Tri-ResNet: A parallel Multi-branch and Transfer Learning Triple-input model for breast cancer detection and classification

Comprehensive Study for Breast Cancer Using Deep Learning and Traditional Machine Learning

Due to its high prevalence and wide dissemination, breast cancer is a particularly dangerous disease. Breast cancer survival chances can be improved by early detection and diagnosis. For medical image analyzers, diagnosing is tough, time-consuming, routine, and repetitive. Medical image analysis could be a useful method for detecting such a disease. Recently, artificial intelligence technology has been utilized to help radiologists identify breast cancer more rapidly and reliably. Convolutional neural networks, among other technologies, are promising medical image recognition and classification tools. This study proposes a framework for automatic and reliable breast cancer classification based on histological and ultrasound data. The system is built on CNN and employs transfer learning technology and metaheuristic optimization. The Manta Ray Foraging Optimization (MRFO) approach is deployed to improve the framework’s adaptability. Using the Breast Cancer Dataset (two classes) and the Breast Ultrasound Dataset (three-classes), eight modern pre-trained CNN architectures are examined to apply the transfer learning technique. The framework uses MRFO to improve the performance of CNN architectures by optimizing their hyperparameters. Extensive experiments have recorded performance parameters, including accuracy, AUC, precision, F1-score, sensitivity, dice, recall, IoU, and cosine similarity. The proposed framework scored 97.73% on histopathological data and 99.01% on ultrasound data in terms of accuracy. The experimental results show that the proposed framework is superior to other state-of-the-art approaches in the literature review.

Highlights A transfer learning strategy improved residue estimates from high-resolution RGB imagery. The best method used probabilistic estimates of expert classifiers to estimate residue cover. This research confirms the utility of RGB imagery to quantify residue cover in agricultural fields. Abstract. Plant residue on the soil surface increases the sustainability of food and fiber production in agricultural systems. Automated assessments of residue cover based on imagery have the potential to reduce labor and human bias associated with in-field measurements. We evaluate the capacity of a transfer learning strategy to improve the determination of residue level from high-resolution RGB images. The imagery for the study was collected from 88 field locations in 40 row crop fields in five Missouri counties between mid-April and early July in 2018 and 2019. At each field location, 50 contiguous 0.3 m × 0.2 m region of interest (ROI) images (ground sampling distance of 0.014 cm pixel--1) were extracted from imagery, resulting in a dataset of 4,400 ROI images; 3,000 were used for cross-validation and training (data collected in 2018) and 1,400 were used for testing (data collected in 2019). The percentage residue for each ROI image (ground truth) was determined by a bullseye grid method (n = 100). Features were extracted from ROI images using the VGGNet-16 CNN model, a pre-trained convolutional neural network model. We extracted 1,472 features per ROI using a global averaging and pooling strategy. The optimum feature set was identified using recursive feature elimination using a support vector machine (RFE-SVM). To estimate crop residue percentage using selected features, expert two-class SVMs were trained to separate adjacent levels of residue cover, where the rationale of the ensemble was to allow each of the two-class SVMs to find the hyperplanes that maximize the margin between the corresponding two consecutive classes. Based on the distance of the samples to these hyperplanes, probabilistic estimates of the data-point belonging to the class were computed. With the combined knowledge of probabilistic estimates from each expert classifier, the percentage crop residue cover of each ROI image was calculated. We tested our approach with 3-, 4-, 5-, and 8-class problems, achieving the best results with the 8-class problem with r2 = 0.93 at the ROI level, r2 = 0.97 at the field-location level, and minimal bias in residue estimates in low residue conditions. These results are superior to other reported estimates of percent residue derived from imagery. This research confirms the utility of high-resolution RGB imagery to quantify residue cover in agricultural systems. Keywords: Convolutional neural network, Soil erosion, Support vector machine, Transfer learning.

