Large Margin Local Estimate With Applications to Medical Image Classification

Deep radial basis function networks with subcategorization for mitosis detection in breast histopathology images

Due to the intra-class diversity of mitotic cells and the morphological overlap with similarly looking imposters, automatic mitosis detection in histopathology slides is still a challenging task. In this paper, we propose a novel mitosis detection model in a weakly supervised way, which consists of a candidate proposal network and a verification network. The candidate proposal network based on patch learning aims to separate both mitotic cells and their mimics from the background as candidate objects, which substantially reduces missed detections in the screening process of candidates. These obtained candidate results are then fed into the verification network for mitosis refinement. The verification network adopts an RBF-based subcategorization scheme to deal with the problems of high intra-class variability of mitosis and the mimics with similar appearance. We utilize the RBF centers to define subcategories containing mitotic cells with similar properties and capture representative RBF center locations through joint training of classification and clustering. Due to the lower intra-class variation within a subcategory, the localized feature space at subcategory level can better characterize a certain type of mitotic figures and can provide a better similarity measurement for distinguishing mitotic cells from nonmitotic cells. Our experiments manifest that this subcategorization scheme helps improve the performance of mitosis detection and achieves state-of-the-art results on the publicly available mitosis datasets using only weak labels.

Enhancing skin lesion classification with advanced deep learning ensemble models: a path towards accurate medical diagnostics

Skin cancer, including the highly lethal malignant melanoma, poses a significant global health challenge with a rising incidence rate. Early detection plays a pivotal role in improving survival rates. This study aims to develop an advanced deep learning-based approach for accurate skin lesion classification, addressing challenges such as limited data availability, class imbalance, and noise. Modern deep neural network designs, such as ResNeXt101, SeResNeXt101, ResNet152V2, DenseNet201, GoogLeNet, and Xception, which are used in the study and ze optimised using the SGD technique. The dataset comprises diverse skin lesion images from the HAM10000 and ISIC datasets. Noise and artifacts are tackled using image inpainting, and data augmentation techniques enhance training sample diversity. The ensemble technique is utilized, creating both average and weighted average ensemble models. Grid search optimizes model weight distribution. The individual models exhibit varying performance, with metrics including recall, precision, F1 score, and MCC. The "Average ensemble model" achieves harmonious balance, emphasizing precision, F1 score, and recall, yielding high performance. The "Weighted ensemble model" capitalizes on individual models' strengths, showcasing heightened precision and MCC, yielding outstanding performance. The ensemble models consistently outperform individual models, with the average ensemble model attaining a macro-average ROC-AUC score of 96 % and the weighted ensemble model achieving a macro-average ROC-AUC score of 97 %. This research demonstrates the efficacy of ensemble techniques in significantly improving skin lesion classification accuracy. By harnessing the strengths of individual models and addressing their limitations, the ensemble models exhibit robust and reliable performance across various metrics. The findings underscore the potential of ensemble techniques in enhancing medical diagnostics and contributing to improved patient outcomes in skin lesion diagnosis.

Generative and Discriminative Learning for Lung X-Ray Analysis Based on Probabilistic Component Analysis

Introduction

The paper presents a hybrid generative/discriminative classification method aimed at identifying abnormalities, such as cancer, in lung X-ray images.

Methods

The proposed method involves a generative model that performs generative embedding in Probabilistic Component Analysis (PrCA). The primary goal of PrCA is to model co-existing information within a probabilistic framework, with the intent to locate the feature vector space for X-ray data based on a defined kernel structure. A kernel-based classifier, grounded in information-theoretic principles, was employed in this study.

Results

The performance of the proposed method is evaluated against nearest neighbour (NN) classifiers and support vector machine (SVM) classifiers, which use a diagonal covariance matrix and incorporate normal linear and non-linear kernels, respectively.

Discussion

The method is found to achieve superior accuracy, offering a viable solution to the class of problems presented. Accuracy rates achieved by the kernels in the NN and SVM models were 95.02% and 92.45%, respectively, suggesting the method's competitiveness with state-of-the-art approaches.

