Feature Selection Module Research Articles

Crack Detection of Concrete Based on Improved CenterNet Model

Cracks on concrete surfaces are vital factors affecting construction safety. Accurate and efficient crack detection can prevent safety-related accidents. Using drones to photograph cracks on a concrete surface and detect them through computer vision technology has the advantages of accurate target recognition, simple practical operation, and low cost. To solve this problem, an improved CenterNet concrete crack-detection model is proposed. Firstly, a channel-space attention mechanism is added to the original model to enhance the ability of the convolution neural network to pay attention to the image. Secondly, a feature selection module is introduced to scale the feature map in the downsampling stage to a uniform size and combine it in the channel dimension. In the upsampling stage, the feature selection module adaptively selects the combined features and fuses them with the output features of the upsampling. Finally, the target size loss is optimized from a Smooth L1 Loss to IoU Loss to lessen its inability to adapt to targets of different sizes. The experimental results show that the improved CenterNet model reduces the FPS by 123.7 Hz, increases the GPU memory by 62 MB, increases the FLOPs by 3.81 times per second, and increases the AP by 15.4% compared with the original model. The GPU memory occupancy remained stable during the training process and exhibited good real-time performance and robustness.

A hybrid deep learning model for urban expressway lane-level mixed traffic flow prediction

Precise real-time traffic flow prediction is crucial for route guidance and traffic fine control. With the development of autonomous driving, the mixed traffic flow state composed of Connected Automated Vehicles (CAVs) and Human-driven Vehicles (HVs) provides new insight into traffic flow prediction. In this paper, we innovatively consider the interaction between heterogeneous traffic flow as well as the mutual effect of traffic flow on different lanes and develop a hybrid model based on deep learning for urban expressway lane-level mixed traffic flow prediction, including three modules. First, the feature selection module is applied to screen the features with a high spatio-temporal correlation to the prediction object and construct the input matrix. Then, it is input to the feature attention module to quantify the importance of the input features on the prediction object, thereby assigning attention weights. Finally, the spatio-temporal information fusion module is adopted to capture the global spatio-temporal dynamics of traffic flow at horizontal and vertical spatial scales, as well as learn the complex coupling characteristics of heterogeneous traffic flow, thus obtaining predictions. An urban expressway mixed traffic flow simulation environment is built to collect experimental datasets for prediction accuracy evaluation. The results indicate that the proposed model outperforms the benchmarks in single-step and multi-step mixed traffic flow predictions on each lane. Furthermore, the proposed model shows the best performance and strong robustness under different penetration rates of connected automated vehicles.

CoBEV: Elevating Roadside 3D Object Detection with Depth and Height Complementarity.

Roadside camera-driven 3D object detection is a crucial task in intelligent transportation systems, which extends the perception range beyond the limitations of vision-centric vehicles and enhances road safety. While previous studies have limitations in using only depth or height information, we find both depth and height matter and they are in fact complementary. The depth feature encompasses precise geometric cues, whereas the height feature is primarily focused on distinguishing between various categories of height intervals, essentially providing semantic context. This insight motivates the development of Complementary-BEV (CoBEV), a novel end-to-end monocular 3D object detection framework that integrates depth and height to construct robust BEV representations. In essence, CoBEV estimates each pixel's depth and height distribution and lifts the camera features into 3D space for lateral fusion using the newly proposed two-stage complementary feature selection (CFS) module. A BEV feature distillation framework is also seamlessly integrated to further enhance the detection accuracy from the prior knowledge of the fusion-modal CoBEV teacher. We conduct extensive experiments on the public 3D detection benchmarks of roadside camera-based DAIR-V2X-I and Rope3D, as well as the private Supremind-Road dataset, demonstrating that CoBEV not only achieves the accuracy of the new state-of-the-art, but also significantly advances the robustness of previous methods in challenging long-distance scenarios and noisy camera disturbance, and enhances generalization by a large margin in heterologous settings with drastic changes in scene and camera parameters. For the first time, the vehicle AP score of a camera model reaches 80% on DAIR-V2X-I in terms of easy mode. The source code will be made publicly available at CoBEV.

