Recognizing Common Skin Diseases in the Philippines Using Image Processing and Machine Learning Classification

Surface electromyogram (sEMG) signals are widely used to control the myoelectric prosthetic arm for amputees. In this study, the authors investigated the usefulness of discrete wavelet transform (DWT) features from multiple levels of approximation and detail coefficients obtained from sEMG signals. DWT is used for de-noising as well as feature extraction in this study and further tested using Support Vector Machine (SVM) classifier and Artificial Neural Network (ANN) Classifier. The performance of SVM classifier is compared with ANN classifier. The classification accuracy of SVM classifiers was found better as compared to ANN in term of speed and robustness. In the first section of paper the authors presented the introduction related to the study. The experimental set-up for recording surface EMG signal is presented in section 2. In section 3 the experimental results are depicted and in section 4 the results are concluded.

Traffic Sign Recognition (TSR) system is a significant component of Intelligent Transport System (ITS) as traffic signs assist the drivers to drive more safely and efficiently. This paper represents a new approach for TSR system where detection of traffic sign is carried out using fuzzy rules based color segmentation method and recognition is accomplished using Speeded Up Robust Features(SURF) descriptor, trained by artificial neural network (ANN) classifier. In the detection step, the region of interest (sign area) is segmented using a set of fuzzy rules depending on the hue and saturation values of each pixel in the HSV color space, post processed to filter unwanted region. Finally the recognition of the traffic sign is implemented using ANN classifier upon the training of SURF features descriptor. The proposed system simulated on offline road scene images captured under different illumination conditions. The detection algorithm shows a high robustness and the recognition rate is quite satisfactory. The performance of the ANN model is illustrated in terms of cross entropy, confusion matrix and receiver operating characteristic (ROC) curves. Also, performances of some classifier such as Support Vector Machine (SVM), Decision Trees, Ensembles Learners (Adaboost) and K-Nearest Neighbor (KNN) classifier are assessed with ANN approach. The simulation results illustrate that recognition using ANN model is higher than classifiers stated above.

This paper presents an S-transform-based Electroencephalogram channel optimization and feature extraction methodology for monitoring mental vigilance level of humans. Vigilance level detection methodology consists of four steps. In the first stage, two types of Electroencephalogram signals (alert and drowsy) are acquired from 30 healthy subjects and decomposed into sub-bands using the S-transform. In the second stage, permutation entropy of the S-transform coefficients is calculated and Electroencephalogram channel optimization is performed. S-transform-based statistical features are computed from the optimized Electroencephalogram channels, in the third stage. In the fourth stage, artificial intelligence techniques such as Least Square-Support Vector Machine, Artificial Neural Network and Naive Bayes Classifier are used for the classification of Electroencephalogram signals using extracted features. The performance of the feature extraction methodology is tested on the Electroencephalogram data of 30 healthy subjects. Experimental results ensured the effectiveness of proposed methodology for the estimation of mental vigilance level by using Electroencephalogram signals. It is observed that the Artificial Neural Network classifier is a good candidate for pre-emptive automatic vigilance level detection system for Brain-Computer Interface applications.

MicroRNAs (miRNAs) have been found in diverse organisms and play critical role in gene expression regulations of many essential cellular processes. Discovery of miRNAs and identification of their target genes are fundamental to the study of such regulatory circuits. To distinguish the real pre-miRNA from other stem loop hairpins with similar stem loop (pseudo pre-miRNA) is an important task in molecular biology. From the analysis of experimentally proved pre-miRNAs, we identified 17 parameters for miRNA formation. These parameters are grouped into two categories: structural and thermodynamic properties of the pre-miRNAs. A set of feature vector was formed from the pre-miRNA-like hairpins of human, mouse and rat. A feed forward multi layer perceptron Artificial Neural Network (ANN) classifier is trained by these feature vectors. This classifier is an application program, that decide whether a given sequence is a pre-miRNA like hairpin sequence or not. If the sequence is a pre-miRNA like hairpin, then the ANN classifier will predict whether it is a real pre-miRNA or a pseudo premiRNA. The approach can classify correctly the precursors of Human Mouse and Rat, with an average sensitivity of 97.40% and specificity of 95.85%. When compared with previous approaches, MiPred, mR-abela, ProMiR and Triplet SVM classifier, current approach was greater in total accuracy.

Discusses the effects of the sample size on the estimates of the error rate of the artificial neural network (ANN) classifiers. Experimental results show that the standard deviation of the estimated error rate of ANN classifiers is independent of the hidden unit size. In addition, it is shown that nevertheless the class distributions are Gaussian, ANN classifiers outperform the quadratic discriminant function when sizes of samples per class are much unequal.

