Search

Accurate histological subtype classification between adenocarcinoma (ADC) and squamous cell carcinoma (SCC) using computed tomography (CT) images is of great importance to assist clinicians in determining treatment and therapy plans for non-small cell lung cancer (NSCLC) patients. Although current deep learning approaches have achieved promising progress in this field, they are often difficult to capture efficient tumor representations due to inadequate training data, and in consequence show limited performance. In this study, we propose a novel and effective reconstruction-assisted feature encoding network (RAFENet) for histological subtype classification by leveraging an auxiliary image reconstruction task to enable extra guidance and regularization for enhanced tumor feature representations. Different from existing reconstruction-assisted methods that directly use generalizable features obtained from shared encoder for primary task, a dedicated task-aware encoding module is utilized in RAFENet to perform refinement of generalizable features. Specifically, a cascade of cross-level non-local blocks are introduced to progressively refine generalizable features at different levels with the aid of lower-level task-specific information, which can successfully learn multi-level task-specific features tailored to histological subtype classification. Moreover, in addition to widely adopted pixel-wise reconstruction loss, we introduce a powerful semantic consistency loss function to explicitly supervise the training of RAFENet, which combines both feature consistency loss and prediction consistency loss to ensure semantic invariance during image reconstruction. Extensive experimental results show that RAFENet effectively addresses the difficult issues that cannot be resolved by existing reconstruction-based methods and consistently outperforms other state-of-the-art methods on both public and in-house NSCLC datasets. Supplementary material is available at https://github.com/lhch1994/Rafenet_sup_material. [ABSTRACT FROM AUTHOR]

In this paper, we study the employment of multiple Unmanned Aerial Vehicles (UAVs) to monitor Points of Interests (PoIs) in a disaster area, e.g., collapsed buildings after an earthquake, where the UAVs can take photos and videos for the people trapped at PoIs, because such valuable information is imperative to make rescue decisions. Unlike most existing studies that ignored the monitoring time of PoIs and simply minimized the longest flying distance among the UAVs, we observe that it takes time to monitor the PoIs. Then, it is possible that the flying distance of a UAV in its flying tour may not be too long, the tour however contains many densely-located PoIs. Therefore, it will take a very long time for the UAV to monitor the PoIs in its tour. In this paper, we first formulate a problem of finding flying tours for $K$ K given UAVs to collaboratively monitor PoIs in a disaster area, such that the maximum spent time of the $K$ K UAVs among their tours is minimized, where the spent time of a UAV in its tour consists of the flying time and the PoI monitoring time. We then propose a novel $5\frac{1}{3}$ 5 1 3 -approximation algorithm for the problem, improving the best approximation ratio 6 so far for the problem of minimizing the longest flying distance among the UAVs. In addition, we extend the proposed algorithm to the case that each UAV may not be able to monitor all PoIs assigned to it, due to its limited maximum flying time (e.g., 30 minutes), and the UAV must return to its depot to replace its battery. We finally evaluate the performance of the proposed algorithms via simulation environments, and experimental results show that the proposed algorithms are very promising. Especially, the maximum spent times of the $K$ K UAVs in their tours by the proposed algorithms are up to 30 percent shorter than those by existing algorithms. In addition, the empirical approximation ratios of the proposed algorithms are no more than 2.4, which are much smaller than their theoretical approximation ratios that are at least $5\frac{1}{3}$ 5 1 3 . [ABSTRACT FROM AUTHOR]

Recently, digital pathology image-based prognosis prediction has become a hot topic in healthcare research to make early decisions on therapy and improve the treatment quality of patients. Therefore, there has been a recent surge of interest in designing deep learning method solving the problem of prognosis prediction with digital pathology images. However, whole slide histopathological images (WSIs) based prognosis prediction is still a challenge due to the large size of pathological images, the heterogeneity of tumors and the high cost of region of interests (ROIs) labeling. In this study, we design a novel two-stage deep learning framework for prognosis prediction (TSDLPP) based on WSIs. Our proposed framework consists of two-stage paradigms: 1) training tissue decomposition network (TDNet) to divide WSIs into cancerous and non-cancerous regions, 2) integrating general prognosis-related densely connected CNN (GPR-DCCNN) and morphology-specific prognosis-related densely connected CNNs (MSPR-DCCNNs) to extract different level features of pathological images. In the end, we apply TSDLPP to the prognosis prediction of breast cancer using The Cancer Genome Atlas (TCGA) datasets. Experiment results demonstrate that TSDLPP obtains superior performance of prognosis prediction compared with the existing state-of-arts methods. [ABSTRACT FROM AUTHOR]