COSTA: A Multi-center TOF-MRA Dataset and A Style Self-Consistency Network for Cerebrovascular Segmentation.

Time-of-flight magnetic resonance angiography (TOF-MRA) is the least invasive and ionizing radiation-free approach for cerebrovascular imaging, but variations in imaging artifacts across different clinical centers and imaging vendors result in inter-site and inter-vendor heterogeneity, making its accurate and robust cerebrovascular segmentation challenging. Moreover, the limited availability and quality of annotated data pose further challenges for segmentation methods to generalize well to unseen datasets. In this paper, we construct the largest and most diverse TOF-MRA dataset (COSTA) from 8 individual imaging centers, with all the volumes manually annotated. Then we propose a novel network for cerebrovascular segmentation, namely CESAR, with the ability to tackle feature granularity and image style heterogeneity issues. Specifically, a coarse-to-fine architecture is implemented to refine cerebrovascular segmentation in an iterative manner. An automatic feature selection module is proposed to selectively fuse global long-range dependencies and local contextual information of cerebrovascular structures. A style self-consistency loss is then introduced to explicitly align diverse styles of TOF-MRA images to a standardized one. Extensive experimental results on the COSTA dataset demonstrate the effectiveness of our CESAR network against state-of-the-art methods. We have made 6 subsets of COSTA with the source code online available, in order to promote relevant research in the community.

LEGAN: A low-light image enhancement generative adversarial network for industrial internet of smart-cameras

The utilization of smart-cameras in the context of the Internet of Things (IoT) has become increasingly prevalent within smart workshops for performing in-situ quality inspection tasks. However, it is worth noting that these smart-cameras may encounter operational challenges when functioning under low-light conditions. The images acquired in such situation are severely degraded, resulting in the performance decline of the subsequent detection algorithms. Focusing on non-stationary noise compression and detail recovery, this paper constructs a novel enhancement model called LEGAN for the industrial internet of smart-cameras system. Firstly, the input undergoes a decomposition process into two branches using the Harr-wavelet technique. These branches are subsequently encoded independently by a series of compact residual blocks, facilitating effective noise suppression. Secondly, in order to enhance detail recovery, a feature selection module is meticulously designed to extract correlations between image foreground–background and low–high frequency signals, ultimately reconstructing a comprehensive feature map. This enables a multi-scale stepwise up-sampling approach that facilitates image recovery based on the reconstructed feature maps. Lastly, the training phase is supervised by an adversarial loss, comprising MSE loss, VGG loss, and discriminating loss, which ensures a harmonious balance between noise suppression and detail recovery. Comparative experiments clearly show the superiority of the LEGAN in terms of noise compression and detail recovery. Moreover, from an industrial practice perspective, the application of the proposed approach to yarn evenness inspection has proven to be highly effective, significantly enhancing detection accuracy in low-light environments.

Dual contrast attention-guided multi-frequency fusion for multi-contrast MRI super-resolution

Objective. Multi-contrast magnetic resonance (MR) imaging super-resolution (SR) reconstruction is an effective solution for acquiring high-resolution MR images. It utilizes anatomical information from auxiliary contrast images to improve the quality of the target contrast images. However, existing studies have simply explored the relationships between auxiliary contrast and target contrast images but did not fully consider different anatomical information contained in multi-contrast images, resulting in texture details and artifacts unrelated to the target contrast images. Approach. To address these issues, we propose a dual contrast attention-guided multi-frequency fusion (DCAMF) network to reconstruct SR MR images from low-resolution MR images, which adaptively captures relevant anatomical information and processes the texture details and low-frequency information from multi-contrast images in parallel. Specifically, after the feature extraction, a feature selection module based on a dual contrast attention mechanism is proposed to focus on the texture details of the auxiliary contrast images and the low-frequency features of the target contrast images. Then, based on the characteristics of the selected features, a high- and low-frequency fusion decoder is constructed to fuse these features. In addition, a texture-enhancing module is embedded in the high-frequency fusion decoder, to highlight and refine the texture details of the auxiliary contrast and target contrast images. Finally, the high- and low-frequency fusion process is constrained by integrating a deeply-supervised mechanism into the DCAMF network. Main results. The experimental results show that the DCAMF outperforms other state-of-the-art methods. The peak signal-to-noise ratio and structural similarity of DCAMF are 39.02 dB and 0.9771 on the IXI dataset and 37.59 dB and 0.9770 on the BraTS2018 dataset, respectively. The image recovery is further validated in segmentation tasks. Significance. Our proposed SR model can enhance the quality of MR images. The results of the SR study provide a reliable basis for clinical diagnosis and subsequent image-guided treatment.

