Two-phase deep convolutional neural network for reducing class skewness in histopathological images based breast cancer detection

Wahab, Noorul; Khan, Asifullah; Lee, Yeon Soo

doi:10.1016/j.compbiomed.2017.04.012

Cited by 138 publications

(61 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Recently, used CNN for the diagnosis of breast cancer images and results are compared against a network trained on a dataset containing handcrafted descriptors [211], [212]. Another recently proposed CNN based technique for breast cancer diagnosis is developed by Wahab et al [213]. In Wahab's work, two phases are involved.…”

Section: Cnn Based Image Classificationmentioning

confidence: 99%

A survey of the recent architectures of deep convolutional neural networks

et al. 2020

Self Cite

View full text Add to dashboard Cite

Deep Convolutional Neural Networks (CNNs) are a special type of Neural Networks, which have shown exemplary performance on several competitions related to Computer Vision and Image Processing. Interesting application areas of CNN include Image Classification and Segmentation, Object Detection, Video Processing, Natural Language Processing, Speech Recognition, etc. The powerful learning ability of deep CNN is largely due to the use of multiple feature extraction stages that can automatically learn representations from the data. Availability of a large amount of data and improvements in the hardware technology have accelerated the research in CNNs, and recently very interesting deep CNN architectures have been reported. In fact, several interesting ideas to bring advancements in CNNs have been explored such as the use of different activation and loss functions, parameter optimization, regularization, and architectural innovations. However, the major improvement in representational capacity of the deep CNN is achieved through architectural innovations. Especially, the idea of exploiting spatial and channel information, depth and width of architecture, and multi-path information processing has gained substantial attention. Similarly, the idea of using a block of layers as a structural unit is also gaining popularity. This survey thus focuses on the intrinsic taxonomy present in the recently reported deep CNN architectures and consequently, classifies the recent innovations in CNN architectures into seven different categories. These seven categories are based on spatial exploitation, depth, multi-path, width, feature-map exploitation, channel boosting, and attention. Additionally, the elementary understanding of CNN components, current challenges and applications of CNN are also provided. CNNs are the best among learning algorithms in understanding images content, and have shown exemplary results in segmentation, classification, detection, and retrieval related tasks [8], [9]. The success of CNNs has captured attention beyond academia. In industry, companies such as Google, Microsoft, AT&T, NEC, and Facebook have developed active research groups for exploring new architectures of CNN [10]. At present, most of the frontrunners of image processing and computer vision competitions are employing deep CNN based models.The attractive feature of CNN is its ability to exploit spatial or time correlation of the data. The topology of CNN is divided into multiple learning stages composed of a combination of the convolutional layers, non-linear processing units, and subsampling layers [11]. CNNs are feedforward multilayered hierarchical networks that are similar to fully connected neural network where each layer, using a bank of convolutional kernels, performs multiple transformations [12]. Convolution operation extracts useful features from locally correlated data points. Output of the convolutional kernels is assigned to non-linear processing unit (activation function), which not only helps in learning abstractions but also emb...

show abstract

Section: Cnn Based Image Classificationmentioning

confidence: 99%

A survey of the recent architectures of deep convolutional neural networks

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Deep learning is a recent development in machine learning and is used in many research fields, such as speech recognition (Martinez et al, 2017), drug discovery and toxicology (Tian et al, 2016), customer relationship management (Singh and Tucker, 2017), computer vision (Lu et al, 2017;Zhang et al, 2017), and bioinformatics (Wahab et al, 2017) with good results.…”

Section: Deep Learningmentioning

confidence: 99%

Deep stacked sparse auto-encoders for prediction of post-operative survival expectancy in thoracic lung cancer surgery

Iraji¹

2019

J Appl Biomed

View full text Add to dashboard Cite

Lung cancer is the leading cause of cancer death in men and women. The prognostic value of survival after lung cancer surgery has an important role in decision-making for surgeons and patients. The combination of clinical features and CT scan information for diagnosis, treatment and survival of patients with lung cancer increases the accuracy of prediction using machine learning. Therefore, creating a computer intelligent method with low error and high accuracy to predict survival is an important challenge, and it is beneficial for decreasing mortality from lung cancer, and for planning treatment. In this work, we implemented a deep stacked sparse auto-encoder (DSSAE) approach on a thoracic surgery data set for 470 patients, and our results contributing to deep learning based on 16 features were more precise than other suggested techniques for predicting postoperative survival expectancy in thoracic lung cancer surgery. The proposed method achieved a sensitivity of 94%, specificity of 82.86% and g-mean of 88.25%.

