neural Network Classifier (MCFFN) is used to carry out the classification process. This classifier provides the stages along with the classes. The accuracy, sensitivity, and specificity of the classification process are taken into account in the calculation of performance measures.

Early Detection of Malignant Tumor in Lungs Using Feed-Forward Neural Network and K-Nearest Neighbor Classifier

Early Detection of Malignant Tumor in Lungs Using Feed-Forward Neural Network and K-Nearest Neighbor Classifier

Monitoring Level of Hypnosis using Stationary Wavelet Transform and Singular Value Decomposition Entropy with Feedforward Neural Network.

Classifying the patient's depth of anesthesia (LoH) level into a few distinct states may lead to inappropriate drug administration. To tackle the problem, this paper presents a robust and computationally efficient framework that predicts a continuous LoH index scale from 0-100 in addition to the LoH state. This paper proposes a novel approach for accurate LoH estimation based on Stationary Wavelet Transform (SWT) and fractal features. The deep learning model adopts an optimized temporal, fractal, and spectral feature set to identify the patient sedation level irrespective of age and the type of anesthetic agent. This feature set is then fed into a multilayer perceptron network (MLP), a class of feed-forward neural networks. A comparative analysis of regression and classification is made to measure the performance of the chosen features on the neural network architecture. The proposed LoH classifier outperforms the state-of-the-art LoH prediction algorithms with the highest accuracy of 97.1% while utilizing minimized feature set and MLP classifier. Moreover, for the first time, the LoH regressor achieves the highest performance metrics (R2 = 0.9, MAE = 1.5) as compared to previous work. This study is very helpful for developing highly accurate monitoring for LoH which is important for intraoperative and postoperative patients' health.

Wavelet Decomposition and Feedforward Neural Network for Classification of Acute Ischemic Stroke based on Electroencephalography

Stroke is one of the leading causes of death in Indonesia. From 2013 to 2018, the prevalence of stroke increased from 7% to 10.9%. There are two types of strokes, namely Hemorrhagic and Acutte Ischemic Stroke (AIS) with the majority of it being AIS. Early detection and diagnosis are essential in stroke as it is a life-threatening disease, and the stroke treatment is based on its type. Currently, the gold imaging standards in stroke diagnosis are Computed Tomography (CT) scan and Magnetic Resonance Imaging (MRI). However, the mentioned services for stroke diagnosis are primarily available in hospitals classified as “class A” (general hospitals with extensive facilities and medical services). Compared to CT scans and MRI, electroencephalography (EEG) is a cost-friendly, non-invasive device studied for various brain-related diseases. This study aimed to determine the optimal epoch length to classify four stroke classes (healthy, minor, moderate, and severe) during resting condition for a machine learning-based AIS computer-aided diagnostics system. 32-channel EEG, CT scan, and NIHSS Scores were the obtained data. The features were delta-theta to alpha-beta ratio (DTABR), delta to alpha ratio (DAR), relative power ratio (RPR), and asymmetry, which were extracted using wavelet decomposition technique. The epoch length was varied by 1s, 2s, 10s, 30s, 60s, and 120s. The severity of stroke were classified using a feedforward neural network. The best performance was obtained at the 60-second epoch length with 89% accuracy using 15 hidden layers. This EEG-based diagnostic system would be expected to be implemented in “class C” hospitals, where only essential medical services and facilities are available, usually in rural areas.

Recognizing elderly peoples by analyzing their walking pattern using body posture skeleton

The increasing age of the population has become a significant concern internationally. During the COVID-19 pandemic situation, it has been seen that the most sensitive and affected class of the population is the class of Elder’s. It is therefore necessary to track the movement and behavior of the old persons. This kind of monitoring could help them in providing assistance in their needy time. Our objective is to develop an approach to classify elderly people using skeleton data for their assistance. OpenPose algorithm is used here to detect human skeletons (joint positions) from the video sequences. OpenPose algorithm with a sliding window of size ‘N’ is used to achieve a real-time posture recognition framework. Posture features from each extracted skeleton are then used to build a classifier for recognizing elderly people. We also introduce here a new dataset that includes old person walk and young person walk video’s. The experimental outcomes reveal that the proposed method has achieved up to 98.45% training accuracy and 96.16% testing accuracy for deep feed-forward neural network (FFNN) classifier. This asserts the effectiveness of the approach.

Investigation of optimized ELM using Invasive Weed-optimization and Cuckoo-Search optimization

Abstract In order to classify data and improve extreme learning machine (ELM), this study explains how a hybrid optimization-driven ELM technique was devised. Input data are pre-processed in order to compute missing values and convert data to numerical values using the exponential kernel transform. The Jaro–Winkler distance is used to identify the relevant features. The feed-forward neural network classifier is used to categorize the data, and it is trained using a hybrid optimization technique called the modified enhanced Invasive Weed, a meta heuristic algorithm, and Cuckoo Search, a non-linear optimization algorithm ELM. The enhanced Invasive Weed optimization (IWO) algorithm and the enhanced Cuckoo Search (CS) algorithm are combined to create the modified CSIWO. The experimental findings presented in this work demonstrate the viability and efficacy of the created ELM method based on CSIWO, with good experimental result as compared to other ELM techniques.

A novel grey prediction model with a feedforward neural network based on a carbon emission dynamic evolution system and its application.