Hybrid ultra-short-term PV power forecasting system for deterministic forecasting and uncertainty analysis

The rapid development of the photovoltaic industry provides a new source of power for the continued operation of the over-consumed energy world. While providing new opportunities for global energy systems, it also poses challenges for power grids. Therefore, it is a priority to fully grasp the characteristics of photovoltaic changes and accurately forecast and analyze them. To enrich the existing research, a novel hybrid prediction system considering meteorological factors is constructed. First, a feature selection module is introduced to select features and assign weights to exogenous meteorological factors, which breaks through the limitations of single-data dimension prediction. Second, shallow and deep learning models are flexibly applied and multi-objective intelligent optimization strategies are introduced to construct deterministic combinatorial prediction models. The module can effectively increase the diversity of prediction models while fully weighing the accuracy and stability of prediction to meet the needs of different information users. Finally, an interval prediction model is constructed to further enrich the PV power prediction system from the perspective of uncertainty analysis. The empirical study is carried out with 5-minute interval data at three sites, and the results show that the hybrid system obtains superior out-of-sample forecasting performance with technical feasibility and general applicability.

Integrated Algorithm Based on Bidirectional Characteristics and Feature Selection for Fire Image Classification

In some fire classification task samples, it is especially important to learn and select limited features. Therefore, enhancing shallow characteristic learning and accurately reserving deep characteristics play a decisive role in the final fire classification task. In this paper, we propose an integrated algorithm based on bidirectional characteristics and feature selection for fire image classification called BCFS-Net. This algorithm is integrated from two modules, a bidirectional characteristics module and feature selection module; hence, it is called an integrated algorithm. The main process of this algorithm is as follows: First, we construct a bidirectional convolution module to obtain multiple sets of bidirectional traditional convolutions and dilated convolutions for the feature mining and learning shallow features. Then, we improve the Inception V3 module. By utilizing the bidirectional attention mechanism and Euclidean distance, feature points with greater correlation between the feature maps generated by convolutions in the Inception V3 module are selected. Next, we comprehensively consider and integrate feature points with richer semantic information from multiple dimensions. Finally, we use convolution to further learn the deep features and complete the final fire classification task. We validated the feasibility of our proposed algorithm in three sets of public fire datasets, and the overall accuracy value in the BoWFire dataset reached 88.9%. The overall accuracy in the outdoor fire dataset reached 96.96%. The overall accuracy value in the Fire Smoke dataset reached 81.66%.

Enhanced Feature Extraction Network Based on Acoustic Signal Feature Learning for Bearing Fault Diagnosis.

The method of acoustic radiation signal detection not only enables contactless measurement but also provides comprehensive state information during equipment operation. This paper proposes an enhanced feature extraction network (EFEN) for fault diagnosis of rolling bearings based on acoustic signal feature learning. The EFEN network comprises four main components: the data preprocessing module, the information feature selection module (IFSM), the channel attention mechanism module (CAMM), and the convolutional neural network module (CNNM). Firstly, the one-dimensional acoustic signal is transformed into a two-dimensional grayscale image. Then, IFSM utilizes three different-sized convolution filters to process input image data and fuse and assign weights to feature information that can attenuate noise while highlighting effective fault information. Next, a channel attention mechanism module is introduced to assign weights to each channel. Finally, the convolutional neural network (CNN) fault diagnosis module is employed for accurate classification of rolling bearing faults. Experimental results demonstrate that the EFEN network achieves high accuracy in fault diagnosis and effectively detects rolling bearing faults based on acoustic signals. The proposed method achieves an accuracy of 98.52%, surpassing other methods in terms of performance. In comparative analysis of antinoise experiments, the average accuracy remains remarkably high at 96.62%, accompanied by a significantly reduced average iteration time of only 0.25 s. Furthermore, comparative analysis confirms that the proposed algorithm exhibits excellent accuracy and resistance against noise.