show abstract

“…Las redes neuronales convolucionales son un tipo de red neuronal que se caracteriza por la compartición de los pesos que se convolucionan a través de la entrada mediante una ventana de OCHOA, Y. | RIVAS, W. | TUSA, E. | MAZÓN, B. movimiento (Wahab, Khan, & Lee, 2017). Esta propiedad convolucional aplicada en una capa de agrupación alcanza invariancia traslacional, que se adapta especialmente a las imágenes.…”

Section: Red Neuronal Convolucional (Cnn)unclassified

“…De este modo, las salidas resultan de calcular el producto punto entre los pesos y las entradas (Qayyum, Anwar, Awais, & Majid, 2017). Luego se aplica la función de activación que introduce efectos no lineales sobre los elementos con el fin de producir un mapa de características, donde cada una de las entradas se considera como la salida de una sola neurona asociada a una pequeña región local de entrada (Wahab et al, 2017).…”

Section: Red Neuronal Convolucional (Cnn)unclassified

“…La capa de agrupación reduce la complejidad computacional de una CNN mediante la aplicación de una o más capas de agrupación a los mapas de características producidos por las capas convolucionales, disminuyendo el tamaño de los mapas. La capa de salida posee un número de nodos igual al número de clases en el conjunto de datos (Wahab et al, 2017). Los autores en (Soriano et al, 2018) utilizan una CNN para clasificar un banco de 1070 mamogramas obtenidos de la base de datos DDSM entre las clases benigno y maligno.…”

Section: Red Neuronal Convolucional (Cnn)unclassified

See 1 more Smart Citation

Revisión de esquemas de clasificación de tejidos cancerígenos en imágenes de mamografía

et al. 2018

View full text Add to dashboard Cite

RESUMENLa presente revisión de esquemas de clasificación analiza los algoritmos y métodos más utilizados dentro de la detección automática y clasificación supervisada de anomalías en imágenes de mamografías. Para ello, se propone el análisis de 36 artículos científicos en base a su metodología de clasificación, considerando además que estas publicaciones se encuentran indexadas en revistas reconocidas en los últimos cinco años. El estudio se ha dividido en cuatro etapas. La primera etapa es el preprocesamiento, que incluye las bases de datos utilizadas para entrenamiento y pruebas. Además, describimos los métodos de filtrados como la ecualización de histogramas (escala de grises) y el filtro Gabor, entre otros métodos que han brindado resultados eficientes. La etapa de los algoritmos de extracción de características presenta coincidencias en la utilización de la transformada discreta de coseno y la transformada discreta de ondícula entre otros algoritmos que se han utilizado para esta tarea. Para los algoritmos de clasificación, se resalta la utilización de las máquinas de vectores de soporte. En adición a la revisión de expresiones matemáticas desarrolladas en esta investigación, una de las contribuciones de este trabajo reside en proporcionar una lista de clasificadores híbridos, que resultan de la combinación de los clasificadores convencionales. Finalmente, se describen las métricas de evaluación como la curva receptora de características operativas y la precisión, que son indicadores cuantitativos del desempeño de los algoritmos revisados en este trabajo investigativo.PALABRAS CLAVE: Algoritmo de clasificación, cáncer de mama, extracción de características, mamografía, reconocimiento de patrones. ABSTRACTThe present revision of classification schemes analyzes the algorithms and methods most used within the automatic detection and supervised classification of abnormalities in mammography images. To do this, we propose the analysis of 36 scientific papers based on their classification methodology, considering also that these publications are indexed in recognized journals in the last five years. The study has been divided into four stages. The first stage is preprocessing, which includes databases used for training and testing. In addition, we describe the filtering methods such as histogram equalization (grayscale) and the Gabor filter, among other methods that have yielded efficient results. The stage of feature extraction algorithms presents coincidences in the use of the discrete cosine transform and the discrete wavelet transform, among other algorithms that have been used for this task. For the classification algorithms, we highlight the use of the support vector machine. In addition to the review of mathematical expressions developed in this research, one of the contributions of this paper is to provide a list of hybrid classifiers, which result from the combination of conventional classifiers. Finally, we describe the evaluation metrics such as the receiver operating characteristic curve and the precis...

show abstract

Two-phase deep convolutional neural network for reducing class skewness in histopathological images based breast cancer detection

Cited by 138 publications

References 43 publications

A survey of the recent architectures of deep convolutional neural networks

A survey of the recent architectures of deep convolutional neural networks

Deep stacked sparse auto-encoders for prediction of post-operative survival expectancy in thoracic lung cancer surgery

Revisión de esquemas de clasificación de tejidos cancerígenos en imágenes de mamografía

Contact Info

Product

Resources

About