Skin Lesion Segmentation Using an Ensemble of Different Image Processing Methods

In recent times, there has been a huge increase in the average number of cases of skin cancer per year, which sometimes become life threatening for humans. Early detection of various skin diseases through automated detection techniques plays a crucial role. However, the presence of numerous artefacts makes this task challenging. Dermoscopic images exhibit various variations, including hair artefacts, markers, and ill-defined boundaries. These artefacts make automatic analysis of skin lesion quite a difficult task. To address these issues, it is essential to have an accurate and efficient automated method which will delineate a skin lesion from the rest of the image. Unfortunately, due to the presence of several types of skin artefacts, there is no such thresholding method that can provide a sufficient segmentation result for every type of skin lesion. To overcome this limitation, an ensemble-based method is proposed that selects the optimal thresholding based on an objective function. A group of state-of-the-art different thresholding methods such as Otsu, Kapur, Harris hawk, and grey level are used. The proposed method obtained superior results (dice score = 0.89 with p-value ≤ 0.05) as compared to other state-of-the-art methods (Otsu = 0.79, Kapur = 0.80, Harris hawk = 0.60, grey level = 0.69, active contour model = 0.72). The experiments conducted in this study utilize the ISIC 2016 dataset, which is publicly available and specifically designed for skin-related research. Accurate segmentation will help in the early detection of many skin diseases.

Semi-Supervised Segmentation of Interstitial Lung Disease Patterns from CT Images via Self-Training with Selective Re-Training

Accurate segmentation of interstitial lung disease (ILD) patterns from computed tomography (CT) images is an essential prerequisite to treatment and follow-up. However, it is highly time-consuming for radiologists to pixel-by-pixel segment ILD patterns from CT scans with hundreds of slices. Consequently, it is hard to obtain large amounts of well-annotated data, which poses a huge challenge for data-driven deep learning-based methods. To alleviate this problem, we propose an end-to-end semi-supervised learning framework for the segmentation of ILD patterns (ESSegILD) from CT images via self-training with selective re-training. The proposed ESSegILD model is trained using a large CT dataset with slice-wise sparse annotations, i.e., only labeling a few slices in each CT volume with ILD patterns. Specifically, we adopt a popular semi-supervised framework, i.e., Mean-Teacher, that consists of a teacher model and a student model and uses consistency regularization to encourage consistent outputs from the two models under different perturbations. Furthermore, we propose introducing the latest self-training technique with a selective re-training strategy to select reliable pseudo-labels generated by the teacher model, which are used to expand training samples to promote the student model during iterative training. By leveraging consistency regularization and self-training with selective re-training, our proposed ESSegILD can effectively utilize unlabeled data from a partially annotated dataset to progressively improve the segmentation performance. Experiments are conducted on a dataset of 67 pneumonia patients with incomplete annotations containing over 11,000 CT images with eight different lung patterns of ILDs, with the results indicating that our proposed method is superior to the state-of-the-art methods.

Intra-class consistency and inter-class discrimination feature learning for automatic skin lesion classification

Automated skin lesion classification has been proved to be capable of improving the diagnostic performance for dermoscopic images. Although many successes have been achieved, accurate classification remains challenging due to the significant intra-class variation and inter-class similarity. In this article, a deep learning method is proposed to increase the intra-class consistency as well as the inter-class discrimination of learned features in the automatic skin lesion classification. To enhance the inter-class discriminative feature learning, a CAM-based (class activation mapping) global-lesion localization module is proposed by optimizing the distance of CAMs for the same dermoscopic image generated by different skin lesion tasks. Then, a global features guided intra-class similarity learning module is proposed to generate the class center according to the deep features of all samples in one class and the history feature of one sample during the learning process. In this way, the performance can be improved with the collaboration of CAM-based inter-class feature discriminating and global features guided intra-class feature concentrating. To evaluate the effectiveness of the proposed method, extensive experiments are conducted on the ISIC-2017 and ISIC-2018 datasets. Experimental results with different backbones have demonstrated that the proposed method has good generalizability and can adaptively focus on more discriminative regions of the skin lesion.

PPsNet: An improved deep learning model for microsatellite instability high prediction in colorectal cancer from whole slide images

BACKGROUND AND OBJECTIVE

Recent studies have shown that colorectal cancer (CRC) patients with microsatellite instability high (MSI-H) are more likely to benefit from immunotherapy. However, current MSI testing methods are not available for all patients due to the lack of available equipment and trained personnel, as well as the high cost of the assay. Here, we developed an improved deep learning model to predict MSI-H in CRC from whole slide images (WSIs).

