Binary Classification Method Research Articles

Improving pre-movement pattern detection with filter bank selection

Objective. Pre-movement decoding plays an important role in detecting the onsets of actions using low-frequency electroencephalography (EEG) signals before the movement of an upper limb. In this work, a binary classification method is proposed between two different states. Approach. The proposed method, referred to as filter bank standard task-related component analysis (FBTRCA), is to incorporate filter bank selection into the standard task-related component analysis (STRCA) method. In FBTRCA, the EEG signals are first divided into multiple sub-bands which start at specific fixed frequencies and end frequencies that follow in an arithmetic sequence. The STRCA method is then applied to the EEG signals in these bands to extract CCPs. The minimum redundancy maximum relevance feature selection method is used to select essential features from these correlation patterns in all sub-bands. Finally, the selected features are classified using the binary support vector machine classifier. A convolutional neural network (CNN) is an alternative approach to select canonical correlation patterns. Main Results. Three methods were evaluated using EEG signals in the time window from 2 s before the movement onset to 1 s after the movement onset. In the binary classification between a movement state and the resting state, the FBTRCA achieved an average accuracy of 0.8968 ± 0.0847 while the accuracies of STRCA and CNN were 0.8228 ± 0.1149 and 0.8828 ± 0.0917, respectively. In the binary classification between two actions, the accuracies of STRCA, CNN, and FBTRCA were 0.6611 ± 0.1432, 0.6993 ± 0.1271, 0.7178 ± 0.1274, respectively. Feature selection using filter banks, as in FBTRCA, produces comparable results to STRCA. Significance. The proposed method provides a way to select filter banks in pre-movement decoding, and thus it improves the classification performance. The improved pre-movement decoding of single upper limb movements is expected to provide people with severe motor disabilities with a more natural, non-invasive control of their external devices.

Automatic Identification of Liquefaction Induced by 2021 Maduo Mw7.3 Earthquake Based on Machine Learning Methods

Rapid extraction of liquefaction induced by strong earthquakes is helpful for earthquake intensity assessment and earthquake emergency response. Supervised classification methods are potentially more accurate and do not need pre-earthquake images. However, the current supervised classification methods depend on the precisely delineated polygons of liquefaction by manual and landcover maps. To overcome these shortcomings, this study proposed two binary classification methods (i.e., random forest and gradient boosting decision tree) based on typical samples. The proposed methods trained the two machine learning methods with different numbers of typical samples, then used the trained binary classification methods to extract the spatial distribution of liquefaction. Finally, a morphological transformation method was used for the postprocessing of the extracted liquefaction. The recognition accuracies of liquefaction were estimated by four evaluation indices, which all showed a score of about 90%. The spatial distribution of liquefaction pits is also consistent with the formation principle of liquefaction. This study demonstrates that the proposed binary classification methods based on machine learning could efficiently and quickly provide the spatial distribution of liquefaction based on post-earthquake emergency satellite images.

A multiple criteria ensemble pruning method for binary classification based on D-S theory of evidence

A multiple criteria ensemble pruning method for binary classification based on D-S theory of evidence

New Labeling Methods for Deep Learning Real-Valued Inter-Residue Distance Prediction.

Much of the recent success in protein structure prediction has been a result of accurate protein contact prediction-a binary classification problem. Dozens of methods, built from various types of machine learning and deep learning algorithms, have been published over the last two decades for predicting contacts. Recently, many groups, including Google DeepMind, have demonstrated that reformulating the problem as a multi-class classification problem is a more promising direction to pursue. As an alternative approach, we recently proposed real-valued distance predictions, formulating the problem as a regression problem. The nuances of protein 3D structures make this formulation appropriate, allowing predictions to reflect inter-residue distances in nature. Despite these promises, the accurate prediction of real-valued distances remains relatively unexplored; possibly due to classification being better suited to machine and deep learning algorithms. Can regression methods be designed to predict real-valued distances as precise as binary contacts? To investigate this, we propose multiple novel methods of input label engineering, which is different from feature engineering, with the goal of optimizing the distribution of distances to cater to the loss function of the deep-learning model. Since an important utility of predicted contacts or distances is to build three-dimensional models, we also tested if predicted distances can reconstruct more accurate models than contacts. Our results demonstrate, for the first time, that deep learning methods for real-valued protein distance prediction can deliver distances as precise as binary classification methods. When using an optimal distance transformation function on the standard PSICOV dataset consisting of 150 representative proteins, the precision of 'top-all' long-range contacts improves from 60.9% to 61.4% when predicting real-valued distances instead of contacts. When building three-dimensional models we observed an average TM-score increase from 0.61 to 0.72, highlighting the advantage of predicting real-valued distances.

