A Bi-FPN-Based Encoder-Decoder Model for Lung Nodule Image Segmentation.

Early detection and analysis of lung cancer involve a precise and efficient lung nodule segmentation in computed tomography (CT) images. However, the anonymous shapes, visual features, and surroundings of the nodules as observed in the CT images pose a challenging and critical problem to the robust segmentation of lung nodules. This article proposes a resource-efficient model architecture: an end-to-end deep learning approach for lung nodule segmentation. It incorporates a Bi-FPN (bidirectional feature network) between an encoder and a decoder architecture. Furthermore, it uses the Mish activation function and class weights of masks with the aim of enhancing the efficiency of the segmentation. The proposed model was extensively trained and evaluated on the publicly available LUNA-16 dataset consisting of 1186 lung nodules. To increase the probability of the suitable class of each voxel in the mask, a weighted binary cross-entropy loss of each sample of training was utilized as network training parameter. Moreover, on the account of further evaluation of robustness, the proposed model was evaluated on the QIN Lung CT dataset. The results of the evaluation show that the proposed architecture outperforms existing deep learning models such as U-Net with a Dice Similarity Coefficient of 82.82% and 81.66% on both datasets.

Skin lesion segmentation using two-phase cross-domain transfer learning framework.

BACKGROUND AND OBJECTIVE: Deep learning (DL) models have been used for medical imaging for a long time but they did not achieve their full potential in the past because of insufficient computing power and scarcity of training data. In recent years, we have seen substantial growth in DL networks because of improved technology and an abundance of data. However, previous studies indicate that even a well-trained DL algorithm may struggle to generalize data from multiple sources because of domain shifts. Additionally, ineffectiveness of basic data fusion methods, complexity of segmentation target and low interpretability of current DL models limit their use in clinical decisions. To meet these challenges, we present a new two-phase cross-domain transfer learning system for effective skin lesion segmentation from dermoscopic images. METHODS: Our system is based on two significant technical inventions. We examine a two- phase cross-domain transfer learning approach, including model-level and data-level transfer learning, by fine-tuning the system on two datasets, MoleMap and ImageNet. We then present nSknRSUNet, a high-performing DL network, for skin lesion segmentation using broad receptive fields and spatial edge attention feature fusion. We examine the trained model's generalization capabilities on skin lesion segmentation to quantify these two inventions. We cross-examine the model using two skin lesion image datasets, MoleMap and HAM10000, obtained from varied clinical contexts. RESULTS: At data-level transfer learning for the HAM10000 dataset, the proposed model obtained 94.63% of DSC and 99.12% accuracy. In cross-examination at data-level transfer learning for the Molemap dataset, the proposed model obtained 93.63% of DSC and 97.01% of accuracy. CONCLUSION: Numerous experiments reveal that our system produces excellent performance and improves upon state-of-the-art methods on both qualitative and quantitative measures.

A multi-class classification framework for disease screening and disease diagnosis of COVID-19 from chest X-ray images.

To accurately diagnose multiple lung diseases from chest X-rays, the critical aspect is to identify lung diseases with high sensitivity and specificity. This study proposed a novel multi-class classification framework that minimises either false positives or false negatives that is useful in computer aided diagnosis or computer aided detection respectively. To minimise false positives or false negatives, we generated respective stacked ensemble from pre-trained models and fully connected layers using selection metric and systematic method. The diversity of base classifiers was based on diverse set of false positives or false negatives generated. The proposed multi-class framework was evaluated on two chest X-ray datasets, and the performance was compared with the existing models and base classifiers. Moreover, we used LIME (Local Interpretable Model-agnostic Explanations) to locate the regions focused by the multi-class classification framework.

Explainable multi-module semantic guided attention based network for medical image segmentation.

Automated segmentation of medical images is crucial for disease diagnosis and treatment planning. Medical image segmentation has been improved based on the convolutional neural networks (CNNs) models. Unfortunately, they are still limited by scenarios in which the segmentation objective has large variations in size, boundary, position, and shape. Moreover, current CNNs have low explainability, restricting their use in clinical decisions. In this paper, we involve substantial use of various attentions in a CNN model and present an explainable multi-module semantic guided attention based network (MSGA-Net) for explainable and highly accurate medical image segmentation, which involves considering the most significant spatial regions, boundaries, scales, and channels. Specifically, we present a multi-scale attention module (MSA) to extract the most salient features at various scales from medical images. Then, we propose a semantic region-guided attention mechanism (SRGA) including location attention (LAM), channel-wise attention (CWA), and edge attention (EA) modules to extract the most important spatial, channel-wise, boundary-related features for interested regions. Moreover, we present a sequence of fine-tuning steps with the SRGA module to gradually weight the significance of interesting regions while simultaneously reducing the noise. In this work, we experimented with three different types of medical images such as dermoscopic images (HAM10000 dataset), multi-organ CT images (CHAOS 2019 dataset), and Brain tumor MRI images (BraTS 2020 dataset). Extensive experiments on all types of medical images revealed that our proposed MSGA-Net substantially increased the overall performance of all metrics over the existing models. Moreover, displaying the attention feature maps has more explainability than state-of-the-art models.

