Publications by authors named "David Dagan Feng"

99 Publications

Predicting distant metastases in soft-tissue sarcomas from PET-CT scans using constrained hierarchical multi-modality feature learning.

Phys Med Biol 2021 Nov 24. Epub 2021 Nov 24.

The University of Sydney, Sydney, 2006, AUSTRALIA.

Objective: Positron emission tomography-computed tomography (PET-CT) is regarded as the imaging modality of choice for the management of soft-tissue sarcomas (STSs). Distant metastases (DM) are the leading cause of death in STS patients and early detection is important to effectively manage tumors with surgery, radiotherapy and chemotherapy. In this study, we aim to early detect DM in patients with STS using their PET-CT data.

Approach: We derive a new convolutional neural network (CNN) method for early DM detection. The novelty of our method is the introduction of a constrained hierarchical multi-modality feature learning approach to integrate functional imaging (PET) features with anatomical imaging (CT) features. In addition, we removed the reliance on manual input, e.g., tumor delineation, for extracting imaging features.

Main Results: Our experimental results on a well-established benchmark PET-CT dataset show that our method achieved the highest accuracy (0.896) and AUC (0.903) scores when compared to the state-of-the-art methods (unpaired student's t-test p-value < 0.05).

Significance: Our method could be an effective and supportive tool to aid physicians in tumor quantification and in identifying image biomarkers for cancer treatment.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1088/1361-6560/ac3d17DOI Listing
November 2021

Machine learning-based noninvasive quantification of single-imaging session dual-tracer 18F-FDG and 68Ga-DOTATATE dynamic PET-CT in oncology.

IEEE Trans Med Imaging 2021 Sep 14;PP. Epub 2021 Sep 14.

68Ga-DOTATATE PET-CT is routinely used for imaging neuroendocrine tumor (NET) somatostatin receptor subtype 2 (SSTR2) density in patients, and is complementary to FDG PET-CT for improving the accuracy of NET detection, characterization, grading, staging, and predicting/monitoring NET responses to treatment. Performing sequential 18F-FDG and 68Ga-DOTATATE PET scans would require 2 or more days and can delay patient care. To align temporal and spatial measurements of 18F-FDG and 68Ga-DOTATATE PET, and to reduce scan time and CT radiation exposure to patients, we propose a single-imaging session dual-tracer dynamic PET acquisition protocol in the study. A recurrent extreme gradient boosting (rXGBoost) machine learning algorithm was proposed to separate the mixed 18F-FDG and 68Ga-DOTATATE time activity curves (TACs) for the region of interest (ROI) based quantification with tracer kinetic modeling. A conventional parallel multi-tracer compartment modeling method was also implemented for reference. Single-scan dual-tracer dynamic PET was simulated from 12 NET patient studies with 18F-FDG and 68Ga-DOTATATE 45-min dynamic PET scans separately obtained within 2 days. Our experimental results suggested an 18F-FDG injection first followed by 68Ga-DOTATATE with a minimum 5 min delayed injection protocol for the separation of mixed 18F-FDG and 68Ga-DOTATATE TACs using rXGBoost algorithm followed by tracer kinetic modeling is highly feasible.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2021.3112783DOI Listing
September 2021

Automatic left ventricular cavity segmentation via deep spatial sequential network in 4D computed tomography.

Comput Med Imaging Graph 2021 07 9;91:101952. Epub 2021 Jun 9.

School of Computer Science, University of Sydney, NSW 2006, Australia. Electronic address:

Automated segmentation of left ventricular cavity (LVC) in temporal cardiac image sequences (consisting of multiple time-points) is a fundamental requirement for quantitative analysis of cardiac structural and functional changes. Deep learning methods for segmentation are the state-of-the-art in performance; however, these methods are generally formulated to work on a single time-point, and thus disregard the complementary information available from the temporal image sequences that can aid in segmentation accuracy and consistency across the time-points. In particular, single time-point segmentation methods perform poorly in segmenting the end-systole (ES) phase image in the cardiac sequence, where the left ventricle deforms to the smallest irregular shape, and the boundary between the blood chamber and the myocardium becomes inconspicuous and ambiguous. To overcome these limitations in automatically segmenting temporal LVCs, we present a spatial sequential network (SS-Net) to learn the deformation and motion characteristics of the LVCs in an unsupervised manner; these characteristics are then integrated with sequential context information derived from bi-directional learning (BL) where both chronological and reverse-chronological directions of the image sequence are used. Our experimental results on a cardiac computed tomography (CT) dataset demonstrate that our spatial-sequential network with bi-directional learning (SS-BL-Net) outperforms existing methods for spatiotemporal LVC segmentation.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.compmedimag.2021.101952DOI Listing
July 2021

Deep Cognitive Gate: Resembling Human Cognition for Saliency Detection.

IEEE Trans Pattern Anal Mach Intell 2021 Mar 23;PP. Epub 2021 Mar 23.