The objective and accurate prediction of carbon dioxide emissions holds great significance for improving governmental energy policies and plans. Therefore, starting from an evolutionary system of carbon emissions, this paper studies the evolution of the system, establishes a grey model of the system, and expands the modeling structure of this model. The modeling mechanism of the classical feedforward neural network model is organically combined with the function of the external influencing factors of carbon emissions, and the grey model of the carbon emission dynamic system is established with a neural network. Then, the properties of the model are studied, the parameters of the model are optimized, and the modeling steps are obtained. Finally, the validity of the model is analyzed by using the carbon emissions of Beijing from 2009 to 2018. The results of the four cases show that the simulation and prediction errors of the new model are all less than 10%, and case 1 shows the best results of 1.56% and 2.07%, respectively, which are used to predict the carbon dioxide emissions in the next 5 years in Beijing. The prediction results are in accordance with the actual trend, which indicates the effectiveness and feasibility of the model.

Diagnosis of Histopathological Images to Distinguish Types of Malignant Lymphomas Using Hybrid Techniques Based on Fusion Features

Malignant lymphoma is one of the types of malignant tumors that can lead to death. The diagnostic method for identifying malignant lymphoma is a histopathological analysis of lymphoma tissue images. Because of the similar morphological characteristics of the lymphoma types, it is difficult for doctors and specialists to manually distinguish the types of lymphomas. Therefore, deep and automated learning techniques aim to solve this problem and help clinicians reconsider their diagnostic decisions. Because of the similarity of the morphological characteristics between lymphoma types, this study aimed to extract features using various algorithms and deep learning models and combine them together into feature vectors. Two datasets have been applied, each with two different systems for the reliable diagnosis of malignant lymphoma. The first system was a hybrid system between DenseNet-121 and ResNet-50 to extract deep features and reduce their dimensions by the principal component analysis (PCA) method, using the support vector machine (SVM) algorithm for classifying low-dimensional deep features. The second system was based on extracting the features using DenseNet-121 and ResNet-50 and combining them with the hand-crafted features extracted by gray level co-occurrence matrix (GLCM), fuzzy color histogram (FCH), discrete wavelet transform (DWT), and local binary pattern (LBP) algorithms and classifying them using a feed-forward neural network (FFNN) classifier. All systems achieved superior results in diagnosing the two datasets of malignant lymphomas. An FFNN classifier with features of ResNet-50 and hand-crafted features reached an accuracy of 99.5%, specificity of 100%, sensitivity of 99.33%, and AUC of 99.86% for the first dataset. In contrast, the same technique reached 100% for all measures to diagnose the second dataset.

Fault Restoration of Six-Axis Force/Torque Sensor Based on Optimized Back Propagation Networks.

Six-axis force/torque sensors are widely installed in manipulators to help researchers achieve closed-loop control. When manipulators work in comic space and deep sea, the adverse ambient environment will cause various degrees of damage to F/T sensors. If the disability of one or two dimensions is restored by self-restoration methods, the robustness and practicality of F/T sensors can be considerably enhanced. The coupling effect is an important characteristic of multi-axis F/T sensors, which implies that all dimensions of F/T sensors will influence each other. We can use this phenomenon to speculate the broken dimension by other regular dimensions. Back propagation neural network (BPNN) is a classical feedforward neural network, which consists of several layers and adopts the back-propagation algorithm to train networks. Hyperparameters of BPNN cannot be updated by training, but they impact the network performance directly. Hence, the particle swarm optimization (PSO) algorithm is adopted to tune the hyperparameters of BPNN. In this work, each dimension of a six-axis F/T sensor is regarded as an element in the input vector, and the relationships among six dimensions can be obtained using optimized BPNN. The average MSE of restoring one dimension and two dimensions over the testing data is and , respectively. Furthermore, the average quote error of one restored dimension and two restored dimensions are and , respectively. The analysis of experimental results illustrates that the proposed fault restoration method based on PSO-BPNN is viable and practical. The F/T sensor restored using the proposed method can reach the original measurement precision.

Natural Language Processing Approaches for Automated Multilevel and Multiclass Classification of Breast Lesions on Free-Text Cytopathology Reports.

The extensive growth and use of electronic health records (EHRs) and extending medical literature have led to huge opportunities to automate the extraction of relevant clinical information that helps in concise and effective clinical decision support. However, processing such information has traditionally been dependent on labor-intensive processes with human errors such as fatigue, oversight, and interobserver variability. Hence, this study aims at the processing of EHRs and performing multilevel and multiclass classification by fetching dominant characteristic features that are sufficient to detect and differentiate various types of breast lesions. In this study, unstructured EHRs on breast lesions obtained through fine-needle aspiration cytology technique are considered. The raw text was normalized into structured tabular form and converted to scores by performing sentiment analysis that helps to decide the total polarity or class label of the EHR. Supervised machine learning approaches, namely random forest and feed-forward neural network trained using Levenberg-Marquardt training function, are used for classification of the collected EHR data set containing 2,879 records that are split in the ratio of 80:20 as training and testing data sets, respectively. Random forest and feed-forward neural network classifiers gave the best performance with an accuracy of 99.36%, an overall receiver operating characteristic-area under the curve of 99.2%, a correlation with ground truth of 98.3%, and a histopathologic correlation of 98.6%. Natural language processing has huge potential to automate the extraction of clinical features from breast lesions. The proposed multilevel and multiclass classification approach is used to classify 13 different types of breast lesions with 20 different labels into five classes to decide the type of treatment that should be given to patients by a physician or oncologist.