Feature Density as an Uncertainty Estimator Method in the Binary Classification Mammography Images Task for a Supervised Deep Learning Model

Fuentes-Fino, Ricardo Javier; Calderón-Ramírez, Saúl; Domínguez, Enrique; López-Rubio, Ezequiel; Hernandez-Vasquez, Marco A.; Molina-Cabello, Miguel A.

doi:10.1007/978-3-031-07802-6_32

Cited by 4 publications

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Convolutional neural networks (CNNs) and transfer learning were shown in this research to achieve very high accuracy in histopathology image classification of breast cancer. [14] and [15] introduce new dimensions to breast cancer detection. Fuentes-Fino et al propose an uncertainty estimator method based on feature density, and Wu et al explore a few-shot learning scheme.…”

Section: A Optimization Techniques In Enhancing Model Performancementioning

confidence: 99%

ACNGCNN: Improving Efficiency of Breast Cancer Detection and Progression using Adversarial Capsule Network with Graph Convolutional Neural Networks

Pallapu,

Syed

2024

ijacsa

View full text Add to dashboard Cite

New diagnostic methods are needed to improve the accuracy and efficiency of breast cancer detection and progression. Although successful, current methods frequently lack precision, accuracy, and timeliness, especially in the early phases Of breast cancer progression. Our research proposes a new model using deep learning to improve breast cancer detection and classification, addressing constraints. Our breast cancer image and sample preprocessing approach combines a nonlocal means filter (NLM) and Generative Adversarial Networks (GAN). The model classifies datasets using LSTM with BiGRUbased Recurrent ShuffleNet V2, a highly efficient and accurate technique for sequential data samples. The integration of a Capsule Network with Graph Convolutional Neural Networks (CNGCNN) significantly improves breast cancer detection. This method was carefully tested on BreaKHis. The results were amazing, showing gains across multiple metrics: 4.9% greater precision, 3.5% higher accuracy, 3.4% higher recall, 2.5% higher AUC (Area Under the Curve), 1.9% higher specificity, and 3.4% decreased delay in the identification of breast cancer stages. Particularly striking was the model's performance in diagnosing illness development, where it displayed 3.5% greater precision, 3.9% higher accuracy, 4.5% higher recall, 3.4% higher AUC, 2.9% higher specificity, and 1.5% lower latency. Significant clinical impacts result from this work. Our methodology enables early diagnosis and precise staging of breast cancer, enabling focused therapies to improve patient outcomes and survival rates. The greater precision and reduced time lag in diagnosing disease progression also allow for more effective monitoring and treatment modifications. Overall, this study marks a considerable improvement in the field of breast cancer diagnostics, delivering a more efficient, accurate, and reliable tool for healthcare providers in their fight against this ubiquitous disease.

show abstract

Section: A Optimization Techniques In Enhancing Model Performancementioning

confidence: 99%

ACNGCNN: Improving Efficiency of Breast Cancer Detection and Progression using Adversarial Capsule Network with Graph Convolutional Neural Networks

Pallapu,

Syed

2024

ijacsa

View full text Add to dashboard Cite

show abstract

An uncertainty estimator method based on the application of feature density to classify mammograms for breast cancer detection

Fuentes-Fino,

Calderón-Ramírez,

Domínguez

et al. 2023

Neural Comput & Applic

View full text Add to dashboard Cite

In the area of medical imaging, one of the factors that can negatively influence the performance of prediction algorithms is the limited number of observations for each class within a labeled dataset. Usually, in order to increase the samples, a second set of unlabeled images is used. However, this set adds two new problems (i) finding patient observations with different pathologies than those observed in the labeled data set and (ii) finding images belonging to a different distribution from the dataset used in the model training process. This way, merging datasets from different sources can have an adverse effect on the distribution of features. Encountering this type of data (better known as out-of-distribution data) within the deployment environments may also lead to varying degrees of performance degradation as can be seen in the different experimental results obtained. In this research, a study of the behavior of Feature Density is made, as a mathematical model for the estimation of predictive uncertainty in supervised classification algorithms, in order to improve the behavior when out-of-distribution data are presented in the dataset. The Feature Density method is based on the estimation of feature density by means of histogram calculation (or Probability Density Function). The advantage of this method over the baseline approach (Mahalanobis distance) is that it does not assume a Gaussian-type distribution of sample characteristics and serves to estimate the uncertainty. This work focuses on the binary classification of mammography X-ray images from three different datasets simulating the condition of a different degree of contamination with out-of-distribution sample. According to the obtained results, the performance of the proposed method depends directly on the architecture of the implemented neural network.

show abstract

Uncertainty Estimation for Complex Text Detection in Spanish

Abreu-Cardenas,

Calderón-Ramírez,

Solís

2023

2023 IEEE 5th International Conference on BioInspired Processing (BIP)

View full text Add to dashboard Cite

Feature Density as an Uncertainty Estimator Method in the Binary Classification Mammography Images Task for a Supervised Deep Learning Model

Cited by 4 publications

References 15 publications

ACNGCNN: Improving Efficiency of Breast Cancer Detection and Progression using Adversarial Capsule Network with Graph Convolutional Neural Networks

ACNGCNN: Improving Efficiency of Breast Cancer Detection and Progression using Adversarial Capsule Network with Graph Convolutional Neural Networks

An uncertainty estimator method based on the application of feature density to classify mammograms for breast cancer detection

Uncertainty Estimation for Complex Text Detection in Spanish

Contact Info

Product

Resources

About