Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: disease classification

Sortuj według:

Ogranicz wyniki do:

Dual-attention EfficientNet based on multi-view feature fusion for cervical squamous intraepithelial lesions diagnosis

Guo Ying, Wang Yongxiong, Yang Huimin, Zhang Jiapeng, Sun Qing

Biocybernetics and Biomedical Engineering

2022

Vol. 42, no. 2

529--542

Cervicograms are widely used in cervical cancer screening but exhibit a high misdiagnosis rate. Even senior experts show only 48% specificity on clinical examinations. Most existing methods only use single-view images applied with acetic acid or Lugol’s iodine solution as their input data, ignoring the fact that non-pathological tissues may show false-positive reactions in these single-view images. This can lead to misdiagnosis in clinical diagnosis. Therefore, it is essential to extract features from multi-view colposcopy images (including the original images) as inputs, because three-view cervicograms provide complementary information. In this work, we propose an improved EfficientNet based on multi-view feature fusion for the automatic diagnosis of cervical squamous intraepithelial lesions. Specifically, EfficientNet-B0 is employed as the backbone network, and three-view images are taken as inputs by channel cascading to reduce misclassification. Additionally, we propose a dual-attention mechanism that implements the feature selection function based on Convolution Block Attention Module (CBAM) and Coordinate Attention (CA). These two attention mechanisms assist each other to enhance the feature representation of HSIL. We leverage a dataset of 3294 clinical cervigrams and obtain 90.0% accuracy with recall, specificity, and F1-Score of 87.1%, 93.0%, and 89.7%, respectively. Experimental results prove that this method can help clinicians with precise disease classification and diagnosis, and out-performs known related works.

Enhanced CAE system for detection of exudates and diagnosis of diabetic retinopathy stages in fundus retinal images using soft computing techniques

Janaki Sathya D., Geetha K.

Polish Journal of Medical Physics and Engineering

2019

Vol. 25, Iss. 2

131--139

Diabetic Retinopathy (DR) is one of the leading causes of visual impairment. Diabetic Retinopathy is the most recent technique of identifying the intensity of acid secretion in the eye for diabetic patients. The identification of DR is performed by visual analysis of retinal images for exudates (fat deposits) and the main patterns are traced by ophthalmologists. This paper proposes a fully automated Computer Assisted Evaluation (CAE) system which comprises of a set of algorithms for exudates detection and to classify the different stages of Diabetics Retinopathy, which are identified as either normal or mild or moderate or severe. Experimental validation is performed on a real fundus retinal image database. The segmentation of exudates is achieved using fuzzy C-means clustering and entropy filtering. An optimal set obtained from the statistical textural features (GLCM and GLHM) is extracted from the segmented exudates for classifying the different stages of Diabetics Retinopathy. The different stages of Diabetic Retinopathy are classified using three classifiers such as Back Propagation Neural Network (BPN), Probabilistic Neural Network (PNN) and Support Vector Machine (SVM). The experimental results show that the SVM classifiers outperformed other classifiers for the examined fundus retinal image dataset. The results obtained confirm that with new a set of texture features, the proposed methodology provides better performance when compared to the other methods available in the literature. These results suggest that our proposed method in this paper can be useful as a diagnostic aid system for Diabetic Retinopathy.