An imbalanced binary classification method based on contrastive learning using multi-label confidence comparisons within sample-neighbors pair

For imbalanced classification, data-level methods can achieve inter-class balance, but the samples generated do not contain new information and cannot avoid the problem of introducing noise. Algorithm-level methods may lead to overfitting of the model, and its classification effect is more dependent on the specific dataset and classification task, which means they lack universality. In addition, how to deeply mine the differences in the distribution of data overlap areas, and how to effectively mine the differences between categories when the absolute number of minority samples is small, are also important challenges in imbalanced classification. This paper proposes an imbalanced binary classification method using multi-label confidence comparisons based on contrastive learning. Different from the previous idea of directly learning its distribution characteristics from minority samples, combined with the idea of contrastive learning, the classification task is redefined as the multi-label matching task by mining the deep features that can represent the commonality and difference between the neighboring samples. Multiple differentiated contrastive sample groups are obtained through random sampling in its neighbor sample pool for each sample. This sample is combined with its contrastive sample groups to form multiple sample-neighbor pairs as training samples in the multi-label matching task. The original dataset is multiplied without introducing noise, laying a foundation for the effective mining of class differences when the absolute number of minority class samples is small. Based on the corresponding reconstruction error generated by Variational AutoEncoder (VAE), for sample-neighbor pairs, a multi-label matching loss between target samples and contrastive sample groups that integrates the idea of contrastive learning is designed. a robust classifier is obtained through simultaneous iterative learning of reconstruction error and multi-label matching loss, which can better mine the distribution differences of overlapping regions. In the testing phase, multiple different contrastive sample groups and the corresponding prediction results of the samples to be classified are obtained, which categories can be judged by integrating the predictions of each group for reverse reasoning. Experimental results on 38 public datasets show that the method outperforms typical imbalanced classification methods in both F1-measure and G-mean.

Predicting Chemical Carcinogens Using a Hybrid Neural Network Deep Learning Method.

Determining environmental chemical carcinogenicity is urgently needed as humans are increasingly exposed to these chemicals. In this study, we developed a hybrid neural network (HNN) method called HNN-Cancer to predict potential carcinogens of real-life chemicals. The HNN-Cancer included a new SMILES feature representation method by modifying our previous 3D array representation of 1D SMILES simulated by the convolutional neural network (CNN). We developed binary classification, multiclass classification, and regression models based on diverse non-congeneric chemicals. Along with the HNN-Cancer model, we developed models based on the random forest (RF), bootstrap aggregating (Bagging), and adaptive boosting (AdaBoost) methods for binary and multiclass classification. We developed regression models using HNN-Cancer, RF, support vector regressor (SVR), gradient boosting (GB), kernel ridge (KR), decision tree with AdaBoost (DT), KNeighbors (KN), and a consensus method. The performance of the models for all classifications was assessed using various statistical metrics. The accuracy of the HNN-Cancer, RF, and Bagging models were 74%, and their AUC was ~0.81 for binary classification models developed with 7994 chemicals. The sensitivity was 79.5% and the specificity was 67.3% for the HNN-Cancer, which outperforms the other methods. In the case of multiclass classification models with 1618 chemicals, we obtained the optimal accuracy of 70% with an AUC 0.7 for HNN-Cancer, RF, Bagging, and AdaBoost, respectively. In the case of regression models, the correlation coefficient (R) was around 0.62 for HNN-Cancer and RF higher than the SVM, GB, KR, DTBoost, and NN machine learning methods. Overall, the HNN-Cancer performed better for the majority of the known carcinogen experimental datasets. Further, the predictive performance of HNN-Cancer on diverse chemicals is comparable to the literature-reported models that included similar and less diverse molecules. Our HNN-Cancer could be used in identifying potentially carcinogenic chemicals for a wide variety of chemical classes.

