Genetic Programming based Feature Manipulation for Skin Cancer Image Classification

<p>Skin image classification involves the development of computational methods for solving problems such as cancer detection in lesion images, and their use for biomedical research and clinical care. Such methods aim at extracting relevant information or knowledge from skin images that can significantly assist in the early detection of disease. Skin images are enormous, and come with various artifacts that hinder effective feature extraction leading to inaccurate classification. Feature selection and feature construction can significantly reduce the amount of data while improving classification performance by selecting prominent features and constructing high-level features. Existing approaches mostly rely on expert intervention and follow multiple stages for pre-processing, feature extraction, and classification, which decreases the reliability, and increases the computational complexity. Since good generalization accuracy is not always the primary objective, clinicians are also interested in analyzing specific features such as pigment network, streaks, and blobs responsible for developing the disease; interpretable methods are favored. In Evolutionary Computation, Genetic Programming (GP) can automatically evolve an interpretable model and address the curse of dimensionality (through feature selection and construction). GP has been successfully applied to many areas, but its potential for feature selection, feature construction, and classification in skin images has not been thoroughly investigated. The overall goal of this thesis is to develop a new GP approach to skin image classification by utilizing GP to evolve programs that are capable of automatically selecting prominent image features, constructing new high level features, interpreting useful image features which can help dermatologist to diagnose a type of cancer, and are robust to processing skin images captured from specialized instruments and standard cameras. This thesis focuses on utilizing a wide range of texture, color, frequency-based, local, and global image properties at the terminal nodes of GP to classify skin cancer images from multiple modalities effectively. This thesis develops new two-stage GP methods using embedded and wrapper feature selection and construction approaches to automatically generating a feature vector of selected and constructed features for classification. The results show that wrapper approach outperforms the embedded approach, the existing baseline GP and other machine learning methods, but the embedded approach is faster than the wrapper approach. This thesis develops a multi-tree GP based embedded feature selection approach for melanoma detection using domain specific and domain independent features. It explores suitable crossover and mutation operators to evolve GP classifiers effectively and further extends this approach using a weighted fitness function. The results show that these multi-tree approaches outperformed single tree GP and other classification methods. They identify that a specific feature extraction method extracts most suitable features for particular images taken from a specific optical instrument. This thesis develops the first GP method utilizing frequency-based wavelet features, where the wrapper based feature selection and construction methods automatically evolve useful constructed features to improve the classification performance. The results show the evidence of successful feature construction by significantly outperforming existing GP approaches, state-of-the-art CNN, and other classification methods. This thesis develops a GP approach to multiple feature construction for ensemble learning in classification. The results show that the ensemble method outperformed existing GP approaches, state-of-the-art skin image classification, and commonly used ensemble methods. Further analysis of the evolved constructed features identified important image features that can potentially help the dermatologist identify further medical procedures in real-world situations.</p>

Genetic Programming based Feature Manipulation for Skin Cancer Image Classification

<p>Skin image classification involves the development of computational methods for solving problems such as cancer detection in lesion images, and their use for biomedical research and clinical care. Such methods aim at extracting relevant information or knowledge from skin images that can significantly assist in the early detection of disease. Skin images are enormous, and come with various artifacts that hinder effective feature extraction leading to inaccurate classification. Feature selection and feature construction can significantly reduce the amount of data while improving classification performance by selecting prominent features and constructing high-level features. Existing approaches mostly rely on expert intervention and follow multiple stages for pre-processing, feature extraction, and classification, which decreases the reliability, and increases the computational complexity. Since good generalization accuracy is not always the primary objective, clinicians are also interested in analyzing specific features such as pigment network, streaks, and blobs responsible for developing the disease; interpretable methods are favored. In Evolutionary Computation, Genetic Programming (GP) can automatically evolve an interpretable model and address the curse of dimensionality (through feature selection and construction). GP has been successfully applied to many areas, but its potential for feature selection, feature construction, and classification in skin images has not been thoroughly investigated. The overall goal of this thesis is to develop a new GP approach to skin image classification by utilizing GP to evolve programs that are capable of automatically selecting prominent image features, constructing new high level features, interpreting useful image features which can help dermatologist to diagnose a type of cancer, and are robust to processing skin images captured from specialized instruments and standard cameras. This thesis focuses on utilizing a wide range of texture, color, frequency-based, local, and global image properties at the terminal nodes of GP to classify skin cancer images from multiple modalities effectively. This thesis develops new two-stage GP methods using embedded and wrapper feature selection and construction approaches to automatically generating a feature vector of selected and constructed features for classification. The results show that wrapper approach outperforms the embedded approach, the existing baseline GP and other machine learning methods, but the embedded approach is faster than the wrapper approach. This thesis develops a multi-tree GP based embedded feature selection approach for melanoma detection using domain specific and domain independent features. It explores suitable crossover and mutation operators to evolve GP classifiers effectively and further extends this approach using a weighted fitness function. The results show that these multi-tree approaches outperformed single tree GP and other classification methods. They identify that a specific feature extraction method extracts most suitable features for particular images taken from a specific optical instrument. This thesis develops the first GP method utilizing frequency-based wavelet features, where the wrapper based feature selection and construction methods automatically evolve useful constructed features to improve the classification performance. The results show the evidence of successful feature construction by significantly outperforming existing GP approaches, state-of-the-art CNN, and other classification methods. This thesis develops a GP approach to multiple feature construction for ensemble learning in classification. The results show that the ensemble method outperformed existing GP approaches, state-of-the-art skin image classification, and commonly used ensemble methods. Further analysis of the evolved constructed features identified important image features that can potentially help the dermatologist identify further medical procedures in real-world situations.</p>