In this current research the aim of study is to predict changes happening in land by comparing an Innovative Artificial Neural Network (ANN) classifier and Mahalanobis Classifier (MC) by digital image processing and also comparing which algorithm gives more accuracy. For the years 2001 and 2011 Landsat7 ETM+(Enhanced Thematic Mapper plus) is used and Landsat 8 is used for 2021 of study region. These satellite images were classified into two groups which are ANN classifier and Mahalanobis classifier, each group contains 3 samples with a total of N=6 samples. The pretest power is assumed to be 80% and with alpha value of 0.05 and Confidence Interval of 95%. The land use and land cover changes have been analyzed with supervised classifiers and percentages of different types of region has been noted. An independent samples-t test from SPSS statistical analysis it was observed that from a single tail test <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{p} &gt; 0.05$</tex> hence there is no significance difference between two groups of classifiers, namely, ANN and MC. The mean and standard deviation of overall classification accuracy is <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$98.69\pm 1.24$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$91.13\pm 6.47$</tex> respectively. The mean and standard deviation for kappa coefficient is <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$0.97\pm 0.016$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$0.87\pm 0.076$</tex> for ANN and MC respectively. From this research, within the limits of the study, it can be concluded that Innovative Artificial Neural Network has performed better than Mahalanobis based classifier.

While many BSN applications require that sensor nodes be able to operate for extended periods of time, they also often require the wireless transmission of copious amounts of sensor data to a data aggregator or base station, where the raw data is processed into application-relevant information. The energy requirements of such streaming can be prohibitive, given the competing considerations of form factor and battery life requirements. Making intelligent decisions on the node about which data to store or transmit, and which to ignore, is a promising method of reducing energy consumption. Artificial neural network (ANN) classifiers are among several competitive techniques for such data selection. However, no systematic metrics exist for determining if an ANN classifier is suited for a particular resource constrained computing environment of a typical BSN node. An especially difficult task is assessing, at the design stage, which classifier architectures are feasible on a given resource-constrained node, what computational resources are required to execute a given classifier, and what classification performance might be achieved by a particular classifier on a given set of resources. This paper describes techniques for quantifying and predicting the performance of ANN classifiers on wearable sensor nodes using scalable synthetic test data. Additionally, the paper shows a comparison of synthetic data with gait data collected using an inertial BSN node, and classification results of the gait data using a cerebellar model arithmetic computer (CMAC) architecture show excellent agreement with theoretical predictions.

Classification is the process of recognition, differentiation and categorizing objects into groups. Breast abnormalities are calcifications which are tumor markers that indicate the presence of cancer in the breast. The aims of this research are to classify the types of breast abnormalities using artificial neural network (ANN) classifier and to evaluate the accuracy performance using receiver operating characteristics (ROC) curve. The methods used in this research are ANN for breast abnormalities classifications and Canny edge detector as a feature extraction method. Previously the ANN classifier provides only the number of benign and malignant cases without providing information for specific cases. However in this research, the type of abnormality for each image can be obtained. The existing MIAS MiniMammographic database classified the mammogram images into three features only namely characteristic of background tissues, class of abnormality and radius of abnormality. However, in this research three other features are added-in. These three features are number of spots, area and shape of abnormalities. Lastly the performance of the ANN classifier is evaluated using ROC curve. It is found that ANN has an accuracy of 97.9% which is considered acceptable.

How to solve multi-category image recognition and meet a certain accuracy is a key issue in the research of high-resolution remote sensing images, and it is of great significance. This article mainly studies artificial neural network in the classification of multi-source remote sensing images. This paper improves the efficiency and accuracy of image segmentation by studying the principle and implementation process of image segmentation algorithm from the two aspects of initial segmentation and region merging; secondly, it studies the method of object feature quantization and the image of different object features on the classification results; and finally, it selects BP neural network. The network classification method classifies the image objects and realizes the extraction and classification of high-resolution remote sensing images. Experiments in this paper show that for multi-source remote sensing image data, the overall accuracy of the two parallel classification algorithms is very similar, and both are close to 85%, which has a good classification effect. When performing large-scale image classification, the terrain types in the image will be more complicated. In this case, the extraction accuracy relative to the artificial neural network classification method will decrease, and the classification time will also become longer. This paper proves through experiments that the classification method of multi-source remote sensing images based on artificial neural network is feasible and has certain advantages.

Small-training sample size problems in artificial neural network classifier design are discussed. A comparison of the artificial neural network (ANN) and nonparametric statistical classifiers in small sample size situations is also presented in terms of the error probability.