Application of Advanced Hybrid Models to Identify the Sustainable Financial Management Clients of Long-Term Care Insurance Policy

The rapid growth of the aging population and the rate of disabled people with physical and mental disorders is increasing the demand for long-term care. The decline in family care could lead to social and economic collapse. In order to reduce the burden of long-term care, long-term care insurance has become one of the most competitive products in the life insurance industry. In the previous literature review, few scholars engaged in the research on this topic with data mining technology, which was motivated to trigger the formation of this study and hoped to increase the different aspects of academic research. The purpose of this study is to develop the long-term insurance business from the original list of insurance clients, to predict whether the sustainable financial management clients will buy the long-term care insurance policies, and to establish a feasible prediction model to assist life insurance companies. This study aims to establish the classified prediction models of Models I~X, to dismantle the data with the percentage split and 10-fold cross validation, plus the application of two kinds of technology as feature selection and data discretization, for the data mining of twenty-three kinds of algorithms in seven different categories (Bayes, Function, Lazy, Meta, Misc, Rule, and Decision Tree) through the data collected from the insurance company database, and to select 20 conditional attributes and 1 decisional attribute (whether to buy the long-term insurance policy or not). The decision attribute is binary classification method for empirical data analysis. The empirical results show that: (1) the marital status, total number of policies purchased, and total amount of policies (including long-term care insurance) are found to be the three important factors affecting the decision attribute; (2) the most stable models are the advanced hybrid Models V and X; and (3) the best classifier is Decision Tree J48 algorithm for the study data used.

Contrasting Profiles of Low-Performing Mathematics Students in Public and Private Schools in the Philippines: Insights from Machine Learning.

Filipino students performed poorly in the 2018 Programme for International Student Assessment (PISA) mathematics assessment, with more than 50% obtaining scores below the lowest proficiency level. Students from public schools also performed worse compared to their private school counterparts. We used machine learning approaches, specifically binary classification methods, to model the variables that best identified the poor performing students (below Level 1) vs. better performing students (Levels 1 to 6) using the PISA data from a nationally representative sample of 15-year-old Filipino students. We analyzed data from students in private and public schools separately. Several binary classification methods were applied, and the best classification model for both private and public school groups was the Random Forest classifier. The ten variables with the highest impact on the model were identified for the private and public school groups. Five variables were similarly important in the private and public school models. However, there were other distinct variables that relate to students’ motivations, family and school experiences that were important in identifying the poor performing students in each school type. The results are discussed in relation to the social and social cognitive experiences of students that relate to socioeconomic contexts that differ between public and private schools.