In the clinical environment, myocardial infarction (MI) as one common cardiovascular disease is mainly evaluated using late gadolinium enhancement (LGE) cardiac magnetic resonance images (CMRIs). Accurate segmentation of the ventricles and the myocardium is a prerequisite for quantitative assessment of cardiac functions and disease progression. Performing the task using LGE images is, however, rather challenging due to heterogeneous image intensity distribution and lack of clear boundaries between adjacent organs and tissues. In this paper we propose a deep neural network method for automatic segmentation of the left ventricle (LV), right ventricle (RV), and left ventricular myocardium (LVM) from LGE CMRIs, which also leverages complementary information from cine and T2-weighted CMRIs if available. In the proposed method, termed SK-Unet, we augment the original U-Net model by adding a squeeze-and-excitation residual (SE-Res) module in the encoder and a selective kernel (SK) module in the decoder. The SE-Res module applies an attention mechanism to enhance informative feature extraction and suppress redundant ones. The SK module offers the ability to adaptively learn task-relevant multi-scale spatial features. We tested our method by participating in the MICCAI 2019 MS-CMRSeg challenge and achieved a mean dice score of 0.922 for LV segmentation, 0.827 for LVM, and 0.874 for RV. The results placed our method at the 1st place in the competition, and our accuracy of 0.827 also greatly surpasses the measured inter-observer agreement of 0.757 for manual segmentation of LVM in LGE CMRIs. The code accompanying our method is made available online at https://github.com/Xiyue-Wang/1st-in-MS-CMRSeg-2019. [ABSTRACT FROM AUTHOR]

In recent years, deep learning methods have achieved great success for vehicle detection tasks in aerial imagery. However, most existing methods focus only on extracting latent vehicle target features, and rarely consider the scene context as vital prior knowledge. In this letter, we propose a scene context attention-based fusion network (SCAF-Net), to fuse the scene context of vehicles into an end-to-end vehicle detection network. First, we propose a novel strategy, patch cover, to keep the original target and scene context information in raw aerial images of a large scale as much as possible. Next, we use an improved YOLO-v3 network as one branch of SCAF-Net, to generate vehicle candidates on each patch. Here, a novel branch for the scene context is utilized to extract the latent scene context of vehicles on each patch without any extra annotations. Then, these two branches above are concatenated together as a fusion network, and we apply an attention-based model to further extract vehicle candidates of each local scene. Finally, all vehicle candidates of different patches, are merged by global nonmax suppress (g-NMS) to output the detection result of the whole original image. Experimental results demonstrate that our proposed method outperforms the comparison methods with both high detection accuracy and speed. Our code is released at https://github.com/minghuicode/SCAF-Net. [ABSTRACT FROM AUTHOR]

A common strategy to discovering cancer associated copy number aberrations (CNAs) from a cohort of cancer samples is to detect recurrent CNAs (RCNAs). Although the previous methods can successfully identify communal RCNAs shared by nearly all tumor samples, detecting subgroup-specific RCNAs and their related subgroup samples from cancer samples with heterogeneity is still invalid for these existing approaches. In this paper, we introduce a novel integrated method called HetRCNA, which can identify statistically significant subgroup-specific RCNAs and their related subgroup samples. Based on matrix decomposition framework with weight constraint, HetRCNA can successfully measure the subgroup samples by coefficients of left vectors with weight constraint and subgroup-specific RCNAs by coefficients of the right vectors and significance test. When we evaluate HetRCNA on simulated dataset, the results show that HetRCNA gives the best performances among the competing methods and is robust to the noise factors of the simulated data. When HetRCNA is applied on a real breast cancer dataset, our approach successfully identifies a bunch of RCNA regions and the result is highly correlated with the results of the other two investigated approaches. Notably, the genomic regions identified by HetRCNA harbor many breast cancer related genes reported by previous researches. [ABSTRACT FROM AUTHOR]

An effective way to facilitate the development of modern oncology precision medicine is the systematical analysis of the known drug sensitivities that have emerged in recent years. Meanwhile, the screening of drug response in cancer cell lines provides an estimable genomic and pharmacological data towards high accuracy prediction. Existing works primarily utilize genomic or functional genomic features to classify or regress the drug response. Here in this work, by the migration and extension of the conventional merchandise recommendation methods, we introduce an innovation model on accurate drug sensitivity prediction by using dual-layer strengthened collaborative topic regression (DS-CTR), which incorporates not only the graphic model to jointly learn drugs and cell lines feature from pharmacogenomics data but also drug and cell line similarity network model to strengthen the correlation of the prediction results. Using Genomics of Drug Sensitivity in Cancer project (GDSC) as benchmark datasets, the 5-fold cross-validation experiment demonstrates that DS-CTR model significantly improves drug response prediction performance compared with four categories of state-of-the-art algorithms as for both Receiver Operator Curve (ROC) and the Area Under Receiver Operator Curve (AUC). By uncovering the unknown cell-drug associations with advanced literature evidences, our novel model DS-CTR is validated and supported. The model also provides the possibility to make the discovery of new anti-cancer therapeutics in the preclinical trials cheaper and faster. [ABSTRACT FROM AUTHOR]