Water Moth Search Algorithm-based Deep Training for Intrusion Detection in IoT

The economic growth and information technology leads to the development of Internet of Things (IoT) industry and has become the emerging field of research. Several intrusion detection techniques are introduced but the detection of intrusion and malicious activities poses a challenging task. This paper devises a novel method, namely the Water Moth Search algorithm (WMSA) algorithm, for training Deep Recurrent Neural Network (Deep RNN) to detect malicious network activities. The WMSA algorithm is newly devised by combining Water Wave optimization (WWO) and the Moth Search Optimization (MSO). The pre-processing is employed for the removal of redundant data. Then, the feature selection is devised using the Wrapper approach, then using the selected features; the Deep RNN classifier effectively detects the intrusion using the selected features. The proposed WMSA-based Deep RNN showed improved results with maximal accuracy, specificity, and sensitivity of 0.96, 0.973 and 0.960.

Efficient Algorithms for Dynamic Incomplete Decision Systems

Attribute reduction is a crucial problem in the process of data mining and knowledge discovery in big data. In incomplete decision systems, the model using tolerance rough set is fundamental to solve the problem by computing the redact to reduce the execution time. However, these proposals used the traditional filter approach so that the reduct was not optimal in the number of attributes and the accuracy of classification. The problem is critical in the dynamic incomplete decision systems which are more appropriate for real-world applications. Therefore, this paper proposes two novel incremental algorithms using the combination of filter and wrapper approach, namely IFWA_ADO and IFWA_DEO, respectively, for the dynamic incomplete decision systems. The IFWA_ADO computes reduct incrementally in cases of adding multiple objects while IFWA_DEO updates reduct when removing multiple objects. These algorithms are also verified on six data sets. Experimental results show that the filter-wrapper algorithms get higher performance than the other filter incremental algorithms.

Advanced feature selection to study the internationalization strategy of enterprises

Firms face an increasingly complex economic and financial environment in which the access to international networks and markets is crucial. To be successful, companies need to understand the role of internationalization determinants such as bilateral psychic distance, experience, etc. Cutting-edge feature selection methods are applied in the present paper and compared to previous results to gain deep knowledge about strategies for Foreign Direct Investment. More precisely, evolutionary feature selection, addressed from the wrapper approach, is applied with two different classifiers as the fitness function: Bagged Trees and Extreme Learning Machines. The proposed intelligent system is validated when applied to real-life data from Spanish Multinational Enterprises (MNEs). These data were extracted from databases belonging to the Spanish Ministry of Industry, Tourism, and Trade. As a result, interesting conclusions are derived about the key features driving to the internationalization of the companies under study. This is the first time that such outcomes are obtained by an intelligent system on internationalization data.

A hybrid Genetic Programming approach to feature detection and image classification

© 2015 IEEE. Image classification is a crucial task in Computer Vision. Feature detection represents a key component of the image classification process, which aims at detecting a set of important features that have the potential to facilitate the classification task. In this paper, we propose a Genetic Programming (GP) approach to image feature detection. The proposed method uses the Speeded Up Robust Features (SURF) method to extract features from regions automatically selected by GP, and adopts a wrapper approach combined with a voting scheme to perform image classification. The proposed approach is evaluated using three datasets of increasing difficulty, and is compared to five popularly used machine learning methods: Support Vector Machines, Random Forest, Naive Bayes, Decision Trees, and Adaptive Boosting. The experimental results show the proposed approach has achieved comparable or better performance than the five existing methods on all three datasets, and reveal its capability to automatically detect good regions from a large image from which good features are automatically constructed.

A hybrid Genetic Programming approach to feature detection and image classification

© 2015 IEEE. Image classification is a crucial task in Computer Vision. Feature detection represents a key component of the image classification process, which aims at detecting a set of important features that have the potential to facilitate the classification task. In this paper, we propose a Genetic Programming (GP) approach to image feature detection. The proposed method uses the Speeded Up Robust Features (SURF) method to extract features from regions automatically selected by GP, and adopts a wrapper approach combined with a voting scheme to perform image classification. The proposed approach is evaluated using three datasets of increasing difficulty, and is compared to five popularly used machine learning methods: Support Vector Machines, Random Forest, Naive Bayes, Decision Trees, and Adaptive Boosting. The experimental results show the proposed approach has achieved comparable or better performance than the five existing methods on all three datasets, and reveal its capability to automatically detect good regions from a large image from which good features are automatically constructed.