Breast cancer is intrusive form of cancer which affects every 1 woman out of 9 in Pakistan. To detect breast cancer at early stage, mammography technique is used which is a manual process and is susceptible to radiologist error. Therefore, this paper proposes a new CAD technique, which relies on customized deep convolutional neural network to detect and classify breast cancer into malignant and benign. Mammogram images from digital database for screening mammography dataset are used to train proposed model. First, region of interest is extracted using region based segmentation technique which is further enhanced using contrast limited adaptive histogram equalization. Later, a customized deep convolution neural network is used to learn features from mammograms. Support vector machine classifier is used to classify breast masses into benign and malignant. 88.7% accuracy is achieved with 0.885 area under the curve. Other parameters like System specificity, sensitivity, precision, F1 score and AUC are recorded as 0.93, 0.841, 0.917, 0.877 and 0.885 respectively.

Breast cancer is one of the most common types of cancer that kills women. When cells become uncontrollably large, cancer develops. As a result, detecting and classifying breast cancer in its early stages is essential so that patients can take the appropriate precautions. On the other hand, mammography images have relatively low sensitivity and effectiveness in detecting breast cancer. Furthermore, MRI (Magnetic Resonance Imaging) has higher detection sensitivity for breast cancer than mammography. In this research, a novel Radial Basis Function Networks model (RBFN) with a Mayfly Optimization Algorithm (MAO) mechanism has Breast MRI scans aid in the early detection of breast cancer. Following the system's training on Magnetic Resonance Imaging (MRI) breast images, a unique Contrast-Limited Adaptive Histogram Equalization (CLAHE) filter is developed for pre-processing noisy MRI image material. Backgrounds were removed before recovering breast cancer photos with a Contrast Limited Histogram Equalization (CLAHE) filter. Furthermore, the new study effort's performance is compared to earlier studies and this model is simulated using Python. The proposed model, RBFN-MAO, also outperforms previous models in terms of performance and precision with an accuracy of 97.54%. In comparison, it is 85.28, 80.95, 76.94, 85.39 and 90.32% for Convolution Neural Networks You Only Look Once (CNN-YOLO), Residual Networks (ResNet50), Diffusion Convolution Neural Networks (DCNN), Support Vector Machine (SVM) and Convolution Neural Network (CNN) models, respectively.

Automated assessments of residue cover based on imagery potentially reduce labor and human bias associated with in-field measurements. Our objective was to evaluate a transfer learning strategy to improve estimates of residue cover derived from high-resolution RGB images. The imagery for the project was collected from 88 locations in 40 row-crop fields in five Missouri counties between mid-April and early July in 2018 and 2019. At each field location, 50 contiguous 0.3 m x 0.2 m region of interest (ROI) images (ground sampling distance of 0.014 cm pixel^-1) were extracted from imagery resulting in a dataset of 4,400 ROI images; 3,000 used for cross validation and training (2018 data) and 1,400 used for testing (2019 data). Percent residue for each ROI image (ground truth) was determined by a bullseye grid method (n = 100; Lory et al. 2021). The ROI images were divided into the maximum number of 224 X 224 sub-images. Features were extracted from each sub-image using the VGGNet16 model, a convolutional neural network model. To reduce the number of features, we only used features from the five max-pooling layers and averaged them based on each kernel. Average values for each kernel were then averaged across all sub-images of the ROI resulting in a 1,472-feature dataset per ROI. Two feature selection strategies were then applied to the features: recursive feature elimination support vector machine classification (RFE-SVM) and forward regression feature selection (FRFS). Best locations outcomes were obtained with RFE-SVM (Table 1). <fig><graphic xlink:href=23092_files/23092-00.jpg id=A1BFF750-7E9F-4664-B86C-B9EE034A5779></graphic></fig> There was no apparent pattern of correlation among selected features and no overlap in outliers suggesting unique characteristics among the two selected feature sets. Results were superior to previous research based on the same data set using 70 manually extracted known features (Upadhyay et al., 2022). Results also compare favorably with other previous published research estimating of residue cover using high-resolution RGB images (r² between 0.75 and 0.90). Implementing a model derived from transfer learning is more computationally complex than implementing a model derived from known manually extracted features. Success of the transfer learning model documented there is an opportunity to improve upon the features developed to date by more conventional methods. Transfer learning through features extracted from VGGNet16 was a successful strategy for estimating residue cover from high-resolution RGB imagery.