METHODS

We established the MSI-H prediction model based on two stages: tumor detection and MSI classification. Previous works applied fine-tuning strategy directly for tumor detection, but ignoring the challenge of vanishing gradient due to the large number of convolutional layers. We added auxiliary classifiers to intermediate layers of pre-trained models to help propagate gradients back through in an effective manner. To predict MSI status, we constructed a pair-wise learning model with a synergic network, named parameter partial sharing network (PPsNet), where partial parameters are shared among two deep convolutional neural networks (DCNNs). The proposed PPsNet contained fewer parameters and reduced the problem of intra-class variation and inter-class similarity. We validated the proposed model on a holdout test set and two external test sets.

RESULTS

144 H&E-stained WSIs from 144 CRC patients (81 cases with MSI-H and 63 cases with MSI-L/MSS) were collected retrospectively from three hospitals. The experimental results indicate that deep supervision based fine-tuning almost outperforms training from scratch and utilizing fine-tuning directly. The proposed PPsNet always achieves better accuracy and area under the receiver operating characteristic curve (AUC) than other solutions with four different neural network architectures on validation. The proposed method finally achieves obvious improvements than other state-of-the-art methods on the validation dataset with an accuracy of 87.28% and AUC of 94.29%.

CONCLUSIONS

The proposed method can obviously increase model performance and our model yields better performance than other methods. Additionally, this work also demonstrates the feasibility of MSI-H prediction using digital pathology images based on deep learning in the Asian population. It is hoped that this model could serve as an auxiliary tool to identify CRC patients with MSI-H more time-saving and efficiently.

A Novel Prediction Model for Brain Glioma Image Segmentation Based on the Theory of Bose-Einstein Condensate

Background

The input image of a blurry glioma image segmentation is, usually, very unclear. It is difficult to obtain the accurate contour line of image segmentation. The main challenge facing the researchers is to correctly determine the area where the points on the contour line belong to the glioma image. This article highlights the mechanism of formation of glioma and provides an image segmentation prediction model to assist in the accurate division of glioma contour points. The proposed prediction model of segmentation associated with the process of the formation of glioma is innovative and challenging. Bose-Einstein Condensate (BEC) is a microscopic quantum phenomenon in which atoms condense to the ground state of energy as the temperature approaches absolute zero. In this article, we propose a BEC kernel function and a novel prediction model based on the BEC kernel to detect the relationship between the process of the BEC and the formation of a brain glioma. Furthermore, the theoretical derivation and proof of the prediction model are given from micro to macro through quantum mechanics, wave, oscillation of glioma, and statistical distribution of laws. The prediction model is a distinct segmentation model that is guided by BEC theory for blurry glioma image segmentation.

Results

Our approach is based on five tests. The first three tests aimed at confirming the measuring range of T and μ in the BEC kernel. The results are extended from −10 to 10, approximating the standard range to T ≤ 0, and μ from 0 to 6.7. Tests 4 and 5 are comparison tests. The comparison in Test 4 was based on various established cluster methods. The results show that our prediction model in image evaluation parameters of P, R, and F is the best amongst all the existent ten forms except for only one reference with the mean value of F that is between 0.88 and 0.93, while our approach returns between 0.85 and 0.99. Test 5 aimed to further compare our results, especially with CNN (Convolutional Neural Networks) methods, by challenging Brain Tumor Segmentation (BraTS) and clinic patient datasets. Our results were also better than all reference tests. In addition, the proposed prediction model with the BEC kernel is feasible and has a comparative validity in glioma image segmentation.

Conclusions

Theoretical derivation and experimental verification show that the prediction model based on the BEC kernel can solve the problem of accurate segmentation of blurry glioma images. It demonstrates that the BEC kernel is a more feasible, valid, and accurate approach than a lot of the recent year segmentation methods. It is also an advanced and innovative model of prediction deducing from micro BEC theory to macro glioma image segmentation.

