Pre-processing Part Research Articles

Detection of bone cancer based on a four-phase framework generative deep belief neural network in deep learning

The medical image processing plays an important role in analyzing and detecting tumors and detecting cancer cells in their early stages. Microscopic images were used as a deep learning approach to recognize bone features. These images were acquired using microscopy, a time-consuming and labor-intensive process that required repeating. Cancer cells cannot be identified by conventional techniques because of the noise in the images. To address these challenges, this paper introduces the Gated Convolutional Neural Network (GCNN) techniques used to identify multiclass bone cancer, including non-tumor, non-viable, and viable tumors. As a first stage, the most basic part of image Preprocessing is to deposing without interrupting the diagnostics information during the removal of noise. The second stage is edge detection, used for image brightness discontinuity detection based on Scaled Region Edge Detector (SRED). It is used for image segmentation and data extraction within regions based on the Screen Cluster Area Segmentation (SCAS) to segment the Region of Interest (RoI) based on Relative Scalar Color Histogram (RSCH). The fourth step is Feature extraction is to find out the features and match the image feature from the Bone Cancer Detection object based on the maximum threshold weights using GCNN with Maximum Standard External Regions (MSER) used for features extracting applied to improve feature quality analysis and Feature support rate. This results in higher classification accuracy, Precision, recall under F-measure, and lower error rates. This enables a more achieved classification than other existing methods.

Parallel quantized dual-level fully connected classifier for bearing fault diagnosis

The diagnosis of bearing failures in rotating mechanical equipment is critical to productivity and the quality of the manufacturing process. Traditional methods generally suffer from high human factor interference and low accuracy. This paper proposes a parallel quantized dual-level fully connected classifier (PQDFCC) for fault identification of bearings. The PQDFCC has two parts. The first part of the PQDFCC contains two types of data pre-processing. The first data pre-processing method is calculating absolute values and sorting the acquired raw signals. The second method of data pre-processing is quantizing the raw data. The second part of PQDFCC is a dual-level fully connected classifier. The first-level fully connected classifier performs the first-level fault classification on the quantized and sorted data; then, the output probability of the first-level classifier is considered as the input of the second-level fully connected classifier; finally, the second-level fully connected classifier provides the classification results of bearing faults. The data pre-processing part of the PQDFCC is a simple calculation that can simplify and not lose fault information; the dual-level fully connected classifier can fully exploit the hidden features of the data with high accuracy. The proposed PQDFCC is tested in the bearing failure, axial piston hydraulic pump, and self-priming centrifugal pump datasets; the experimental results show that the PQDFCC achieves 100% fault diagnosis accuracy.

Imagined speech classification exploiting EEG power spectrum features.

Imagined speech recognition has developed as a significant topic of research in the field of brain-computer interfaces. This innovative technique has great promise as a communication tool, providing essential help to those with impairments. An imagined speech recognition model is proposed in this paper to identify the ten most frequently used English alphabets (e.g., A, D, E, H, I, N, O, R, S, T) and numerals (e.g., 0 to 9). A novel electroencephalogram (EEG) dataset was created by measuring the brain activity of 30 people while they imagined these alphabets and digits. As part of signal preprocessing, EEG signals are filtered before extracting delta, theta, alpha, and beta band power features. These features are used as input for classification using support vector machines, k-nearest neighbors, and random forest (RF) classifiers. It is identified that the RF classifier outperformed the others in terms of classification accuracy. Classification accuracies of 99.38% and 95.39% were achieved at the coarse-level and fine-level, respectively with the RF classifier. From our study, it is also revealed that the beta frequency band and the frontal lobe of the brain played crucial roles in imagined speech recognition. Furthermore, a comparative analysis against state-of-the-art techniques is conducted to demonstrate the efficacy of our proposed model.