Recently, research intergrading medicine and Artificial Intelligence has attracted extensive attention. Mobile health has emerged as a promising paradigm for improving people's work and life in the future. However, high mobility of mobile devices and limited resources pose challenges for users to deal with the applications in mobile health that require large amount of computational resources. In this paper, a novel computation offloading mechanism is proposed in the environments combining of the Internet of Vehicles and Multi-Access Edge Computing. Through the proposed mechanism, mobile health applications are divided into several parts and can be offloaded to appropriate nearby vehicles while meeting the requirements of application completion time, energy consumption, and resource utilization. A particle swarm optimization based approach is proposed to optimize the the aforementioned computation offloading problem in a specific medical application. Evaluations of the proposed algorithms against local computing method serves as baseline method are conducted via extensive simulations. The average task completion time saved by our proposed task allocation scheme increases continually compared with the local solution. Specially, the global resource utilization rate increased from 71.8% to 94.5% compared with the local execution time. [ABSTRACT FROM AUTHOR]

Glioblastoma multiforme (GBM) is one of the most malignant brain tumors with very short prognosis expectation. To improve patients’ clinical treatment and their life quality after surgery, researches have developed tremendous in silico models and tools for predicting GBM prognosis based on molecular datasets and have earned great success. However, pathology still plays the most critical role in cancer diagnosis and prognosis in the clinic at present. Recent advancement of storing and processing histopathological images has drawn attention of researchers. Models based on histopathological images are developed, which show great potential for computer-aided pathological diagnoses. But models based on both molecular and histopathological images that could predict GBM prognosis with high accuracy are not present yet. In our previous research, we used the simple MKL method to integrate multi-omics data to improve GBM prognosis prediction successfully. In this paper, we have developed a novel multiple kernel learning (MKL) method, named histopathological integrating multiple kernel learning (HI-MKL), that could integrate both histopathological images and multi-omics data efficiently. By using datasets from The Cancer Genome Atlas project, we have built a system that could predict the GBM prognosis with high accuracy. Our research shows that HI-MKL is an accurate, robust, and generalized MKL method, which performs well in a GBM prognosis task. [ABSTRACT FROM AUTHOR]

Long non-coding RNAs (lncRNA) play crucial roles in a variety of biological processes and complex diseases. Massive studies have indicated that lncRNAs interact with related proteins to exert regulation of cellular biological processes. Because it is time-consuming and expensive to determine lncRNA-protein interaction by experiment, more accurate predictions of interaction by computational methods are imperative. We propose a novel computational approach, predicting lncRNA-protein interaction using graph regularized nonnegative matrix factorization (LPGNMF), to discover unobserved lncRNA-protein association. First, we calculate lncRNA similarity and protein similarity by integrating the lncRNA expression information and gene ontology information. Subsequently, we utilize graph regularized nonnegative matrix factorization framework to predict potential interactions for all lncRNA simultaneously. In the cross validation test, LPGNMF achieves an AUC of 85.2 percent, higher than those of other compared methods. In addition, novel lncRNA-protein interactions detected by LPGNMF are validated by literatures or database. The results indicate that our method is effective to discover potential lncRNA-protein interaction. [ABSTRACT FROM AUTHOR]

Breast cancer is a highly aggressive type of cancer with very low median survival. Accurate prognosis prediction of breast cancer can spare a significant number of patients from receiving unnecessary adjuvant systemic treatment and its related expensive medical costs. Previous work relies mostly on selected gene expression data to create a predictive model. The emergence of deep learning methods and multi-dimensional data offers opportunities for more comprehensive analysis of the molecular characteristics of breast cancer and therefore can improve diagnosis, treatment, and prevention. In this study, we propose a Multimodal Deep Neural Network by integrating Multi-dimensional Data (MDNNMD) for the prognosis prediction of breast cancer. The novelty of the method lies in the design of our method's architecture and the fusion of multi-dimensional data. The comprehensive performance evaluation results show that the proposed method achieves a better performance than the prediction methods with single-dimensional data and other existing approaches. The source code implemented by TensorFlow 1.0 deep learning library can be downloaded from the Github: https://github.com/USTC-HIlab/MDNNMD. [ABSTRACT FROM AUTHOR]