Deep metric attention learning for skin lesion classification in dermoscopy images

Currently, convolutional neural networks (CNNs) have made remarkable achievements in skin lesion classification because of their end-to-end feature representation abilities. However, precise skin lesion classification is still challenging because of the following three issues: (1) insufficient training samples, (2) inter-class similarities and intra-class variations, and (3) lack of the ability to focus on discriminative skin lesion parts. To address these issues, we propose a deep metric attention learning CNN (DeMAL-CNN) for skin lesion classification. In DeMAL-CNN, a triplet-based network (TPN) is first designed based on deep metric learning, which consists of three weight-shared embedding extraction networks. TPN adopts a triplet of samples as input and uses the triplet loss to optimize the embeddings, which can not only increase the number of training samples, but also learn the embeddings robust to inter-class similarities and intra-class variations. In addition, a mixed attention mechanism considering both the spatial-wise and channel-wise attention information is designed and integrated into the construction of each embedding extraction network, which can further strengthen the skin lesion localization ability of DeMAL-CNN. After extracting the embeddings, three weight-shared classification layers are used to generate the final predictions. In the training procedure, we combine the triplet loss with the classification loss as a hybrid loss to train DeMAL-CNN. We compare DeMAL-CNN with the baseline method, attention methods, advanced challenge methods, and state-of-the-art skin lesion classification methods on the ISIC 2016 and ISIC 2017 datasets, and test its generalization ability on the PH2 dataset. The results demonstrate its effectiveness.

Special Convolutional Neural Network for Identification and Positioning of Interstitial Lung Disease Patterns in Computed Tomography Images

In this paper, automated detection of interstitial lung disease patterns in high resolution computed tomography images is achieved by developing a faster region-based convolutional network based detector with GoogLeNet as a backbone. GoogLeNet is simplified by removing few inception models and used as the backbone of the detector network. The proposed framework is developed to detect several interstitial lung disease patterns without doing lung field segmentation. The proposed method is able to detect the five most prevalent interstitial lung disease patterns: fibrosis, emphysema, consolidation, micronodules and ground-glass opacity, as well as normal. Five-fold cross-validation has been used to avoid bias and reduce over-fitting. The proposed framework performance is measured in terms of F-score on the publicly available MedGIFT database. It outperforms state-of-the-art techniques. The detection is performed at slice level and could be used for screening and differential diagnosis of interstitial lung disease patterns using high resolution computed tomography images.

A Dataset of Pulmonary Lesions With Multiple-Level Attributes and Fine Contours

Lung cancer is a life-threatening disease and its diagnosis is of great significance. Data scarcity and unavailability of datasets is a major bottleneck in lung cancer research. In this paper, we introduce a dataset of pulmonary lesions for designing the computer-aided diagnosis (CAD) systems. The dataset has fine contour annotations and nine attribute annotations. We define the structure of the dataset in detail, and then discuss the relationship of the attributes and pathology, and the correlation between the nine attributes with the chi-square test. To demonstrate the contribution of our dataset to computer-aided system design, we define four tasks that can be developed using our dataset. Then, we use our dataset to model multi-attribute classification tasks. We discuss the performance in 2D, 2.5D, and 3D input modes of the classification model. To improve performance, we introduce two attention mechanisms and verify the principles of the attention mechanisms through visualization. Experimental results show the relationship between different models and different levels of attributes.

Improved bag-of-features using grey relational analysis for classification of histology images

An efficient classification method to categorize histopathological images is a challenging research problem. In this paper, an improved bag-of-features approach is presented as an efficient image classification method. In bag-of-features, a large number of keypoints are extracted from histopathological images that increases the computational cost of the codebook construction step. Therefore, to select the a relevant subset of keypoints, a new keypoints selection method is introduced in the bag-of-features method. To validate the performance of the proposed method, an extensive experimental analysis is conducted on two standard histopathological image datasets, namely ADL and Blue histology datasets. The proposed keypoint selection method reduces the extracted high dimensional features by 95% and 68% from the ADL and Blue histology datasets respectively with less computational time. Moreover, the enhanced bag-of-features method increases classification accuracy by from other considered classification methods.