Survival prediction models: an introduction to discrete-time modeling

BackgroundPrediction models for time-to-event outcomes are commonly used in biomedical research to obtain subject-specific probabilities that aid in making important clinical care decisions. There are several regression and machine learning methods for building these models that have been designed or modified to account for the censoring that occurs in time-to-event data. Discrete-time survival models, which have often been overlooked in the literature, provide an alternative approach for predictive modeling in the presence of censoring with limited loss in predictive accuracy. These models can take advantage of the range of nonparametric machine learning classification algorithms and their available software to predict survival outcomes.MethodsDiscrete-time survival models are applied to a person-period data set to predict the hazard of experiencing the failure event in pre-specified time intervals. This framework allows for any binary classification method to be applied to predict these conditional survival probabilities. Using time-dependent performance metrics that account for censoring, we compare the predictions from parametric and machine learning classification approaches applied within the discrete time-to-event framework to those from continuous-time survival prediction models. We outline the process for training and validating discrete-time prediction models, and demonstrate its application using the open-source R statistical programming environment.ResultsUsing publicly available data sets, we show that some discrete-time prediction models achieve better prediction performance than the continuous-time Cox proportional hazards model. Random survival forests, a machine learning algorithm adapted to survival data, also had improved performance compared to the Cox model, but was sometimes outperformed by the discrete-time approaches. In comparing the binary classification methods in the discrete time-to-event framework, the relative performance of the different methods varied depending on the data set.ConclusionsWe present a guide for developing survival prediction models using discrete-time methods and assessing their predictive performance with the aim of encouraging their use in medical research settings. These methods can be applied to data sets that have continuous time-to-event outcomes and multiple clinical predictors. They can also be extended to accommodate new binary classification algorithms as they become available. We provide R code for fitting discrete-time survival prediction models in a github repository.

A Deep Learning-Based Compact Weighted Binary Classification Technique to Discriminate between Targets and Clutter in SAR Images

The proposed approach is a deep learning-based compact weighted binary classification (DL-CWBC) method to discriminate between targets and clutter in synthetic aperture radar (SAR) images. A new modified cross-entropy error function is proposed to improve the probability of detection by controlling the rate of false alarms (FAs). The unique feature of a CWBC algorithm is reducing the FA rate and maximizing the probability of target detection without missing any target. For pre-processing, targets and clutter are detected through a constant false alarm rate (CFAR) as a conventional detection algorithm. These are then manually divided into two classes. The classified targets and clutter were trained through a ResNet-101 network. There is a trade-off between the minimization of the FA rate and the maximization of the detection probability for targets of interest (TOIs). The weighted coefficient of the modified cross-entropy error function tries to maximize the performance of this trade-off. In addition, the proposed approach enables us not to miss any targets by an extreme distinction decision. Above all, the DL-CWBC algorithm performs very well despite its simplicity.

A Comparative Study of Machine-learning Methods for X-Ray Binary Classification

X-ray binaries (XRBs) consist of a compact object that accretes material from an orbiting secondary star. The most secure method we have for determining if the compact object is a black hole is to determine its mass: This is limited to bright objects and requires substantial time-intensive spectroscopic monitoring. With new X-ray sources being discovered with different X-ray observatories, developing efficient, robust means to classify compact objects becomes increasingly important. We compare three machine-learning classification methods (Bayesian Gaussian Processes (BGPs), K-Nearest Neighbors (KNN), Support Vector Machines) for determining whether the compact objects are neutron stars or black holes (BHs) in XRB systems. Each machine-learning method uses spatial patterns that exist between systems of the same type in 3D color–color–intensity diagrams. We used lightcurves extracted using 6 yr of data with MAXI/GSC for 44 representative sources. We find that all three methods are highly accurate in distinguishing pulsing from nonpulsing neutron stars (NPNS) with 95% of NPNS and 100% of pulsars accurately predicted. All three methods have high accuracy in distinguishing BHs from pulsars (92%) but continue to confuse BHs with a subclass of NPNS, called bursters, with KNN doing the best at only 50% accuracy for predicting BHs. The precision of all three methods is high, providing equivalent results over 5–10 independent runs. In future work, we will suggest a fourth dimension be incorporated to mitigate the confusion of BHs with bursters. This work paves the way toward more robust methods to efficiently distinguish BHs, NPNS, and pulsars.

Fock state-enhanced expressivity of quantum machine learning models

The data-embedding process is one of the bottlenecks of quantum machine learning, potentially negating any quantum speedups. In light of this, more effective data-encoding strategies are necessary. We propose a photonic-based bosonic data-encoding scheme that embeds classical data points using fewer encoding layers and circumventing the need for nonlinear optical components by mapping the data points into the high-dimensional Fock space. The expressive power of the circuit can be controlled via the number of input photons. Our work sheds some light on the unique advantages offered by quantum photonics on the expressive power of quantum machine learning models. By leveraging the photon-number dependent expressive power, we propose three different noisy intermediate-scale quantum-compatible binary classification methods with different scaling of required resources suitable for different supervised classification tasks.