Automated assessments of residue cover based on imagery potentially reduce labor and human bias associated with in-field measurements. Our objective was to evaluate a transfer learning strategy to improve estimates of residue cover derived from high-resolution RGB images. The imagery for the project was collected from 88 locations in 40 row-crop fields in five Missouri counties between mid-April and early July in 2018 and 2019. At each field location, 50 contiguous 0.3 m x 0.2 m region of interest (ROI) images (ground sampling distance of 0.014 cm pixel^-1) were extracted from imagery resulting in a dataset of 4,400 ROI images; 3,000 used for cross validation and training (2018 data) and 1,400 used for testing (2019 data). Percent residue for each ROI image (ground truth) was determined by a bullseye grid method (n = 100; Lory et al. 2021). The ROI images were divided into the maximum number of 224 X 224 sub-images. Features were extracted from each sub-image using the VGGNet16 model, a convolutional neural network model. To reduce the number of features, we only used features from the five max-pooling layers and averaged them based on each kernel. Average values for each kernel were then averaged across all sub-images of the ROI resulting in a 1,472-feature dataset per ROI. Two feature selection strategies were then applied to the features: recursive feature elimination support vector machine classification (RFE-SVM) and forward regression feature selection (FRFS). Best locations outcomes were obtained with RFE-SVM (Table 1). <fig><graphic xlink:href=23092_files/23092-00.jpg id=A1BFF750-7E9F-4664-B86C-B9EE034A5779></graphic></fig> There was no apparent pattern of correlation among selected features and no overlap in outliers suggesting unique characteristics among the two selected feature sets. Results were superior to previous research based on the same data set using 70 manually extracted known features (Upadhyay et al., 2022). Results also compare favorably with other previous published research estimating of residue cover using high-resolution RGB images (r² between 0.75 and 0.90). Implementing a model derived from transfer learning is more computationally complex than implementing a model derived from known manually extracted features. Success of the transfer learning model documented there is an opportunity to improve upon the features developed to date by more conventional methods. Transfer learning through features extracted from VGGNet16 was a successful strategy for estimating residue cover from high-resolution RGB imagery.

Early diagnosis of breast cancer is the most reliable and practical approach to mitigate cancer. Computer-aided detection or computer-aided diagnosis is one of the software technologies designed to assist doctors in detecting or diagnosing cancer and to reduce mortality using medical image analysis. Recently, Convolution Neural Networks became very popular in medical image analysis helping to process vast amount of data to detect and classify cancer in a fast and efficient manner. In this paper, we implemented deep neural networks ResNet18, InceptionV3 and ShuffleNet for binary classification of breast cancer in histopathological images. We have used networks pre-trained by the transfer learning on the ImageNet database and with fine-tuned output layers trained on histopathological images from the public dataset BreakHis. The highest average accuracy achieved for binary classification of benign or malignant cases was 98.73% for ResNet18, followed by 97.65% for ShuffleNet and 97.44% for Inception-V3Net.

A common disease suffered by Indonesian women is breast cancer. Early awareness of breast cancer is very important to minimize the negative impact and increase the chances of recovery for breast cancer patients. Breast cancer detection efforts using CT scan image technology. CT scan images provide a detailed picture of the internal structure of the breast, allowing the identification of pathological changes that may be early signs of breast cancer. The purpose of the study is to utilize CNN algorithm for breast cancer classification using CT scan images. The dataset used consists of three labels namely benign cancer, malignant cancer, normal. The three data sets consist of 1096 data. CNN is a type of algorithm in the field of artificial intelligence that has proven successful in pattern recognition on image data. The collected breast CT scan image dataset includes breast cancer and non-breast cancer cases. The data is used to train and test the CNN model. Furthermore, breast cancer classification through CT scans is carried out by applying the CNN method. The results of the research conducted obtained an accuracy of 97.26%. In Benign classification with precision 0.99 (99%), recall 0.96 (96%), f1-score 0.98 (98%), support 186, then Malignant classification with precision 93% or with points 0.93, recall 98% with points 0.98, and f1-score 96% with points 0.96, and support 202. The last is the normal classification with 99% precision with 0.99 points, 97% recall with 0.97 points, 98% f1-score with 0.93 points, and 269 support.