WR-SVM Model Based on the Margin Radius Approach for Solving the Minimum Enclosing Ball Problem in Support Vector Machine Classification

The generalization error of conventional support vector machine (SVM) depends on the ratio of two factors; radius and margin. The traditional SVM aims to maximize margin but ignore minimization of radius, which decreases the overall performance of the SVM classifier. However, different approaches are developed to achieve a trade-off between the margin and radius. Still, the computational cost of all these approaches is high due to the requirements of matrix transformation. Furthermore, a conventional SVM tries to set the best hyperplane between classes, and due to some robust kernel tricks, an SVM is used in many non-linear and complex problems. The configuration of the best hyperplane between classes is not effective; therefore, it is required to bind a class within its limited area to enhance the performance of the SVM classifier. The area enclosed by a class is called its Minimum Enclosing Ball (MEB), and it is one of the emerging problems of SVM. Therefore, a robust solution is needed to improve the performance of the conventional SVM to overcome the highlighted issues. In this research study, a novel weighted radius SVM (WR-SVM) is proposed to determine the tighter bounds of MEB. The proposed solution uses a weighted mean to find tighter bounds of radius, due to which the size of MEB decreases. Experiments are conducted on nine different benchmark datasets and one synthetic dataset to demonstrate the effectiveness of our proposed model. The experimental results reveal that the proposed WR-SVM significantly performed well compared to the conventional SVM classifier. Furthermore, experimental results are compared with F-SVM and traditional SVM in terms of classification accuracy to demonstrate the significance of the proposed WR-SVM.

Skin Lesion Classification Using Densely Connected Convolutional Networks with Attention Residual Learning

Skin lesion classification is an effective approach aided by computer vision for the diagnosis of skin cancer. Though deep learning models presented advantages over traditional methods and brought tremendous breakthroughs, a precise diagnosis is still challenging because of the intra-class variation and inter-class similarity caused by the diversity of imaging methods and clinicopathology. In this paper, we propose a densely connected convolutional network with an attention and residual learning (ARDT-DenseNet) method for skin lesion classification. Each ARDT block consists of dense blocks, transition blocks and attention and residual modules. Compared to a residual network with the same number of convolutional layers, the size of the parameters of the densely connected network proposed in this paper has been reduced by half, while the accuracy of skin lesion classification is preserved. Our improved densely connected network adds an attention mechanism and residual learning after each dense block and transition block without introducing additional parameters. We evaluate the ARDT-DenseNet model with the ISIC 2016 and ISIC 2017 datasets. Our method achieves an ACC of 85.7% and an AUC of 83.7% in skin lesion classification with ISIC 2016 and an average AUC of 91.8% in skin lesion classification with ISIC 2017. The experimental results show that the method proposed in this paper has achieved a significant improvement in skin lesion classification, which is superior to that of the state-of-the-art method.

Handling imbalanced medical image data: A deep-learning-based one-class classification approach

In clinical settings, a lot of medical image datasets suffer from the imbalance problem which hampers the detection of outliers (rare health care events), as most classification methods assume an equal occurrence of classes. In this way, identifying outliers in imbalanced datasets has become a crucial issue. To help address this challenge, one-class classification, which focuses on learning a model using samples from only a single given class, has attracted increasing attention. Previous one-class modeling usually uses feature mapping or feature fitting to enforce the feature learning process. However, these methods are limited for medical images which usually have complex features. In this paper, a novel method is proposed to enable deep learning models to optimally learn single-class-relevant inherent imaging features by leveraging the concept of imaging complexity. We investigate and compare the effects of simple but effective perturbing operations applied to images to capture imaging complexity and to enhance feature learning. Extensive experiments are performed on four clinical datasets to show that the proposed method outperforms four state-of-the-art methods.

Attention Residual Learning for Skin Lesion Classification

Automated skin lesion classification in dermoscopy images is an essential way to improve the diagnostic performance and reduce melanoma deaths. Although deep convolutional neural networks (DCNNs) have made dramatic breakthroughs in many image classification tasks, accurate classification of skin lesions remains challenging due to the insufficiency of training data, inter-class similarity, intra-class variation, and the lack of the ability to focus on semantically meaningful lesion parts. To address these issues, we propose an attention residual learning convolutional neural network (ARL-CNN) model for skin lesion classification in dermoscopy images, which is composed of multiple ARL blocks, a global average pooling layer, and a classification layer. Each ARL block jointly uses the residual learning and a novel attention learning mechanisms to improve its ability for discriminative representation. Instead of using extra learnable layers, the proposed attention learning mechanism aims to exploit the intrinsic self-attention ability of DCNNs, i.e., using the feature maps learned by a high layer to generate the attention map for a low layer. We evaluated our ARL-CNN model on the ISIC-skin 2017 dataset. Our results indicate that the proposed ARL-CNN model can adaptively focus on the discriminative parts of skin lesions, and thus achieve the state-of-the-art performance in skin lesion classification.