Research on pathologic myopia recognition based on vision transformer

Currently, the diagnosis of pathological myopia is mostly done through manual diagnosis, which not only requires experienced ophthalmologists but is also time-consuming and labour-intensive. In order to improve the diagnostic efficiency and accuracy, and to prevent irreversible visual impairment caused by missed diagnosis, misdiagnosis, and delayed treatment, this paper presents a fine-grained image analysis task of classifying fundus images of patients with pathological myopia and non-pathological myopia. To accurately identify subtle differences in features among similar fundus images, a pathological myopia recognition model based on Vision Transformer (ViT) is proposed. The model incorporates a feature selection module using self-attention mechanism that can effectively select important features in the fundus images, thereby eliminating the influence of irrelevant regions on recognition. Experimental results demonstrate that this method outperforms traditional ViT models, achieving high accuracy in pathological myopia recognition.

Tri-SeizureDualNet: A novel multimodal brain seizure detection using triple stream skipped feature extraction module entrenched dual parallel attention transformer

The timely and accurate detection of epileptic seizures is highly needed to enhance the quality of a patient’s life. The state-of-the-art works to design and utilize many deep learning techniques to detect and predict seizures using EEG, fMRI, and a combination of EEG and fMRI modalities respectively. Whereas the existing models are highly vulnerable to complexity, and overfitting issues due to improper feature analysis. To this end, we design a novel seizure detection model named Triple stream skipped feature extraction module and Dual parallel attention transformer Network (Tri-SeizureDualNet) using EEG and fMRI modalities. The major scope of this research is to enhance the detection accuracy of seizures with better performance. The data acquired for this research is from the CHB-MIT database for EEG and the UW Madison database for fMRI. The designed model tends to perform feature extraction, feature selection, feature engineering, and classification respectively with high feature extraction accuracy and less complexity. Beforehand, we perform pre-processing on both the EEG and fMRI data using ICA, PCA, and high pass filtering. Followed by we perform feature extraction in three streams with skip connections from both modalities to improve the extractor capability and feature extraction accuracy. The extracted features are selected in the feature selection module using the Humming Bird Optimization (HBO) algorithm based on the feature importance ratio. The selected features from both modalities were fused and provided to the Dual Parallel Attention Transformer (DPAT) for feature engineering and context learning respectively. The utilized DPAT is composed of two regularized attention modules with a gating mechanism to picture the feature similarity rate. At last, the softmax layer classifies the features into three classes such as focal onset, general onset, and unknown onset seizures respectively. The experimental results were conducted with two utilized datasets such as CHB-MIT and UW Madison datasets separately for EEG and fMRI data in which the proposed work achieves an overall accuracy rate of 99.67% and AUC rate of 0.989 respectively. The results reveal that the proposed research outperforms the existing works.

Attention-based dual-path feature fusion network for automatic skin lesion segmentation

Automatic segmentation of skin lesions is a critical step in Computer Aided Diagnosis (CAD) of melanoma. However, due to the blurring of the lesion boundary, uneven color distribution, and low image contrast, resulting in poor segmentation result. Aiming at the problem of difficult segmentation of skin lesions, this paper proposes an Attention-based Dual-path Feature Fusion Network (ADFFNet) for automatic skin lesion segmentation. Firstly, in the spatial path, a Boundary Refinement (BR) module is designed for the output of low-level features to filter out irrelevant background information and retain more boundary details of the lesion area. Secondly, in the context path, a Multi-scale Feature Selection (MFS) module is constructed for high-level feature output to capture multi-scale context information and use the attention mechanism to filter out redundant semantic information. Finally, we design a Dual-path Feature Fusion (DFF) module, which uses high-level global attention information to guide the step-by-step fusion of high-level semantic features and low-level detail features, which is beneficial to restore image detail information and further improve the pixel-level segmentation accuracy of skin lesion. In the experiment, the ISIC 2018 and PH2 datasets are employed to evaluate the effectiveness of the proposed method. It achieves a performance of 0.890/ 0.925 and 0.933 /0.954 on the F1-score and SE index, respectively. Comparative analysis with state-of-the-art segmentation methods reveals that the ADFFNet algorithm exhibits superior segmentation performance.