CytoPipeline and CytoPipelineGUI: a Bioconductor R package suite for building and visualizing automated pre-processing pipelines for flow cytometry data

BackgroundWith the increase of the dimensionality in flow cytometry data over the past years, there is a growing need to replace or complement traditional manual analysis (i.e. iterative 2D gating) with automated data analysis pipelines. A crucial part of these pipelines consists of pre-processing and applying quality control filtering to the raw data, in order to use high quality events in the downstream analyses. This part can in turn be split into a number of elementary steps: signal compensation or unmixing, scale transformation, debris, doublets and dead cells removal, batch effect correction, etc. However, assembling and assessing the pre-processing part can be challenging for a number of reasons. First, each of the involved elementary steps can be implemented using various methods and R packages. Second, the order of the steps can have an impact on the downstream analysis results. Finally, each method typically comes with its specific, non standardized diagnostic and visualizations, making objective comparison difficult for the end user.ResultsHere, we present CytoPipeline and CytoPipelineGUI, two R packages to build, compare and assess pre-processing pipelines for flow cytometry data. To exemplify these new tools, we present the steps involved in designing a pre-processing pipeline on a real life dataset and demonstrate different visual assessment use cases. We also set up a benchmarking comparing two pre-processing pipelines differing by their quality control methods, and show how the package visualization utilities can provide crucial user insight into the obtained benchmark metrics.ConclusionCytoPipeline and CytoPipelineGUI are two Bioconductor R packages that help building, visualizing and assessing pre-processing pipelines for flow cytometry data. They increase productivity during pipeline development and testing, and complement benchmarking tools, by providing user intuitive insight into benchmarking results.

Analyzing Sentiments using Optimized Novel Ensemble Fuzzy and DL based Approach with Efficient Feature Selection and Extraction Models

The most popular and active area of data mining study is sentiment analysis. Twitter is a crucial platform for collecting and distributing people’s thoughts, feelings, views, and attitudes regarding specific entities. There are several social media networks available today. In light of this, sentiment analysis in the Natural Language Processing (NLP) field became fascinating. Various methods for analyzing sentiment have been explored. However, improvements are still required regarding reliability and system efficacy. Additionally, user emotional expressions typically take the form of naturally occurring human-written textual data with numerous noises and ambiguities. The intricate contextual significance of sentiment expressions is difficult for present studies on sentiment analysis to precisely capture and interpret, particularly in linguistics with complex frameworks. To address these issues, we presented a new integrated fuzzy neural network. The proposed framework is developed for effective and efficient feature selection and hybrid approach based sentiment analysis. Ensemble of novel Deep Convolutional Neuro-Fuzzy Inference System (DCNFIS) and Deep Learning-based (DL) Long short-term memory neural network multilayer stacked bidirectional LSTM neural network analyzes the sentiments. The provided dataset is initially cleaned up and filtered out as part of preprocessing. Utilizing the preprocessed data, sentiment-based features are extracted using the inception-ResNet-V2 model. Then, the relevant features are selected by employing the Enhanced Reptile Search Algorithm (ERSA). The Al-Biruni Earth Radius (BER) optimization algorithm is used to optimize the hyperparameters of the ensemble approach, which analyzes the sentiment categories such as positive, negative, very positive, very negative, and neutral. Finally, an effectiveness assessment of the suggested and present classifiers is presented. Using three distinct research datasets, we conducted an experimental evaluation of the suggested model. While differentiating from the proposed approach, the proposed approach yields a greater performance of 98.97% recall, 99.06% precision, 99.13% accuracy and 99.01% F1-score. The experimental investigation analyzes that the proposed approach gains superior performance over existing approaches

Sign Language Classifier based on Machine Learning

Sign language represents an efficient way for individuals with hearing impairments to communicate. We propose a sign recognition system into which several tools are integrated to help with the image pre-processing part. By doing so, a machine learning model was developed that does not require a lot of processing power because instead of using the images themselves, it uses extracted data from them to connect this model to a mobile interface that the users will use to recognise signed letters successfully. The communication between the client and the model is sustained through a local server.

An Iterative Curve-Fitting Baseline Correction Method for Raman Spectra Driven by Neural Network.

Baseline correction is a vital part of spectral preprocessing, especially for Raman spectra. Iterative polynomial fitting is an easy but less accurate way to find baselines compared to other methods such as wavelet transform and penalized least squares (PLS) methods. The polynomial fitting methods can also get distorted results in certain conditions. In this paper, a neural network model for detecting the trend of the baseline was proposed to improve the correction accuracy of the fitting methods. The model selects the function basis according to the baseline trend instead of using a fixed polynomial function to match the baseline for a more precise fit. We also propose a way to generate simulation data, these data can be used to train the neural network model. The model provides reliable results for real spectral data with noise. Our method provides a new idea to correct the baseline with a strange shape. In addition, we examine the limitations of conventional iterative polynomial fitting, adaptive iteratively reweighted PLS and explain why our approach surpasses these methods.