Automatic detection of COVID-19 from chest CT scan and chest X-Rays images using deep learning, transfer learning and stacking.

One of the promising methods for early detection of Coronavirus Disease 2019 (COVID-19) among symptomatic patients is to analyze chest Computed Tomography (CT) scans or chest x-rays images of individuals using Deep Learning (DL) techniques. This paper proposes a novel stacked ensemble to detect COVID-19 either from chest CT scans or chest x-ray images of an individual. The proposed model is a stacked ensemble of heterogenous pre-trained computer vision models. Four pre-trained DL models were considered: Visual Geometry Group (VGG 19), Residual Network (ResNet 101), Densely Connected Convolutional Networks (DenseNet 169) and Wide Residual Network (WideResNet 50 2). From each pre-trained model, the potential candidates for base classifiers were obtained by varying the number of additional fully-connected layers. After an exhaustive search, three best-performing diverse models were selected to design a weighted average-based heterogeneous stacked ensemble. Five different chest CT scans and chest x-ray images were used to train and evaluate the proposed model. The performance of the proposed model was compared with two other ensemble models, baseline pre-trained computer vision models and existing models for COVID-19 detection. The proposed model achieved uniformly good performance on five different datasets, consisting of chest CT scans and chest x-rays images. In relevance to COVID-19, as the recall is more important than precision, the trade-offs between recall and precision at different thresholds were explored. Recommended threshold values which yielded a high recall and accuracy were obtained for each dataset.

Deep learning-based improved snapshot ensemble technique for COVID-19 chest X-ray classification.

COVID-19 has proven to be a deadly virus, and unfortunately, it triggered a worldwide pandemic. Its detection for further treatment poses a severe threat to researchers, scientists, health professionals, and administrators worldwide. One of the daunting tasks during the pandemic for doctors in radiology is the use of chest X-ray or CT images for COVID-19 diagnosis. Time is required to inspect each report manually. While a CT scan is the better standard, an X-ray is still useful because it is cheaper, faster, and more widely used. To diagnose COVID-19, this paper proposes to use a deep learning-based improved Snapshot Ensemble technique for efficient COVID-19 chest X-ray classification. In addition, the proposed method takes advantage of the transfer learning technique using the ResNet-50 model, which is a pre-trained model. The proposed model uses the publicly accessible COVID-19 chest X-ray dataset consisting of 2905 images, which include COVID-19, viral pneumonia, and normal chest X-ray images. For performance evaluation, the model applied the metrics such as AU-ROC, AU-PR, and Jaccard Index. Furthermore, it also obtained a multi-class micro-average of 97% specificity, 95% f 1-score, and 95% classification accuracy. The obtained results demonstrate that the performance of the proposed method outperformed those of several existing methods. This method appears to be a suitable and efficient approach for COVID-19 chest X-ray classification.

A stacked ensemble for the detection of COVID-19 with high recall and accuracy.

The main challenges for the automatic detection of the coronavirus disease (COVID-19) from computed tomography (CT) scans of an individual are: a lack of large datasets, ambiguity in the characteristics of COVID-19 and the detection techniques having low sensitivity (or recall). Hence, developing diagnostic techniques with high recall and automatic feature extraction using the available data are crucial for controlling the spread of COVID-19. This paper proposes a novel stacked ensemble capable of detecting COVID-19 from a patient's chest CT scans with high recall and accuracy. A systematic approach for designing a stacked ensemble from pre-trained computer vision models using transfer learning (TL) is presented. A novel diversity measure that results in the stacked ensemble with high recall and accuracy is proposed. The stacked ensemble proposed in this paper considers four pre-trained computer vision models: the visual geometry group (VGG)-19, residual network (ResNet)-101, densely connected convolutional network (DenseNet)-169 and wide residual network (WideResNet)-50-2. The proposed model was trained and evaluated with three different chest CT scans. As recall is more important than precision, the trade-offs between recall and precision were explored in relevance to COVID-19. The optimal recommended threshold values were found for each dataset.

Cancer microarray data feature selection using multi-objective binary particle swarm optimization algorithm.

Cancer investigations in microarray data play a major role in cancer analysis and the treatment. Cancer microarray data consists of complex gene expressed patterns of cancer. In this article, a Multi-Objective Binary Particle Swarm Optimization (MOBPSO) algorithm is proposed for analyzing cancer gene expression data. Due to its high dimensionality, a fast heuristic based pre-processing technique is employed to reduce some of the crude domain features from the initial feature set. Since these pre-processed and reduced features are still high dimensional, the proposed MOBPSO algorithm is used for finding further feature subsets. The objective functions are suitably modeled by optimizing two conflicting objectives i.e., cardinality of feature subsets and distinctive capability of those selected subsets. As these two objective functions are conflicting in nature, they are more suitable for multi-objective modeling. The experiments are carried out on benchmark gene expression datasets, i.e., Colon, Lymphoma and Leukaemia available in literature. The performance of the selected feature subsets with their classification accuracy and validated using 10 fold cross validation techniques. A detailed comparative study is also made to show the betterment or competitiveness of the proposed algorithm.