MDFN: A Multi-level Dynamic Fusion Network with self-calibrated edge enhancement for lung nodule segmentation

Precise segmentation of lung nodules from the surrounding tissues provides radiologists with distinct boundary details, instrumental to the meticulous diagnosis and subsequent treatment of lung cancer. Although extensively applied in practice, the existing deep learning approaches are generally confronted with under-segmentation for intractable nodules with heterogeneous textures, blurred boundaries, adhesive structures and the loss of crucial spatial and edge details during the deep convolutional encoding. Aiming at these issues, we take full advantage of the complementarity between edge and region to enhance the restoration of complex edges and propose a 2D Multi-level Dynamic Fusion Network (MDFN) for precise lung nodule segmentation and fine-grained edge details. Firstly, we innovatively construct a multi-scale spatial and channel feature selection module (MSCFS) to enrich the receptive fields and capture highly correlated multi-scale contexts. Secondly, a self-calibrated encoder with MSCFS is adopted to establish the spatial dependency between long distance and channel to avert the loss of both global and local features. Thirdly, a novel deep feature fusion decoder is designed to separately generate the edge and region features of the same and different levels and dynamically integrate them to restore fine-grained nodule edge morphology. Various experiments on LUNA16 public dataset confirm that MDFN outperforms other advanced methods on the quantitative index appraisal and qualitative visualization analysis, with an average DSC of 89.19% and Hausdorff distance of 2.353 mm. By contrast, with the powerful blend of multi-level edge and region features, MDFN achieves more smooth and consistent boundary details, especially for nodules adhesive to surrounding issues.

FSE-Net: feature selection and enhancement network for mammogram classification

Objective. Early detection and diagnosis allow for intervention and treatment at an early stage of breast cancer. Despite recent advances in computer aided diagnosis systems based on convolutional neural networks for breast cancer diagnosis, improving the classification performance of mammograms remains a challenge due to the various sizes of breast lesions and difficult extraction of small lesion features. To obtain more accurate classification results, many studies choose to directly classify region of interest (ROI) annotations, but labeling ROIs is labor intensive. The purpose of this research is to design a novel network to automatically classify mammogram image as cancer and no cancer, aiming to mitigate or address the above challenges and help radiologists perform mammogram diagnosis more accurately. Approach. We propose a novel feature selection and enhancement network (FSE-Net) to fully exploit the features of mammogram images, which requires only mammogram images and image-level labels without any bounding boxes or masks. Specifically, to obtain more contextual information, an effective feature selection module is proposed to adaptively select the receptive fields and fuse features from receptive fields of different scales. Moreover, a feature enhancement module is designed to explore the correlation between feature maps of different resolutions and to enhance the representation capacity of low-resolution feature maps with high-resolution feature maps. Main results. The performance of the proposed network has been evaluated on the CBIS-DDSM dataset and INbreast dataset. It achieves an accuracy of 0.806 with an AUC of 0.866 on the CBIS-DDSM dataset and an accuracy of 0.956 with an AUC of 0.974 on the INbreast dataset. Significance. Through extensive experiments and saliency map visualization analysis, the proposed network achieves the satisfactory performance in the mammogram classification task, and can roughly locate suspicious regions to assist in the final prediction of the entire images.