Automated lung cancer diagnosis using swarm intelligence with deep learning

ABSTRACT In worldwide, lung cancer is amajor threatening issue for humans that increase the mortality rate. Here, the existing techniques are suffered from huge false-positiverates. Such kind of limitations encourages the researchers to designa novel automated lung cancer identification model for providing highly accurate outcome in the early phase. Thus, the major goal ofthis research work is to design and initiate the lung cancer identification framework using an enhanced deep learning model. The standard dataset attained for the analysis is offered to pre-processing part. Then, the Frangi filter is utilized for removingthe vessel from the image. The adoption of modified Fuzzy C-Means Clustering (FCM) with local thresholding is employed for the nodule segmentation, where parameter tuning is executed by Modified Dimension Range-based Cat Swarm Optimization (MDR-CSO). The resultant texture and shape features are subjected to the Enhanced Recurrent Neural Network (ERNN) for diagnosing lung cancer. Throughout theresult analysis, the accuracy and MCC rate of the developed model is95% and 91%. Thus, the result analysis of the offered method providesa better lung cancer detection rate than the classical techniquesthroughout the experimental analysis.

STRESS PREDICTION USING MACHINE LEARNING

Every year tens of millions of people suffer from depression and few of them get proper treatment on time. So, it is crucial to detect human stress and relaxation automatically via social media on a timely basis. It is very important to detect and manage stress before it goes into a severe problem. A huge number of informal messages are posted every day in social networking sites, blogs and discussion forums. This paper describes an approach to detect the stress using the information from social media networking sites, like tweeter.This paper presents a method to detect expressions of stress and relaxation on tweeter dataset i.e. working on sentiment analysis to find emotions or feelings about daily life. Sentiment analysis works the automatic extraction of sentiment related information from text. Here using TensiStrengthframework for sentiment strength detection on social networking sites to extract sentiment strength from the informal English text. TensiStrength is a system todetect the strength of stress and relaxation expressed in social media text messages. TensiStrength uses a lexical approach and a set of rules to detect direct and indirect expressions of stress or relaxation. This classifies both positive and negative emotions based on the strength scale from -5 to +5 indications of sentiments. Stressed sentences from the conversation are considered &categorised into stress and relax. TensiStrength is robust, it can be applied to a widevarietyofdifferent social web contexts. Theeffectiveness of TensiStrength depends on the nature of the tweets.In human being there is inborn capability to differentiate the multiple senses of an ambiguous word in a particular context, but machine executes only according to the instructions. The major drawback of machine translation is Word Sense Disambiguation. There is a fact that a single word can have multiple meanings or "senses." In the pre-processing part- of- speech disambiguation is analysed and the drawback of WSD overcomes in the proposed method by unigram, bigram and trigram to give better result on ambiguous words. Here, SVM with Ngram gives better resultPrecision is65% and Recall is 67% .But, the main objective of this technique is to find the explicit and implicit amounts of stress and relaxation expressed in tweets. Keywords: Stress Detection, Data Mining, TensiStrength, wordsense disambiguation.

Wind speed forecasting using deep learning and preprocessing techniques

ABSTRACT Most forecasting algorithms are tuned to a specific location or dataset and will not perform well in other situations. Some wind speed data might contain outliers, missed values, or noise which affects the forecasting performance tremendously. This paper proposes a hybrid-forecasting model that includes pre-processing and deep learning techniques to bridge the gap in developing a generic forecasting algorithm that is unspecific of location or dataset. The proposed model includes a preprocessing part that consists of missed value and outlier handling, decomposition, Kalman filtering, and smoothing. This is an important step because no matter how accurate the forecasting model is, results will vary tremendously if the dataset is corrupted. In addition to that, three different deep-base learning algorithms RNN, GRU, and LSTM will be used based on the characteristics of each subseries to reduce the complexity of the overall forecasting model. The proposed model performed the best across the seven tested sites from different locations with different climates and geography. Compared to other forecasting models such as LSTM standalone and EWT-LSTM, a performance improvement in accuracy by 50% as well as a 25% reduction in processing time was achieved with the proposed forecasting model