Deep transfer learning-based fully automated detection and classification of Alzheimer's disease on brain MRI.

To employ different automated convolutional neural network (CNN)-based transfer learning (TL) methods for both binary and multiclass classification of Alzheimer's disease (AD) using brain MRI. Herein, we applied three popular pre-trained CNN models (ResNet101, Xception, and InceptionV3) using a fine-tuned approach of TL on 3D T1-weighted brain MRI from a subset of ADNI dataset (n = 305 subjects). To evaluate power of TL, the aforementioned networks were also trained from scratch for performance comparison. Initially, Unet network segmentedthe MRI scans into characteristic components of gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF). The proposed networks were trained and tested over the pre-processed and augmented segmented and whole images for both binary (NC/AD + progressive mild cognitive impairment (pMCI)+stable MCI (sMCI)) and 4-class (AD/pMCI/sMCI/NC) classification. Also, two independent test sets from the OASIS (n = 30) and AIBL (n = 60) datasets were used to externally assess the performance of the proposed algorithms. The proposed TL-based CNN models achieved better performance compared to the training CNN models from scratch. On the ADNI test set, InceptionV3-TL achieved the highest accuracy of 93.75% and AUC of 92.0% for binary classification, as well as the highest accuracy of 93.75% and AUC of 96.0% for multiclass classification of AD on the whole images. On the OASIS test set, InceptionV3-TL outperformed two other models by achieving 93.33% accuracy with 93.0% AUC in binary classification of AD on the whole images. On the AIBL test set, InceptionV3-TL also outperformed two other models in both binary and multiclass classification tasks on the whole MR images and achieved accuracy/AUC of 93.33%/95.0% and 90.0%/93.0%, respectively. The GM segment as input provided the highest performance in both binary and multiclass classification of AD, as compared to the WM and CSF segments. This study demonstrates the potential of applying deep TL approach for automated detection and classification of AD using brain MRI with high accuracy and robustness across internal and external test data, suggesting that these models can possibly be used as a supportive tool to assist clinicians in creating objective opinion and correct diagnosis. We used CNN-based TL approaches and the augmentation techniques to overcome the insufficient data problem. Our study provides evidence that deep TL algorithms can be used for both binary and multiclass classification of AD with high accuracy.

Social media data analysis framework for disaster response

This paper presents a social media data analysis framework applied to multiple datasets. The method developed uses machine learning classifiers, where filtering binary classifiers based on deep bidirectional neural networks are trained on benchmark datasets of disaster responses for earthquakes and floods and extreme flood events. The classifiers consist of learning from discrete handcrafted features and fine-tuning approaches using deep bidirectional Transformer neural networks on these disaster response datasets. With the development of the multiclass classification approach, we compare the state-of-the-art results in one of the benchmark datasets containing the largest number of disaster-related categories. The multiclass classification approaches developed in this research with support vector machines provide a precision of 0.83 and 0.79 compared to Bernoulli naïve Bayes, which are 0.59 and 0.76, and multinomial naïve Bayes, which are 0.79 and 0.91, respectively. The binary classification methods based on the MDRM dataset show a higher precision with deep learning methods (DistilBERT) than BoW and TF-IDF, while in the case of UnifiedCEHMET dataset show a high performance for accuracy with the deep learning method in terms of severity, with a precision of 0.92 compared to BoW and TF-IDF method which has a precision of 0.68 and 0.70, respectively.