Saliency detection by human refers to the ability to identify pertinent information using our perceptive and cognitive capabilities. While human perception is attracted by visual stimuli, our cognitive capability is derived from the inspiration of constructing concepts of reasoning. Saliency detection has gained intensive interest with the aim of resembling human perceptual system. However, saliency related to human cognition, particularly the analysis of complex salient regions (cogitating process), is yet to be fully exploited. We propose to resemble human cognition, coupled with human perception, to improve saliency detection. We recognize saliency in three phases (Seeing - Perceiving - Cogitating), mimicking human's perceptive and cognitive thinking of an image. In our method, Seeing phase is related to human perception, and we formulate the Perceiving and Cogitating phases related to the human cognition systems via deep neural networks (DNNs) to construct a new module (Cognitive Gate) that enhances the DNN features for saliency detection. To the best of our knowledge, this is the first work that established DNNs to resemble human cognition for saliency detection. In our experiments, our approach outperformed 17 benchmarking DNN methods on six well-recognized datasets, demonstrating that resembling human cognition improves saliency detection.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2021.3068277DOI Listing
March 2021

Recurrent feature fusion learning for multi-modality pet-ct tumor segmentation.

Comput Methods Programs Biomed 2021 May 11;203:106043. Epub 2021 Mar 11.

School of Computer Science, University of Sydney, NSW, Australia; Australian Research Council Training Centre for Innovative Bioengineering, NSW, Australia. Electronic address:

Background And Objective: [18f]-fluorodeoxyglucose (fdg) positron emission tomography - computed tomography (pet-ct) is now the preferred imaging modality for staging many cancers. Pet images characterize tumoral glucose metabolism while ct depicts the complementary anatomical localization of the tumor. Automatic tumor segmentation is an important step in image analysis in computer aided diagnosis systems. Recently, fully convolutional networks (fcns), with their ability to leverage annotated datasets and extract image feature representations, have become the state-of-the-art in tumor segmentation. There are limited fcn based methods that support multi-modality images and current methods have primarily focused on the fusion of multi-modality image features at various stages, i.e., early-fusion where the multi-modality image features are fused prior to fcn, late-fusion with the resultant features fused and hyper-fusion where multi-modality image features are fused across multiple image feature scales. Early- and late-fusion methods, however, have inherent, limited freedom to fuse complementary multi-modality image features. The hyper-fusion methods learn different image features across different image feature scales that can result in inaccurate segmentations, in particular, in situations where the tumors have heterogeneous textures.

Methods: we propose a recurrent fusion network (rfn), which consists of multiple recurrent fusion phases to progressively fuse the complementary multi-modality image features with intermediary segmentation results derived at individual recurrent fusion phases: (1) the recurrent fusion phases iteratively learn the image features and then refine the subsequent segmentation results; and, (2) the intermediary segmentation results allows our method to focus on learning the multi-modality image features around these intermediary segmentation results, which minimize the risk of inconsistent feature learning.

Results: we evaluated our method on two pathologically proven non-small cell lung cancer pet-ct datasets. We compared our method to the commonly used fusion methods (early-fusion, late-fusion and hyper-fusion) and the state-of-the-art pet-ct tumor segmentation methods on various network backbones (resnet, densenet and 3d-unet). Our results show that the rfn provides more accurate segmentation compared to the existing methods and is generalizable to different datasets.

Conclusions: we show that learning through multiple recurrent fusion phases allows the iterative re-use of multi-modality image features that refines tumor segmentation results. We also identify that our rfn produces consistent segmentation results across different network architectures.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.cmpb.2021.106043DOI Listing
May 2021

Short-Term Lesion Change Detection for Melanoma Screening With Novel Siamese Neural Network.

IEEE Trans Med Imaging 2021 03 2;40(3):840-851. Epub 2021 Mar 2.

Short-term monitoring of lesion changes has been a widely accepted clinical guideline for melanoma screening. When there is a significant change of a melanocytic lesion at three months, the lesion will be excised to exclude melanoma. However, the decision on change or no-change heavily depends on the experience and bias of individual clinicians, which is subjective. For the first time, a novel deep learning based method is developed in this paper for automatically detecting short-term lesion changes in melanoma screening. The lesion change detection is formulated as a task measuring the similarity between two dermoscopy images taken for a lesion in a short time-frame, and a novel Siamese structure based deep network is proposed to produce the decision: changed (i.e. not similar) or unchanged (i.e. similar enough). Under the Siamese framework, a novel structure, namely Tensorial Regression Process, is proposed to extract the global features of lesion images, in addition to deep convolutional features. In order to mimic the decision-making process of clinicians who often focus more on regions with specific patterns when comparing a pair of lesion images, a segmentation loss (SegLoss) is further devised and incorporated into the proposed network as a regularization term. To evaluate the proposed method, an in-house dataset with 1,000 pairs of lesion images taken in a short time-frame at a clinical melanoma centre was established. Experimental results on this first-of-a-kind large dataset indicate that the proposed model is promising in detecting the short-term lesion change for objective melanoma screening.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2020.3037761DOI Listing
March 2021

SPST-CNN: Spatial pyramid based searching and tagging of liver's intraoperative live views via CNN for minimal invasive surgery.