A sequential cross-product knowledge accumulation, extraction and transfer framework for machine learning-based production process modelling

ABSTRACT Machine learning is a promising method to model production processes and predict product quality. It is challenging to accurately model complex systems due to data scarcity, as mass customisation leads to various high-variety low-volume products. This study conceptualised knowledge accumulation, extraction, and transfer (KAET) to exploit the knowledge embedded in similar entities to address data scarcity. A sequential cross-product KAET (SeqTrans) is proposed to conduct KAET, integrating data preparation and preprocessing, feature selection (FS), feature learning (FL), and transfer learning (TL). The FS and FL modules conduct knowledge extraction and help address various practical challenges such as changing operating conditions and unbalanced datasets. In this paper, sequential TL is introduced to production modelling to conduct knowledge transfer among multiple entities. The first case study of auxetic material performance prediction demonstrates the effectiveness of sequential TL. Compared with conventional TL, sequential TL can achieve the same test mean square errors with 300 fewer training examples when facing data scarcity. In the second case study, balancing anomaly detection models were constructed for two gas turbines in the same series using real-world production data. With SeqTrans, the F1-score of the anomaly detection model of the data-poor engine was improved from 0.769 to 0.909.

Enhancing neural collaborative filtering using hybrid feature selection for recommendation.

The past decade has seen substantial growth in online transactions. Accordingly, many professionals and researchers utilize deep learning models to design and develop recommender systems to suit the needs of online personal services. These systems can model the interactions between users and items. However, existing approaches focus on either modeling global or local item correlation and rarely consider both cases, thus failing to represent user-item correlation very well. Therefore, this article proposes a deep collaborative recommendation system based on a convolutional neural network with an outer product matrix and a hybrid feature selection module to capture local and global higher-order interaction between users and items. Moreover, we incorporated the weights of generalized matrix factorization to optimize the overall network performance and prevent overfitting. Finally, we conducted extensive experiments on two real-world datasets with different sparsity to confirm that our proposed approach outperforms the baseline methods we have used in the experiment.

SSDStacked-BLS with Extended Depth and Width: Infrared Fault Diagnosis of Rolling Bearings under Dual Feature Selection

In fault diagnosis, broad learning systems (BLS) have been applied in recent years. However, the best fault diagnosis cannot be guaranteed by width node extension alone, so a stacked broad learning system (stacked BLS) was proposed. Most of the methods for choosing the number of depth layers used optimization algorithms that tend to increase computation time. In addition, the data under single feature selection are not sufficiently representative, and effective features are easily lost. To solve these problems, this article proposes an infrared fault diagnosis model for rolling bearings based on integration of principal component analysis and singular value decomposition (IPS) and the stacked BLS with self-selected depth model (SSDStacked-BLS). First, 72 second-order statistical features are extracted from the pre-processed infrared images of rolling bearings. Next, feature selection is performed using IPS. he IPS feature selection module consists of principal component analysis (PCA) and singular value decomposition (SVD). The feature selection is performed by PCA and SVD separately, which are then stitched together to form a new feature. This ensures a comprehensive coverage of infrared image features. Finally, the acquired features are input into SSDStacked-BLS. This model establishes a data storage group for the residual training characteristics of stacked BLS, adding one block at a time. The accuracy rate of each newly added block is output and saved to the data storage group. If the diagnostic rate fails to increase three consecutive times, the block stacking is stopped and the results are output. IPS-SSDStacked-BLS achieved an accuracy of 0.9667 in 0.1775 s. This is almost five times faster than stacked BLS optimized using the grid search method. Compared with the original BLS, its accuracy was 0.0445 higher and the time was approximated. Compared with IPS-SVM, IPS-RF, IPS-1DCNN and 2DCNN, IPS-SSDStacked-BLS was more advantageous in terms of accuracy and time consumption.

A combined prediction system for PM2.5 concentration integrating spatio-temporal correlation extracting, multi-objective optimization weighting and non-parametric estimation

Air pollution nowadays has seriously hindered the sustainable development. PM2.5 greatly affects air quality and human health, even facilitates virus transmission, making its concentration prediction is crucial. However, previous studies are limited to single PM2.5 concentration series, neglecting optimization of prediction stability, and lacking uncertainty analysis. To address these issues, this research proposed a combined PM2.5 prediction system (CPPs) based on modular concept. Firstly, the temporal and spatial correlations of PM2.5 were fully extracted by data pretreatment and feature selection modules. Subsequently, the results of single submodel prediction module were integrated by multi-objective slime mould algorithm in combination weighted module, achieving Pareto optimality theoretically. Eventually, interval forecasting module analyzed the predictive uncertainty. Notably, a truly accurate metric for predictive directional accuracy was proposed for the first time. Validating CPPs using PM2.5 data from Shanghai during COVID-19 epidemic showed superior performance in point and interval forecasts. This research achieves optimization of prediction accuracy and stability as well as uncertainty analysis based on multiple data sources, contributes to improved air quality and public health protection.