A novel fused convolutional neural network for biomedical image classification

With the advent of biomedical imaging technology, the number of captured and stored biomedical images is rapidly increasing day by day in hospitals, imaging laboratories and biomedical institutions. Therefore, more robust biomedical image analysis technology is needed to meet the requirement of the diagnosis and classification of various kinds of diseases using biomedical images. However, the current biomedical image classification methods and general non-biomedical image classifiers cannot extract more compact biomedical image features or capture the tiny differences between similar images with different types of diseases from the same category. In this paper, we propose a novel fused convolutional neural network to develop a more accurate and highly efficient classifier for biomedical images, which combines shallow layer features and deep layer features from the proposed deep neural network architecture. In the analysis, it was observed that the shallow layers provided more detailed local features, which could distinguish different diseases in the same category, while the deep layers could convey more high-level semantic information used to classify the diseases among the various categories. A detailed comparison of our approach with traditional classification algorithms and popular deep classifiers across several public biomedical image datasets showed the superior performance of our proposed method for biomedical image classification. In addition, we also evaluated the performance of our method in modality classification of medical images using the ImageCLEFmed dataset. Graphical abstract The graphical abstract shows the fused, deep convolutional neural network architecture proposed for biomedical image classification. In the architecture, we can clearly see the feature-fusing process going from shallow layers and the deep layers.

Multiscale Rotation-Invariant Convolutional Neural Networks for Lung Texture Classification

We propose a new multiscale rotation-invariant convolutional neural network (MRCNN) model for classifying various lung tissue types on high-resolution computed tomography. MRCNN employs Gabor-local binary pattern that introduces a good property in image analysis-invariance to image scales and rotations. In addition, we offer an approach to deal with the problems caused by imbalanced number of samples between different classes in most of the existing works, accomplished by changing the overlapping size between the adjacent patches. Experimental results on a public interstitial lung disease database show a superior performance of the proposed method to state of the art.

Dual discriminative local coding for tissue aging analysis

In aging research, morphological age of tissue helps to characterize the effects of aging on different individuals. While currently manual evaluations are used to estimate morphological ages under microscopy, such operation is difficult and subjective due to the complex visual characteristics of tissue images. In this paper, we propose an automated method to quantify morphological ages of tissues from microscopy images. We design a new sparse representation method, namely dual discriminative local coding (DDLC), that classifies the tissue images into different chronological ages. DDLC in- corporates discriminative distance learning and dual-level local coding into the basis model of locality-constrained linear coding thus achieves higher discriminative capability. The morphological age is then computed based on the classification scores. We conducted our study using the publicly avail- able terminal bulb aging database that has been commonly used in existing microscopy imaging research. To represent these images, we also design a highly descriptive descriptor that combines several complementary texture features extracted at two scales. Experimental results show that our method achieves significant improvement in age classification when compared to the existing approaches and other popular classifiers. We also present promising results in quantification of morphological ages.

A novel end-to-end classifier using domain transferred deep convolutional neural networks for biomedical images

BACKGROUND AND OBJECTIVES

Highly accurate classification of biomedical images is an essential task in the clinical diagnosis of numerous medical diseases identified from those images. Traditional image classification methods combined with hand-crafted image feature descriptors and various classifiers are not able to effectively improve the accuracy rate and meet the high requirements of classification of biomedical images. The same also holds true for artificial neural network models directly trained with limited biomedical images used as training data or directly used as a black box to extract the deep features based on another distant dataset. In this study, we propose a highly reliable and accurate end-to-end classifier for all kinds of biomedical images via deep learning and transfer learning.