Car make and model recognition system using rear-lamp features and convolutional neural networks

Recognizing cars based on their features is a difficult task. We propose a solution that uses a convolutional neural network (CNN) and image binarization method for car make and model classification. Unlike many previous works in this area, we use a feature extraction method combined with a binarization method. In the first stage of the pre-processing part we normalize and change the size of an image. The image is then used to recognize where the rear-lamps are placed on the image. We extract the region and use the image binarization method. The binarized image is used as input to the CNN network that finds the features of a specific car model. We have tested the combinations of three different neural network architectures and eight binarization methods. The convolutional neural network with parameters of the highest quality metrics value is used to find the characteristics of the rear lamps on the binary image. The convolutional network is tested with four different gradient algorithms. We have tested the method on two data sets which differ in the way the images were taken. Each data set consists of three subsets of the same car, but is scaled to different image dimensions. Compared to related works that are based on CNN, we use rear view images in different position and light exposure. The proposed method gives better results compared to most available methods. It is also less complex, and faster to train compared to other methods. The proposed approach achieves an average accuracy of 93,9% on the first data set and 84,5% on the second set.

A Novel Denoising Algorithm Based on Wavelet and Non-Local Moment Mean Filtering

Denoising is the basis and premise of image processing and an important part of image preprocessing. Denoising can effectively improve image quality, which contributes to subsequent image processing such as image segmentation, feature extraction, and so on. In this paper, we propose a novel image denoising method based on wavelet transform and nonlocal moment mean filtering approach (NMM). The noisy image is firstly denoised by a wavelet-based soft-thresholding denoising technique and NMM is then utilized to further eliminate the rest noises. Meanwhile, the fusion of moment invariants increases the robustness of our denoising algorithm due to the invariance of image scaling, translation, and rotation of color moments. Experiments show that our algorithm achieves a better denoising effect compared with some other denoising approaches.

Preprocessing and Artificial Intelligence for Increasing Explainability in Mental Health

This paper shows the added value of using the existing specific domain knowledge to generate new derivated variables to complement a target dataset and the benefits of including these new variables into further data analysis methods. The main contribution of the paper is to propose a methodology to generate these new variables as a part of preprocessing, under a double approach: creating 2nd generation knowledge-driven variables, catching the experts criteria used for reasoning on the field or 3rd generation data-driven indicators, these created by clustering original variables. And Data Mining and Artificial Intelligence techniques like Clustering or Traffic light Panels help to obtain successful results. Some results of the project INSESS-COVID19 are presented, basic descriptive analysis gives simple results that even though they are useful to support basic policy-making, especially in health, a much richer global perspective is acquired after including derivated variables. When 2nd generation variables are available and can be introduced in the method for creating 3rd generation data, added value is obtained from both basic analysis and building new data-driven indicators.

Revealing the History and Mystery of RNA-Seq.

Advances in RNA-sequencing technologies have led to the development of intriguing experimental setups, a massive accumulation of data, and high demand for tools to analyze it. To answer this demand, computational scientists have developed a myriad of data analysis pipelines, but it is less often considered what the most appropriate one is. The RNA-sequencing data analysis pipeline can be divided into three major parts: data pre-processing, followed by the main and downstream analyses. Here, we present an overview of the tools used in both the bulk RNA-seq and at the single-cell level, with a particular focus on alternative splicing and active RNA synthesis analysis. A crucial part of data pre-processing is quality control, which defines the necessity of the next steps; adapter removal, trimming, and filtering. After pre-processing, the data are finally analyzed using a variety of tools: differential gene expression, alternative splicing, and assessment of active synthesis, the latter requiring dedicated sample preparation. In brief, we describe the commonly used tools in the sample preparation and analysis of RNA-seq data.

