This study aims to enable effective breast ultrasound image classification by combining deep features with conventional handcrafted features to classify the tumors. In particular, the deep features are extracted from a pre-trained convolutional neural network model, namely the VGG19 model, at six different extraction levels. The deep features extracted at each level are analyzed using a features selection algorithm to identify the deep feature combination that achieves the highest classification performance. Furthermore, the extracted deep features are combined with handcrafted texture and morphological features and processed using features selection to investigate the possibility of improving the classification performance. The cross-validation analysis, which is performed using 380 breast ultrasound images, shows that the best combination of deep features is obtained using a feature set, denoted by $CONV$ features that include convolution features extracted from all convolution blocks of the VGG19 model. In particular, the $CONV$ features achieved mean accuracy, sensitivity, and specificity values of $94.2\%$, $93.3\%$, and $94.9\%$, respectively. The analysis also shows that the performance of the $CONV$ features degrades substantially when the features selection algorithm is not applied. The classification performance of the $CONV$ features is improved by combining these features with handcrafted morphological features to achieve mean accuracy, sensitivity, and specificity values of $96.1\%$, $95.7\%$, and $96.3\%$, respectively. Furthermore, the cross-validation analysis demonstrates that the $CONV$ features and the combined $CONV$ and morphological features outperform the handcrafted texture and morphological features as well as the fine-tuned VGG19 model. The generalization performance of the $CONV$ features and the combined $CONV$ and morphological features is demonstrated by performing the training using the 380 breast ultrasound images and the testing using another dataset that includes 163 images. The results suggest that the combined $CONV$ and morphological features can achieve effective breast ultrasound image classifications that increase the capability of detecting malignant tumors and reduce the potential of misclassifying benign tumors.

Globally, Skin carcinoma is among the most lethal diseases. Millions of people are diagnosed with this cancer every year. Sill, early detection can decrease the medication cost and mortality rate substantially. The recent improvement in automated cancer classification using deep learning methods has reached a human-level performance requiring a large amount of annotated data assembled in one location, yet, finding such conditions usually is not feasible. Recently, federated learning (FL) has been proposed to train decentralized models in a privacy-preserved fashion depending on labeled data at the client-side, which is usually not available and costly. To address this, we propose FedPerl, a semi-supervised federated learning method. Our method is inspired by peer learning from educational psychology and ensemble averaging from committee machines. FedPerl builds communities based on clients' similarities. Then it encourages communities' members to learn from each other to generate more accurate pseudo labels for the unlabeled data. We also proposed the peer anonymization (PA) technique to anonymize clients. As a core component of our method, PA is orthogonal to other methods without additional complexity, and reduces the communication cost while enhances performance. Finally, we propose a dynamic peer learning policy that controls the learning stream to avoid any degradation in the performance, especially for the individual clients. Our experimental setup consists of 71,000 skin lesion images collected from 5 publicly available datasets. We test our method in four different scenarios in SSFL. With few annotated data, FedPerl is on par with a state-of-the-art method in skin lesion classification in the standard setup while outperforming SSFLs and the baselines by 1.8% and 15.8%, respectively. Also, it generalizes better to an unseen client while being less sensitive to noisy ones.

Medical image segmentation is one of the major challenges addressed by machine learning methods. However, these methods profoundly depend on a large amount of annotated data, which is expensive and time‐consuming. Semi‐supervised learning (SSL) approaches this by leveraging an abundant amount of unlabeled data. Recently, MixUp regularizer has been introduced to SSL methods by augmenting the model with new data points through linear interpolation at the input space. While this provides the model with new data, it is limited and may lead to inconsistent soft labels. It is argued that the linear interpolation at different representations provides the network with novel training signals and overcomes the inconsistency of the soft labels. This paper proposes ROAM as an SSL method that explores the manifold and performs linear interpolation on randomly selected layers to generate virtual data that has never been seen before, which encourages the network to be less confident for interpolated points. Hence it avoids overfitting, enhances the generalization, and shows less sensitivity to the domain shift. Extensive experiments are conducted on publicl datasets on whole‐brain and lung segmentation. ROAM achieves state‐of‐the‐art results in fully supervised (89.5%) and semi‐supervised (87.0%) settings with relative improvements up to 2.40% and 16.50%, respectively.

Globally, Skin carcinoma is among the most lethal diseases. Millions of people are diagnosed with this cancer every year. Sill, early detection can decrease the medication cost and mortality rate substantially. The recent improvement in automated cancer classification using deep learning methods has reached a human-level performance requiring a large amount of annotated data assembled in one location, yet, finding such conditions usually is not feasible. Recently, federated learning (FL) has been proposed to train decentralized models in a privacy-preserved fashion depending on labeled data at the client-side, which is usually not available and costly. To address this, we propose \verb!FedPerl!, a semi-supervised federated learning method. Our method is inspired by peer learning from educational psychology and ensemble averaging from committee machines. FedPerl builds communities based on clients' similarities. Then it encourages communities members to learn from each other to generate more accurate pseudo labels for the unlabeled data. We also proposed the peer anonymization (PA) technique to anonymize clients. As a core component of our method, PA is orthogonal to other methods without additional complexity and reduces the communication cost while enhancing performance. Finally, we propose a dynamic peer-learning policy that controls the learning stream to avoid any degradation in the performance, especially for individual clients. Our experimental setup consists of 71,000 skin lesion images collected from 5 publicly available datasets. We test our method in four different scenarios in SSFL. With few annotated data, FedPerl is on par with a state-of-the-art method in skin lesion classification in the standard setup while outperforming SSFLs and the baselines by 1.8% and 15.8%, respectively. Also, it generalizes better to unseen clients while being less sensitive to noisy ones.

Self-supervised Learning (SSL) has recently gained much attention due to the high cost and data limitation in the training of supervised learning models. The current paradigm in the SSL is to utilize data augmentation at the input space to create different views of the same images and train a model to maximize the representations between similar images and minimize them for different ones. While this approach achieves state-of-the-art (SOTA) results in various downstream tasks, it still lakes the opportunity to investigate the latent space augmentation. This paper proposes TriMix, a novel concept for SSL that generates virtual embeddings through linear interpolation of the data, thus providing the model with novel representations. Our strategy focuses on training the model to extract the original embeddings from virtual ones, hence, better representation learning. Additionally, we propose a self-consistency term that improves the consistency between the virtual and actual embeddings. We validate TriMix on eight benchmark datasets consisting of natural and medical images with an improvement of 2.71% and 0.41% better than the second-best models for both data types. Further, our approach outperformed the current methods in semi-supervised learning, particularly in low data regimes. Besides, our pre-trained models showed better transfer to other datasets.