Application of Gesture Estimation Method Based on Computer

This paper will study the computer vision pose estimation algorithm and apply it to fall detection. At present, the problem of aging cannot be ignored. There are more and more elderly people living alone. Due to their poor physical fitness, they are very easy to fall in daily life and bring great harm to the body and mind of the elderly. Therefore, they are in great need of means to monitor the daily life of the elderly in real time and make judgments on fall events. Driven by the rapid development of computer vision technology, human pose estimation has been widely used in fields such as motion recognition, motion posture scoring, and assisted rehabilitation therapy. The core of pose estimation is to detect the joint information and bone information of the human body from the image, and connect them to form a bone image to help the computer understand the human body information. The use of pose estimation for fall detection requires the extraction of human joint point information and bone information. In order to extract human joint point coordinates and bone images from video sequences, this paper has done a lot of research on computer vision pose estimation. First, the background subtraction method is studied. After that, the openpose pose estimation algorithm is studied. then studied the attitude estimation of Azure Kinect DK depth camera, focusing on its TOF principle and body tracking principle. After comparison, it is found that traditional pose estimation such as openpose is based on optical RGB images. In this paper, we mainly use Azure Kinect DK depth camera for fall detection. In the fall detection method, this paper uses the threshold method combined with pattern recognition method to detect falls. The thresholding method uses the head fall speed and the body tilt angle. The tilt angle of the upper body is used to solve the bending and squatting errors. The pattern recognition method transforms fall detection into two classification problems, fall and no-fall, and uses a support vector machine binary classification method combined with a directed gradient histogram to extract features from skeletal images for classification. The final experimental data show that this paper implements fall detection based on human posture estimation using Azure Kinect DK depth camera with high accuracy and practicality.

A novel diagnostic approach for the classification of small B-cell lymphoid neoplasms based on the NanoString platform

Small B-cell lymphoid neoplasms (SBCLNs) are a heterogeneous group of diseases characterized by malignant clonal proliferation of mature B-cells. However, the classification of SBCLNs remains a challenge, especially in cases where histopathological analysis is unavailable or those with atypical laboratory findings or equivocal pathologic data. In this study, gene expression profiling of 1039 samples from 27 gene expression omnibus (GEO) datasets was first investigated to select highly and differentially expressed genes among SBCLNs. Samples from 57 SBCLN cases and 102 nonmalignant control samples were used to train a classifier using the NanoString platform. The classifier was built by employing a cascade binary classification method based on the random forest algorithm with 35 refined gene signatures. Cases were successively classified as chronic lymphocytic leukemia/small lymphocytic lymphoma, conventional mantle cell lymphoma, follicular lymphoma, leukemic non-nodal mantle cell lymphoma, marginal zone lymphoma, lymphoplasmacytic lymphoma/Waldenström's macroglobulinemia, and other undetermined. The classifier algorithm was then validated using an independent cohort of 197 patients with SBCLNs. Under the distribution of our validation cohort, the overall sensitivity and specificity of proposed algorithm model were >95%, respectively, for all the cases with tumor cell content greater than 0.72. Combined with additional genetic aberrations including IGH-BCL2 translocation, MYD88 L265P mutation, and BRAF V600E mutation, the optimal sensitivity and specificity were respectively found at 0.88 and 0.98. In conclusion, the established algorithm demonstrated to be an effective and valuable ancillary diagnostic approach for the sub-classification and pathologic investigation of SBCLN in daily practice.

A generalised severity number to predict liquefaction damage with lateral spreading

This study introduces a generalisation of the classical one-dimensional liquefaction severity indexes to extend their predictive capability for the occurrence of lateral spreading. After a critical overview of the most used indexes, the rationale for extension to bi-dimensional conditions determined by non-horizontal geomorphology is presented, together with the rule to achieve generalisation. The efficacy of the new index is demonstrated with a performance-based study on two cases, the earthquakes of 20 May 2012 (Mw 6·1) at Terre del Reno (Emilia-Romagna, Italy) and 11 February 2011 (Mw = 6·2) at Christchurch (New Zealand). Stratigraphic attributes including thickness, depth, composition and relative density of the liquefiable layers, obtained over the whole territories from rich datasets of cone penetration tests, are coupled with topographic information derived from the digital elevation model to provide the input for the analysis. Consistency assessment and spatial interpolation of data are carried out with geostatistical tools implemented in a geographic information system platform. Validation against a post-earthquake damage survey, quantified with a binary classification method, shows the paramount role of the bi-dimensional conditions.