Salp swarm algorithm with iterative mapping and local escaping for multi-level threshold image segmentation: a skin cancer dermoscopic case study

AbstractIf found and treated early, fast-growing skin cancers can dramatically prolong patients’ lives. Dermoscopy is a convenient and reliable tool during the fore-period detection stage of skin cancer, so the efficient processing of digital images of dermoscopy is particularly critical to improving the level of a skin cancer diagnosis. Notably, image segmentation is a part of image preprocessing and essential technical support in the process of image processing. In addition, multi-threshold image segmentation (MIS) technology is extensively used due to its straightforward and effective features. Many academics have coupled different meta-heuristic algorithms with MIS to raise image segmentation quality. Nonetheless, these meta-heuristic algorithms frequently enter local optima. Therefore, this paper suggests an improved salp swarm algorithm (ILSSA) method that combines iterative mapping and local escaping operator to address this drawback. Besides, this paper also proposes the ILSSA-based MIS approach, which is triumphantly utilized to segment dermoscopic images of skin cancer. This method uses two-dimensional (2D) Kapur’s entropy as the objective function and employs non-local means 2D histogram to represent the image information. Furthermore, an array of benchmark function test experiments demonstrated that ILSSA could alleviate the local optimal problem more effectively than other compared algorithms. Afterward, the skin cancer dermoscopy image segmentation experiment displayed that the proposed ILSSA-based MIS method obtained superior segmentation results than other MIS peers and was more adaptable at different thresholds.

Design of Boosted Cascade and RTLBP Features based Face Recognition Module Using ML in OpenCV

Over the past ten years, face recognition in images has become more and more important. Since facial recognition is a natural talent for humans, it could be challenging to train a computer to recognize faces. The detecting difficulty comprises three stages. Without the requirement for picture difference or skin color identification, the detector operates in real-time at fifteen frames per second. Proposed and developed is a face recognition system based on Real Time Local Binary Pattern (RTLBP) and Boosted Cascade of Simple Features (BCSF). Preprocessing, face identification and recognition, and annotation are the three stages of the system. Segmentation, filtering, thresholding, and improving image quality are all part of preprocessing. The face, which is crucial for identification, is identified using the BCSF approach. Every face in the database has a correlation and RTLBP calculation. Correlation and RTLBP analysis between the captured face and the database face is then used to conduct recognition. One statistical technique that falls within the large category of factor analysis is the Boosted Cascade of Simple Features. By reducing the massive quantity of data stored to the extent of the feature space needed to represent the data efficiently, the BCSF seeks to achieve this goal. The RTLBP is designed to recognize faces by using a wide 1-D pixel vector consisting of a 2-D face picture in compact main parts of the space function. This is referred to as self-space projection. The identity of the vectors in the covariance matrix, which is centred on a set of fingerprint images, helps determine the proper space. Using OpenCV, Open Computer Vision (OpenCV), Python, Eigen-face, Fisher Face, LBPH, and Haar Cascade programming, an algorithm for a camera-based, real-time face recognition system is developed in this project. The name of the detected individual appears in the top left corner of a highlighted rectangle that annotates the face. The process of recognition takes 10.18 seconds, with a 99.93% success rate.

Saraiki Language Hybrid Stemmer Using Rule-Based and LSTM-Based Sequence-To-Sequence Model Approach

Converting a word to its original form, is called stemming, which is extremely important in the field of Natural language processing (NLP). It’s an integral part of the linguistic pre-processing of every Natural language processing application. Stemming converts inflectional word forms into their root word. Much work has been done for stemming in different national and regional languages like English, French, Arabic, German, Urdu, and Hindi. Many regional languages still need work to build digital resources using Natural language processing. Saraiki is one of the widely spoken regional languages in Pakistan. Almost eighty million people use this language for communication. There are very limited digital resources using the Saraiki language available to support advancement in Natural language processing technologies. The current research aims to propose a hybrid stemmer to stem Saraiki Work. The hybrid stemmer contains two hundred prefix and postfix rules and Long short-term memory based sequence-to-sequence model for converting Saraiki words into the stem. Firstly, Saraiki text * Corresponding Author: mubasher@isp.edu.pk was pre-processed, and a rule set was implemented. Secondly, the Long short-term memory based sequence-to-sequence model was deployed to stem the Saraiki word correctly. In the last step, The Saraiki Stemmer performance was evaluated by accurately finding stem word accuracy using a rule-set and Long short-term memory sequence to sequence model. After experiments, using the rule set correctly, stem word accuracy was 68.53%, while the Long short-term memory based sequence-to-sequence model produced 93.0% accuracy of correctly stem words. This work contributes significantly to the regional linguistic field by introducing stemmer for the Saraiki language.

