Método de mineração de dados para identificação de câncer de mama baseado na seleção de variáveis

Holsbach, Nicole; Fogliatto, Flávio Sanson; Anzanello, Michel J.

doi:10.1590/1413-81232014194.01722013

Cited by 6 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These image processing steps include morphological feature computing (Sahiner et al 2001, Chen et al 2003, Armato and Sensakovic 2004, Joo et al 2004, Way et al 2006, Cheng et al 2010, which is still difficult to solve (Arbelaez et al 2011), and image decomposition (Sun et al 2013, Yang et al 2013, followed by statistical summaries and presentations (Sorensen 2010, Gmez et al 2012, Sun et al 2013, Yang et al 2013 for the calculation of textural features (Tourassi 1999). In feature integration by classifier, the widely used techniques are based on the KNN (k-nearest neighbor) method (Sahan et al 2007, Holsbach et al 2014, LDA (linear discriminant analysis) (Perez et al 2013(Perez et al , 2014 and SVM (support vector machines) (Akay et al 2009, Wang et al 2009, Krishnan et al 2010. The extraction of meaningful features is highly dependent on the quality of each intermediate result in the image processing steps (Tourassi 1999, Sahiner et al 2001, Joo et al 2004, Armato and Sensakovic 2004, Cheng et al 2010, which often requires recursive trial and error to obtain satisfactory results.…”

Section: Introductionmentioning

confidence: 99%

A deep learning framework for supporting the classification of breast lesions in ultrasound images

et al. 2017

View full text Add to dashboard Cite

In this research, we exploited the deep learning framework to differentiate the distinctive types of lesions and nodules in breast acquired with ultrasound imaging. A biopsy-proven benchmarking dataset was built from 5151 patients cases containing a total of 7408 ultrasound breast images, representative of semi-automatically segmented lesions associated with masses. The dataset comprised 4254 benign and 3154 malignant lesions. The developed method includes histogram equalization, image cropping and margin augmentation. The GoogLeNet convolutionary neural network was trained to the database to differentiate benign and malignant tumors. The networks were trained on the data with augmentation and the data without augmentation. Both of them showed an area under the curve of over 0.9. The networks showed an accuracy of about 0.9 (90%), a sensitivity of 0.86 and a specificity of 0.96. Although target regions of interest (ROIs) were selected by radiologists, meaning that radiologists still have to point out the location of the ROI, the classification of malignant lesions showed promising results. If this method is used by radiologists in clinical situations it can classify malignant lesions in a short time and support the diagnosis of radiologists in discriminating malignant lesions. Therefore, the proposed method can work in tandem with human radiologists to improve performance, which is a fundamental purpose of computer-aided diagnosis.

show abstract

Section: Introductionmentioning

confidence: 99%

A deep learning framework for supporting the classification of breast lesions in ultrasound images

et al. 2017

View full text Add to dashboard Cite

show abstract

“…A wide variety of machine learning classifiers have been developed for early diagnosis of breast cancer 16 17 18 . The widely used techniques are based on support vector machines (SVM) 18 19 20 , k-nearest neighbor (KNN) method 21 22 and linear discriminant analysis (LDA) 23 24 . However, the discriminative power of these methods is limited due to the computational costs of identifying definitive features for subset characterization and optimization.…”

mentioning

confidence: 99%

Discrimination of Breast Cancer with Microcalcifications on Mammography by Deep Learning

Wang

Yang

Cai

et al. 2016

Sci Rep

209

120

View full text Add to dashboard Cite

Microcalcification is an effective indicator of early breast cancer. To improve the diagnostic accuracy of microcalcifications, this study evaluates the performance of deep learning-based models on large datasets for its discrimination. A semi-automated segmentation method was used to characterize all microcalcifications. A discrimination classifier model was constructed to assess the accuracies of microcalcifications and breast masses, either in isolation or combination, for classifying breast lesions. Performances were compared to benchmark models. Our deep learning model achieved a discriminative accuracy of 87.3% if microcalcifications were characterized alone, compared to 85.8% with a support vector machine. The accuracies were 61.3% for both methods with masses alone and improved to 89.7% and 85.8% after the combined analysis with microcalcifications. Image segmentation with our deep learning model yielded 15, 26 and 41 features for the three scenarios, respectively. Overall, deep learning based on large datasets was superior to standard methods for the discrimination of microcalcifications. Accuracy was increased by adopting a combinatorial approach to detect microcalcifications and masses simultaneously. This may have clinical value for early detection and treatment of breast cancer.

show abstract

“…A primeira consistiu na validac ¸ão cruzada pelo técnica k -fold , que utilizou toda a base de dados e consistiu em realizar 'k' partic ¸ões (1 para teste e k-1 para treino), alternando os dados de treino e teste por 'k' vezes. Fez-se uso do k = 10 em todos os 6 modelos de classificac ¸ão [13]. Já na segunda configurac ¸ão, a base de dados foi dividida entre treino e teste, na seguinte proporc ¸ão: 20% para teste e 80% para o treinamento, pelo comando 'split', com uso do parâmetro 'stratify' e 'random state', sendo este último igual a 300 (escolha aleatória), para garantir que, cada vez que o código fosse inicializado, a divisão fosse a mesma.…”

Section: B Validac ¸ãO Cruzada E Divisão Treinamento/testeunclassified

Modelos de Aprendizado de Máquina Aplicados à Detecção de Câncer de Mama

Neto¹,

Silva²,

Campos³

et al. 2021

Anais Do 15. Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

O objetivo deste trabalho é construir um algoritmo classificador binário de câncer de mama, baseado em dados de exame de sangue e fisiológicos (Idade, Índice de Massa Corporal, Glicose, Insulina, Modelo de Avaliação da Homeostase, Leptina, Adiponectina, Resistina e Proteína Quimiotática de Monócitos-1) de 116 pacientes. Para tal estudo, realizou-se comparativo de desempenho dos seguintes modelos de aprendizado de máquina: Árvore de Decisão, Floresta Aleatória, K-Vizinhos mais Próximos, Redes Neurais Artificiais, Máquinas de Vetores de Suporte e Regressão Logística. Quanto a metodologia aplicada nos dados, teve-se duas: validação cruzada por k-fold (k=10); divisão dos dados em 80% treino e 20% teste. Para a primeira, avaliou-se média da acurácia e da sensibilidade. Na segunda, valores de acurácia, sensibilidade, especificidade e área sob a curva da característica de operação do receptor. Além disso, a partir do valor-p, por correlação de Pearson e Spearman, dependendo da variável de entrada, realizou-se o teste apenas com as variáveis com nível de significância de 5%, que foram: Glicose, Insulina, Resistina e Modelo de Avaliação da Homeostase. Como melhor classificador final, teve-se a Floresta Aleatória, no método treino/teste e com 9 variáveis, com 83,3% de acurácia, 100% de sensibilidade, 64% de especificidade e 0,881 de área sob a curva.

show abstract

Método de mineração de dados para identificação de câncer de mama baseado na seleção de variáveis

Abstract: Método de mineração de dados para identificação de câncer de mama baseado na seleção de variáveisA data mining method for breast cancer identification based on a selection of variables

Cited by 6 publications

References 20 publications

A deep learning framework for supporting the classification of breast lesions in ultrasound images

A deep learning framework for supporting the classification of breast lesions in ultrasound images

Discrimination of Breast Cancer with Microcalcifications on Mammography by Deep Learning

Modelos de Aprendizado de Máquina Aplicados à Detecção de Câncer de Mama

Contact Info

Product

Resources

About