Breast cancer can be considered a deadly disease affecting women over the globe. Since severity of breast cancer is high at advanced stages, early detection processes find useful to increase the survival rate. The recently presented medical imaging modalities and deep learning (DL) models pave the way to design effective breast cancer, classification models. In this view, this study develops a new deep transfer learning enabled breast cancer detection and classification (DTL-BCDC) model using mammogram images. The proposed DTL-BDCD technique mainly intends to identify the presence of breast cancer. Primarily, the DTL-BDCD model involves contrast enhancement using CLAHE technique and adaptive weighted segmentation (AWS) technique is used for determining the infected regions. Besides, Densely Connected Networks (DenseNet-169) model is employed for feature extraction and multilayer perceptron (MLP) is utilized for breast cancer classification. The design of DenseNet169 with MLP model for breast cancer detection shows the novelty of the work. In order to demonstrate the enhanced outcomes of the DTL-BDCD model, a wide range of simulations take place on benchmark datasets and the comparison study reported the betterment of the DTL-BDCD model over the recent approaches.

Plant residue on the soil surface increases the sustainability food and fiber production in agricultural systems. Automated assessments of residue cover based on imagery has the potential to reduce labor and human bias associated with in-field measurements. Our objective was to evaluate the capacity of a transfer learning strategy to improve estimates of residue cover derived from high-resolution RGB images. The imagery for the project was collected from 88 locations in 40 row crop fields in five Missouri counties between mid-April and early July in 2018 and 2019. At each field location, 50 contiguous 0.3 m x 0.2 m region of interest (ROI) images (ground sampling distance of 0.014 cm pixel-1) were extracted from imagery resulting in a dataset of 4,400 ROI images; 3,000 used for cross validation and training (data collected in 2018) and 1,400 used for testing (data collected in 2019). The percent residue for each ROI image (ground truth) was determined by a bullseye grid method (n = 100). Features were extracted from ROI images using the VGGNet16 model, a convolutional neural network model. To reduce feature numbers, we averaged the features based on each kernel resulting 1,472 feature dataset per ROI. After the extraction, we compared three feature selection strategies: recursive feature elimination support vector machine classification (RFE-SVM), sequential forward feature selection classification (SFFS-SVM) and forward regression feature selection (FRFS). Best locations outcomes were obtained with RFE-SVM (r2 = 0.93, MAE = 4.9, with three outliers) and FRFS (r2 = 0.94, MAE = 5.2, with two outliers). The three models had no apparent pattern of correlation among selected features and limited overlap in outliers suggesting unique characteristics among the three selected feature sets. These results were superior to previous research based the same data set using 70 manually extracted known features. This suggested that transfer learning through features extracted from VGGNet16 pre-trained on ImageNet was a successful strategy for estimating residue cover. This research also confirmed the utility of high-resolution RGB imagery to quantify residue cover in agricultural systems.

Breast cancer is the most common cancer among women and the second leading cause of cancer-related death worldwide. Early detection through mammography is crucial, yet it often faces challenges such as false negatives and false positives that may cause more harm. This study develops a mammogram image classification model to distinguish between Cancer and Non-Cancer labels by integrating Convolutional Neural Network (CNN), Support Vector Machine (SVM), and an attention mechanism. DenseNet-201, Inception-v3, and VGG-16 were three CNN architectures that were employed as feature extractors, with preprocessing using Contrast Limited Adaptive Histogram Equalization (CLAHE) and data augmentation to enhance image quality and diversity. Four models were compared: CNN, CNN-SVM, CNN with attention mechanism, and CNN-SVM with attention mechanism. Evaluation metrics included accuracy, precision, recall, F1-score, F2-score, and the Friedman test to analyze statistical significance across models and data pipelines. The best performance was achieved by the CNN-SVM with attention mechanism based on DenseNet-201 trained on CLAHE-filtered data, reaching macro-average F1-score and F2-score of 95.2462%. Pipeline variations significantly influenced model performance. The Friedman test confirmed significant differences under the CLAHE pipeline, while the Wilcoxon test showed that CNN-SVM with attention mechanism outperformed CNN with attention mechanism.