Research on Mutual Information Feature Selection Algorithm Based on Genetic Algorithm

<p>Feature selection is an important part of data preprocessing. Feature selection algorithms that use mutual information as evaluation can effectively handle different types of data, so it has been widely used. However, the potential relationship between relevance and redundancy in the evaluation criteria is often ignored, so that effective feature subsets cannot be selected. Optimize the evaluation criteria of the mutual information feature selection algorithm and propose a mutual information feature selection algorithm based on dynamic penalty factors (Dynamic Penalty Factor Mutual Information Feature Selection Algorithm, DPMFS). The penalty factor is dynamically calculated with different selected features, so as to achieve a relative balance between relevance and redundancy, and effectively play the synergy between relevance and redundancy, and select a suitable feature subset. Experimental results verify that the DPMFS algorithm can effectively improve the classification accuracy of the feature selection algorithm. Compared with the traditional chi-square, MIM and MIFS feature selection algorithms, the average classification accuracy of the random forest classifier for the six standard datasets is increased by 3.73%, 3.51% and 2.44%, respectively.</p> <p> </p>

Auto Segmentation of Lymph Node Microscopy Images

The manual histology assessment on the biopsy tissue sample still remains the gold standard procedure for cancer and its progression in human body. Auto nuclei segmentation is an important method to measure cellularity but often suffered an issue due to the present of overlapping nuclei. The implementation of auto segmentation of cells could speed up the process of histology assessment for cancer cases. The first step to implement, a wide data profile of normal and cancerous need to be compile and analyze further as a reference guide. Tissue data profile can be collected based on cellularity property of the tissue which can be automatically segmented using MATLAB software. The objective of the study is to develop an auto nuclei segmentation using MATLAB software to measure cellularity between normal and cancerous cells of lymph node tissue. Histological images of the tissue were analyzed using MATLAB software by using thresholding method and the result was compared with ImageJ. The pre-processing part of the image processing incudes converting the image into 8-bit grayscale image. The segmentation parts include adaptive filtering to remove the noise using Wiener filter and the thresholding Otsu method. Results from the ImajeJ and manual counting on the cellularity shows a comparable results to the automated cellularity measured using MATLAB. The cellularity of the cancerous lymph nodes was found lower than normal lymph nodes.

Hyperspectral Image Denoising via Adversarial Learning

Due to sensor instability and atmospheric interference, hyperspectral images (HSIs) often suffer from different kinds of noise which degrade the performance of downstream tasks. Therefore, HSI denoising has become an essential part of HSI preprocessing. Traditional methods tend to tackle one specific type of noise and remove it iteratively, resulting in drawbacks including inefficiency when dealing with mixed noise. Most recently, deep neural network-based models, especially generative adversarial networks, have demonstrated promising performance in generic image denoising. However, in contrast to generic RGB images, HSIs often possess abundant spectral information; thus, it is non-trivial to design a denoising network to effectively explore both spatial and spectral characteristics simultaneously. To address the above issues, in this paper, we propose an end-to-end HSI denoising model via adversarial learning. More specifically, to capture the subtle noise distribution from both spatial and spectral dimensions, we designed a Residual Spatial-Spectral Module (RSSM) and embed it in an UNet-like structure as the generator to obtain clean images. To distinguish the real image from the generated one, we designed a discriminator based on the Multiscale Feature Fusion Module (MFFM) to further improve the quality of the denoising results. The generator was trained with joint loss functions, including reconstruction loss, structural loss and adversarial loss. Moreover, considering the lack of publicly available training data for the HSI denoising task, we collected an additional benchmark dataset denoted as the Shandong Feicheng Denoising (SFD) dataset. We evaluated five types of mixed noise across several datasets in comparative experiments, and comprehensive experimental results on both simulated and real data demonstrate that the proposed model achieves competitive results against state-of-the-art methods. For ablation studies, we investigated the structure of the generator as well as the training process with joint losses and different amounts of training data, further validating the rationality and effectiveness of the proposed method.