EpiSemble: A Novel Ensemble-based Machine-learning Framework for Prediction of DNA N6-methyladenine Sites Using Hybrid Features Selection Approach for Crops

Aim: The study aimed to develop a robust and more precise 6mA methylation prediction tool that assists researchers in studying the epigenetic behaviour of crop plants. Background: N6-methyladenine (6mA) is one of the predominant epigenetic modifications involved in a variety of biological processes in all three kingdoms of life. While in vitro approaches are more precise in detecting epigenetic alterations, they are resource-intensive and time-consuming. Artificial intelligence- based in silico methods have helped overcome these bottlenecks Methods: A novel machine learning framework was developed through the incorporation of four techniques: ensemble machine learning, hybrid approach for feature selection, the addition of features, such as Average Mutual Information Profile (AMIP), and bootstrap samples. In this study, four different feature sets, namely di-nucleotide frequency, GC content, AMIP, and nucleotide chemical properties were chosen for the vectorization of DNA sequences. Nine machine learning models, including support vector machine, random forest, k-nearest neighbor, artificial neural network, multiple logistic regression, decision tree, naïve Bayes, AdaBoost, and gradient boosting were employed using relevant features extracted through the feature selection module. The top three best-performing models were selected and a robust ensemble model was developed to predict sequences with 6mA sites. Results: EpiSemble, a novel ensemble model was developed for the prediction of 6mA methylation sites. Using the new model, an improvement in accuracy of 7.0%, 3.74%, and 6.65% was achieved over existing models for RiceChen, RiceLv, and Arabidopsis datasets, respectively. An R package, EpiSemble, based on the new model was developed and made available at https://cran.rproject. org/web/packages/EpiSemble/index.html. Conclusion: The EpiSemble model added AMIP as a novel feature, integrated feature selection modules, bootstrapping of samples, and ensemble technique to achieve an improved output for accurate prediction of 6mA sites in plants. To our knowledge, this is the first R package developed for predicting epigenetic sites of genomes in crop plants, which is expected to help plant researchers in their future explorations.

VitaDNet: A Deep Learning-Based Approach for Vitamin-D Deficiency Prediction

Vitamin D (VD) deficiency is a very common disease among elderly people. The lack of VD causes various diseases related to skin, eyes and throat. The previous epidemiological studies tried to predict the vitamin B6 and VD levels from the blood samples. Since this is laborious and time-taking, it is very difficult for the homely people to work on it. There is a strong requirement for the noninvasive method as there is a necessity to detect the deficiency at the early stage. Certain crucial parameters that could be used for analysis are based on the intake of anthropogenic parameters along with the commonly known body vitals. These parameters include the body mass index (BMI), waist circumference (WC), waist-to-height ratio (WHR) and body roundness index (BRI). The dataset used for the prediction of VD has been collected from 501 patients in the age of 40–75 years old. The prediction of VD levels in the body has various complications, like the sex, previous health records, inherent health conditions and body pathology. To consolidate all those parameters and to analyse, a robust model is required to associate the parameters which are used to predict the deficit of VD. A binary set of gated recurrent units (GRUs) are used along with the auto-encoders. The feature extraction and selection module in the network are composed of two different patch-based networks which makes the three-stage network robust. Despite these difficulties, the model is robust enough to predict the levels of VD in the body based on the anthropogenic parameters. To support this network, a sub-VitaDNet module is proposed based on the food taken. Through this network, the food taken is continuously observed and the levels of VD are predicted. Hence, the authors believe that the model is robust enough to predict the VD levels in the body.