Automatic kidney segmentation using 2.5D ResUNet and 2.5D DenseUNet for malignant potential analysis in complex renal cyst based on CT images

Bosniak renal cyst classification has been widely used in determining the complexity of a renal cyst. However, it turns out that about half of patients undergoing surgery for Bosniak category III, take surgical risks that reward them with no clinical benefit at all. This is because their pathological results reveal that the cysts are actually benign not malignant. This problem inspires us to use recently popular deep learning techniques and study alternative analytics methods for precise binary classification (benign or malignant tumor) on Computerized Tomography (CT) images. To achieve our goal, two consecutive steps are required–segmenting kidney organs or lesions from CT images then classifying the segmented kidneys. In this paper, we propose a study of kidney segmentation using 2.5D ResUNet and 2.5D DenseUNet for efficiently extracting intra-slice and inter-slice features. Our models are trained and validated on the public data set from Kidney Tumor Segmentation (KiTS19) challenge in two different training environments. As a result, all experimental models achieve high mean kidney Dice scores of at least 95% on the KiTS19 validation set consisting of 60 patients. Apart from the KiTS19 data set, we also conduct separate experiments on abdomen CT images of four Thai patients. Based on the four Thai patients, our experimental models show a drop in performance, where the best mean kidney Dice score is 87.60%.

A novel multi-objective medical feature selection compass method for binary classification

The use of Artificial Intelligence in medical decision support systems has been widely studied. Since a medical decision is frequently the result of a multi-objective optimization problem, a popular challenge combining Artificial Intelligence and Medicine is Multi-Objective Feature Selection (MOFS). This article proposes a novel approach for MOFS applied to medical binary classification. It is built upon a Genetic Algorithm and a 3-Dimensional Compass that aims at guiding the search towards a desired trade-off between: Number of features, Accuracy and Area Under the ROC Curve (AUC). This method, the Genetic Algorithm with multi-objective Compass (GAwC), outperforms all other competitive genetic algorithm-based MOFS approaches on several real-world medical datasets. Moreover, by considering AUC as one of the objectives, GAwC guarantees the classification quality of the solution it provides thus making it a particularly interesting approach for medical problems where both healthy and ill patients should be accurately detected. Finally, GAwC is applied to a real-world medical classification problem and its results are discussed and justified both from a medical point of view and in terms of classification quality.

Skin cancer diagnosis using convolutional neural networks for smartphone images: A comparative study

In the past few years many researches investigated algorithms to diagnose skin cancer lesions where melanoma is the deadliest type of skin cancer. In this paper, we propose three different methods for binary classification (melanoma and nevus) using smartphone images since there is a lack in research on it. The applied methods depend on different versions of CNN architectures namely; Alex-net, Mobilenet-V2, and Resnet-50. These three methods are divided into two groups; one group is the three nets with transfer learning applied on the original data, and the second group is the same methods used for group one but applied on augmented data. The proposed methods are applied on PAD-UFES-20 which is a smartphone images dataset. The augmented data have improved the performance of each method relative to its performance for original data. The best performance is achieved when using Alex-net with transfer learning (TL) and data augmentation, as it achieved 99% for accuracy, 96.364% for sensitivity, 99.792% for specificity, 99.167% for precision, and 97.661% for F1-score. Mobilenet-V2 with TL on augmented dataset achieved promising results if it used as a mobile application as it achieved 94.071% for accuracy, 84.547% for sensitivity, 96.25% for specificity, 84.223% for precision, and 84.2% for F1-score. And finally, Resnet-50 with TL on augmented dataset achieved 94.918% for accuracy, 77.275% for sensitivity, 98.918% for specificity, 95.192% for precision, and 84.737% for F1-score.