METHODS

We first apply domain transferred deep convolutional neural network for building a deep model; and then develop an overall deep learning architecture based on the raw pixels of original biomedical images using supervised training. In our model, we do not need the manual design of the feature space, seek an effective feature vector classifier or segment specific detection object and image patches, which are the main technological difficulties in the adoption of traditional image classification methods. Moreover, we do not need to be concerned with whether there are large training sets of annotated biomedical images, affordable parallel computing resources featuring GPUs or long times to wait for training a perfect deep model, which are the main problems to train deep neural networks for biomedical image classification as observed in recent works.

RESULTS

With the utilization of a simple data augmentation method and fast convergence speed, our algorithm can achieve the best accuracy rate and outstanding classification ability for biomedical images. We have evaluated our classifier on several well-known public biomedical datasets and compared it with several state-of-the-art approaches.

CONCLUSIONS

We propose a robust automated end-to-end classifier for biomedical images based on a domain transferred deep convolutional neural network model that shows a highly reliable and accurate performance which has been confirmed on several public biomedical image datasets.

Holistic classification of CT attenuation patterns for interstitial lung diseases via deep convolutional neural networks

Interstitial lung diseases (ILD) involve several abnormal imaging patterns observed in computed tomography (CT) images. Accurate classification of these patterns plays a significant role in precise clinical decision making of the extent and nature of the diseases. Therefore, it is important for developing automated pulmonary computer-aided detection systems. Conventionally, this task relies on experts' manual identification of regions of interest (ROIs) as a prerequisite to diagnose potential diseases. This protocol is time consuming and inhibits fully automatic assessment. In this paper, we present a new method to classify ILD imaging patterns on CT images. The main difference is that the proposed algorithm uses the entire image as a holistic input. By circumventing the prerequisite of manual input ROIs, our problem set-up is significantly more difficult than previous work but can better address the clinical workflow. Qualitative and quantitative results using a publicly available ILD database demonstrate state-of-the-art classification accuracy under the patch-based classification and shows the potential of predicting the ILD type using holistic image.

A swarm-trained k-nearest prototypes adaptive classifier with automatic feature selection for interval data

Some complex data types are capable of modeling data variability and imprecision. These data types are studied in the symbolic data analysis field. One such data type is interval data, which represents ranges of values and is more versatile than classic point data for many domains. This paper proposes a new prototype-based classifier for interval data, trained by a swarm optimization method. Our work has two main contributions: a swarm method which is capable of performing both automatic selection of features and pruning of unused prototypes and a generalized weighted squared Euclidean distance for interval data. By discarding unnecessary features and prototypes, the proposed algorithm deals with typical limitations of prototype-based methods, such as the problem of prototype initialization. The proposed distance is useful for learning classes in interval datasets with different shapes, sizes and structures. When compared to other prototype-based methods, the proposed method achieves lower error rates in both synthetic and real interval datasets.

Combining Generative and Discriminative Representation Learning for Lung CT Analysis With Convolutional Restricted Boltzmann Machines

The choice of features greatly influences the performance of a tissue classification system. Despite this, many systems are built with standard, predefined filter banks that are not optimized for that particular application. Representation learning methods such as restricted Boltzmann machines may outperform these standard filter banks because they learn a feature description directly from the training data. Like many other representation learning methods, restricted Boltzmann machines are unsupervised and are trained with a generative learning objective; this allows them to learn representations from unlabeled data, but does not necessarily produce features that are optimal for classification. In this paper we propose the convolutional classification restricted Boltzmann machine, which combines a generative and a discriminative learning objective. This allows it to learn filters that are good both for describing the training data and for classification. We present experiments with feature learning for lung texture classification and airway detection in CT images. In both applications, a combination of learning objectives outperformed purely discriminative or generative learning, increasing, for instance, the lung tissue classification accuracy by 1 to 8 percentage points. This shows that discriminative learning can help an otherwise unsupervised feature learner to learn filters that are optimized for classification.

Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning

Remarkable progress has been made in image recognition, primarily due to the availability of large-scale annotated datasets and deep convolutional neural networks (CNNs). CNNs enable learning data-driven, highly representative, hierarchical image features from sufficient training data. However, obtaining datasets as comprehensively annotated as ImageNet in the medical imaging domain remains a challenge. There are currently three major techniques that successfully employ CNNs to medical image classification: training the CNN from scratch, using off-the-shelf pre-trained CNN features, and conducting unsupervised CNN pre-training with supervised fine-tuning. Another effective method is transfer learning, i.e., fine-tuning CNN models pre-trained from natural image dataset to medical image tasks. In this paper, we exploit three important, but previously understudied factors of employing deep convolutional neural networks to computer-aided detection problems. We first explore and evaluate different CNN architectures. The studied models contain 5 thousand to 160 million parameters, and vary in numbers of layers. We then evaluate the influence of dataset scale and spatial image context on performance. Finally, we examine when and why transfer learning from pre-trained ImageNet (via fine-tuning) can be useful. We study two specific computer-aided detection (CADe) problems, namely thoraco-abdominal lymph node (LN) detection and interstitial lung disease (ILD) classification. We achieve the state-of-the-art performance on the mediastinal LN detection, and report the first five-fold cross-validation classification results on predicting axial CT slices with ILD categories. Our extensive empirical evaluation, CNN model analysis and valuable insights can be extended to the design of high performance CAD systems for other medical imaging tasks.

Histopathological Image Classification Using Discriminative Feature-Oriented Dictionary Learning

In histopathological image analysis, feature extraction for classification is a challenging task due to the diversity of histology features suitable for each problem as well as presence of rich geometrical structures. In this paper, we propose an automatic feature discovery framework via learning class-specific dictionaries and present a low-complexity method for classification and disease grading in histopathology. Essentially, our Discriminative Feature-oriented Dictionary Learning (DFDL) method learns class-specific dictionaries such that under a sparsity constraint, the learned dictionaries allow representing a new image sample parsimoniously via the dictionary corresponding to the class identity of the sample. At the same time, the dictionary is designed to be poorly capable of representing samples from other classes. Experiments on three challenging real-world image databases: 1) histopathological images of intraductal breast lesions, 2) mammalian kidney, lung and spleen images provided by the Animal Diagnostics Lab (ADL) at Pennsylvania State University, and 3) brain tumor images from The Cancer Genome Atlas (TCGA) database, reveal the merits of our proposal over state-of-the-art alternatives. Moreover, we demonstrate that DFDL exhibits a more graceful decay in classification accuracy against the number of training images which is highly desirable in practice where generous training is often not available.

Capturing heterogeneous group differences using mixture-of-experts: Application to a study of aging

In MRI studies, linear multi-variate methods are often employed to identify regions or connections that are affected due to disease or normal aging. Such linear models inherently assume that there is a single, homogeneous abnormality pattern that is present in all affected individuals. While kernel-based methods can implicitly model a non-linear effect, and therefore the heterogeneity in the affected group, extracting and interpreting information about affected regions is difficult. In this paper, we present a method that explicitly models and captures heterogeneous patterns of change in the affected group relative to a reference group of controls. For this purpose, we use the Mixture-of-Experts (MOE) framework, which combines unsupervised modeling of mixtures of distributions with supervised learning of classifiers. MOE approximates the non-linear boundary between the two groups with a piece-wise linear boundary, thus allowing discovery of multiple patterns of group differences. In the case of patient/control comparisons, each such pattern aims to capture a different dimension of a disease, and hence to identify patient subgroups. We validated our model using multiple simulation scenarios and performance measures. We applied this method to resting state functional MRI data from the Baltimore Longitudinal Study of Aging, to investigate heterogeneous effects of aging on brain function in cognitively normal older adults (>85years) relative to a reference group of normal young to middle-aged adults (<60years). We found strong evidence for the presence of two subgroups of older adults, with similar age distributions in each subgroup, but different connectivity patterns associated with aging. While both older subgroups showed reduced functional connectivity in the Default Mode Network (DMN), increases in functional connectivity within the pre-frontal cortex as well as the bilateral insula were observed only for one of the two subgroups. Interestingly, the subgroup showing this increased connectivity (unlike the other subgroup) was, cognitively similar at baseline to the young and middle-aged subjects in two of seven cognitive domains, and had a faster rate of cognitive decline in one of seven domains. These results suggest that older individuals whose baseline cognitive performance is comparable to that of younger individuals recruit their "cognitive reserve" later in life, to compensate for reduced connectivity in other brain regions.