Background: Breast cancer causes millions of deaths all over the world every year. It has become the most common type of cancer in women. Early detection will help in better prognosis and increase the chance of survival. Automating the classification using Computer-Aided Diagnosis (CAD) systems can make the diagnosis less prone to errors. Multi-class classification and Binary classification of breast cancer is a challenging problem. Convolutional neural network architectures extract specific feature descriptors from images, which cannot represent different types of breast cancer. This leads to false positives in classification, which is undesirable in disease diagnosis. Methods: The current paper presents an ensemble Convolutional neural network for multi-class classification and Binary classification of breast cancer. The feature descriptors from each network are combined to produce the final classification. In this paper, histopathological images are taken from the publicly available BreakHis dataset and classified into 8 classes. Results: The proposed ensemble model can perform better when compared to the methods proposed in the literature. The results showed that the proposed model could be a viable approach for breast cancer classification. Conclusion: In this paper, an approach for multi-class classification on the breast images for cancer detection is proposed. The proposed architecture can be a viable option for the classification of histopathology images.

Breast cancer is a serious medical condition and a leading cause of death among women. Early and accurate diagnosis is crucial for improving patient outcomes. This study explores the use of Convolutional Neural Networks (CNNs) with Transfer Learning using DenseNet121 and ResNet50 models to enhance breast cancer classification via mammography. Transfer Learning enables CNN models to leverage knowledge learned from larger datasets such as ImageNet to improve performance on specific breast cancer datasets. The dataset comprised medical images with three breast variations: benign, malignant, and normal, totaling 531 data points. Data was split with a 70% training and 30% validation ratio. Two CNN models, AlexNet and ResNet50, were evaluated to compare their performance in classifying these breast cancer types. The experimental results show that AlexNet achieved a training accuracy of 98.01%, while ResNet50 achieved 64.07%. AlexNet demonstrated superior performance in identifying complex patterns in mammography images, resulting in more accurate classification of different breast cancer types. These findings highlight the potential of deep learning applications to support more precise and effective medical diagnostics for breast cancer. This research contributes significantly to the development of AI technologies in healthcare aimed at improving early detection of breast cancer. The implications of this study could expand our understanding of Transfer Learning applications in medical contexts, driving further advancements in this field to enhance patient care and prognosis

The Deep Convolutional Neural Network (DCNN) is one of the most powerful and successful deep learning approaches. DCNNs have already provided superior performance in different modalities of medical imaging including breast cancer classification, segmentation, and detection. Breast cancer is one of the most common and dangerous cancers impacting women worldwide. In this paper, we have proposed a method for breast cancer classification with the Inception Recurrent Residual Convolutional Neural Network (IRRCNN) model. The IRRCNN is a powerful DCNN model that combines the strength of the Inception Network (Inception-v4), the Residual Network (ResNet), and the Recurrent Convolutional Neural Network (RCNN). The IRRCNN shows superior performance against equivalent Inception Networks, Residual Networks, and RCNNs for object recognition tasks. In this paper, the IRRCNN approach is applied for breast cancer classification on two publicly available datasets including BreakHis and Breast Cancer (BC) classification challenge 2015. The experimental results are compared against the existing machine learning and deep learning-based approaches with respect to image-based, patch-based, image-level, and patient-level classification. The IRRCNN model provides superior classification performance in terms of sensitivity, area under the curve (AUC), the ROC curve, and global accuracy compared to existing approaches for both datasets.