Traffic Sign Recognition (TSR) system is a vital component of intelligent transport system. It plays an important role by enhancing the safety of the drivers, pedestrians and vehicles as traffic signs provide important information of the traffic environment of the road and assist the drivers to drive more safely and easily by guiding and warning. This paper represents road sign detection and recognition system based on speeded up robust feature (SURF) descriptor that is invariant to rotation, skew and occlusion of the sign and artificial neural network (ANN) classifier. The proposed system consists of three steps. In the first step the road scene image is segmented using color based thresholding algorithm followed by some post processing like morphological analysis and filtering using region properties to extract the road sign as a blob of region. Next, shape verification is carried out by using translation, scaling and rotation invariant features vector named Distance to Borders (DtBs) of the segmented blob. Finally the recognition of the traffic sign is implemented using ANN classifier with the training of SURF descriptor. The proposed system simulated on offline road scene images shows a high success rate in the recognition stage. The performance of the classifier is illustrated in terms of cross entropy, confusion matrix and receiver operating characteristic (ROC) curve.

Identification of ECG beats from cross-spectrum information aided learning vector quantization

For the localization of epileptic seizure, it has a guiding significance to detect high frequency oscillations (HFOs) which are novel biomarkers of the seizure onset zone. In this paper, a new method including Fuzzy entropy (FuzzyEn) and Fuzzy neural network (FNN) for automatic HFOs detection is proposed. FuzzyEn and short time energy which is calculated by mean square method are used as the inputs of the FNN classifier. The T-S interference model is applied to the consequent part of the fuzzy rules. The least squares estimation and error back propagation algorithms are used to adjust the consequent and antecedent parameters, respectively. The output of the FNN classifier is utilized to classify the signal as HFOs or as normal activity. The simulation results indicate that FuzzyEn can extract the HFOs features more effectively compared with approximate entropy. Besides, the sensitivity and specificity of the FNN classifier proposed in this paper are obviously higher than those of the artificial neural network (ANN) classifier which has the same number of hidden layers, neurons and epochs with the FNN classifier. It is clear that the classification performance of the FNN classifier is better compared with the ANN classifier.

Parkinson's disease (PD) is a neurological disorder which is progressive in nature. Although there is no cure to this disease, symptomatic treatments are available. These treatments can slow the progressive development of the symptoms. Medications can treat some of the symptoms of the PD up to a great extent that in turn may help the patients to live a normal life. Besides these medications, the patients can also be provided with various therapies based on the types of their symptoms. But for providing any treatment, detection of its symptoms at an early stage is very important. This can reduce its future complexities considerably. Early diagnosis along with proper medications may treat the symptoms of PD significantly. This motivates to propose a new and effective methodology for detection and analysis of PD. In this work, an approach has been proposed for identification of PD patients by using Electroencephalogram (EEG) signals. Here, the EEG signals of normal persons and PD patients are processed in three stages. First, the raw EEG signals are pre-processed for removal of noises and artefacts present. Out of various techniques, Wavelet transform is used for this purpose. In the MATLAB environment, de-noising can be executed by using the in-built functions. Performances of the de-noising techniques are examined with the performance parameters namely Root Mean Square Error (RMSE) as well as Signal to Noise Ratio (SNR). In the second stage, statistical features are extracted from the pre-processed EEG signals. In this work, five statistical features are considered for performing the classification. In the final stage, the extracted features are classified using Artificial Neural Network (ANN) and Support Vector Machine (SVM) techniques. ANN is an efficient classifier that predicts the human brain's working manners. SVM on the other hand has been proven as one of the most prevailing classification algorithms that gives highly accurate and robust results. This is a novel approach of analyzing the performances of the classification techniques by evaluating the best performing feature. In both the classifiers accuracy, precision, sensitivity and specificity are calculated from the confusion matrix evaluated from the values of the statistical features. In ANN, results using six different training algorithms at different hidden layers are calculated and compared. This proves the training algorithm Levenberg-Marquardt back-propagation with hidden layer 20 as the best combination for performing the classification. From the results it is seen that both ANN and SVM classifiers provide significant classification accuracies of 94.7% and 96.5% respectively. Amongst the five considered features, Mean performs the best in terms of classification accuracy.

A method to automatically detect epileptic seizure regions from long term EEG recordings using Support Vector Machine (SVM) and Artificial Neural Network (ANN) classifier is proposed in this paper. This method uses a combination of various features derived from multichannel EEG signals. Features are extracted from a 4 second window to create a feature vector. Classifier (SVM/ANN) is trained using feature vectors from a carefully chosen training set. Feature vectors from a new data set when fed to the trained models will give an output which is then processed using different rules to remove interictal spikes and correctly detect the seizure regions. Results of applying this on long term EEG recordings of 27 epileptic patients revealed that, the proposed method is capable of very high degree of discrimination between the interictal region and ictal(seizure) region. The proposed method is a generalized seizure detection method which is not patient specific and has an average detection accuracy of nearly 75%.