J Biomed Inform 2020 06 1;106:103430. Epub 2020 May 1.

School of Information Technologies, The University of Sydney, Australia.

Laparoscopic liver surgery is challenging to perform because of compromised ability of the surgeon to localize subsurface anatomy due to minimal invasive visibility. While image guidance has the potential to address this barrier, intraoperative factors, such as insufflations and variable degrees of organ mobilization from supporting ligaments, may generate substantial deformation. The navigation ability in terms of searching and tagging within liver views has not been characterized, and current object detection methods do not account for the mechanics of how these features could be applied to the liver images. In this research, we have proposed spatial pyramid based searching and tagging of liver's intraoperative views using convolution neural network (SPST-CNN). By exploiting a hybrid combination of an image pyramid at input and spatial pyramid pooling layer at deeper stages of SPST-CNN, we reveal the gains of full-image representations for searching and tagging variable scaled liver live views. SPST-CNN provides pinpoint searching and tagging of intraoperative liver views to obtain up-to-date information about the location and shape of the area of interest. Downsampling input using image pyramid enables SPST-CNN framework to deploy input images with a diversity of resolutions for achieving scale-invariance feature. We have compared the proposed approach to the four recent state-of-the-art approaches and our method achieved better mAP up to 85.9%.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.jbi.2020.103430DOI Listing
June 2020

A Residual Based Attention Model for EEG Based Sleep Staging.

IEEE J Biomed Health Inform 2020 10 3;24(10):2833-2843. Epub 2020 Mar 3.

Sleep staging is to score the sleep state of a subject into different sleep stages such as Wake and Rapid Eye Movement (REM). It plays an indispensable role in the diagnosis and treatment of sleep disorders. As manual sleep staging through well-trained sleep experts is time consuming, tedious, and subjective, many automatic methods have been developed for accurate, efficient, and objective sleep staging. Recently, deep learning based methods have been successfully proposed for electroencephalogram (EEG) based sleep staging with promising results. However, most of these methods directly take EEG raw signals as input of convolutional neural networks (CNNs) without considering the domain knowledge of EEG staging. Apart from that, to capture temporal information, most of the existing methods utilize recurrent neural networks such as LSTM (Long Short Term Memory) which are not effective for modelling global temporal context and difficult to train. Therefore, inspired by the clinical guidelines of sleep staging such as AASM (American Academy of Sleep Medicine) rules where different stages are generally characterized by EEG waveforms of various frequencies, we propose a multi-scale deep architecture by decomposing an EEG signal into different frequency bands as input to CNNs. To model global temporal context, we utilize the multi-head self-attention module of the transformer model to not only improve performance, but also shorten the training time. In addition, we choose residual based architecture which makes training end-to-end. Experimental results on two widely used sleep staging datasets, Montreal Archive of Sleep Studies (MASS) and sleep-EDF datasets, demonstrate the effectiveness and significant efficiency (up to 12 times less training time) of our proposed method over the state-of-the-art.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/JBHI.2020.2978004DOI Listing
October 2020

Cloud-Based Automated Clinical Decision Support System for Detection and Diagnosis of Lung Cancer in Chest CT.

IEEE J Transl Eng Health Med 2020 4;8:4300113. Epub 2019 Dec 4.

6Biomedical and Multimedia Information Technology Research Group, School of Information TechnologiesThe University of SydneySydneyNSW2006Australia.

Lung cancer is a major cause for cancer-related deaths. The detection of pulmonary cancer in the early stages can highly increase survival rate. Manual delineation of lung nodules by radiologists is a tedious task. We developed a novel computer-aided decision support system for lung nodule detection based on a 3D Deep Convolutional Neural Network (3DDCNN) for assisting the radiologists. Our decision support system provides a second opinion to the radiologists in lung cancer diagnostic decision making. In order to leverage 3-dimensional information from Computed Tomography (CT) scans, we applied median intensity projection and multi-Region Proposal Network (mRPN) for automatic selection of potential region-of-interests. Our Computer Aided Diagnosis (CAD) system has been trained and validated using LUNA16, ANODE09, and LIDC-IDR datasets; the experiments demonstrate the superior performance of our system, attaining sensitivity, specificity, AUROC, accuracy, of 98.4%, 92%, 96% and 98.51% with 2.1 FPs per scan. We integrated cloud computing, trained and validated our Cloud-Based 3DDCNN on the datasets provided by Shanghai Sixth People's Hospital, as well as LUNA16, ANODE09, and LIDC-IDR. Our system outperformed the state-of-the-art systems and obtained an impressive 98.7% sensitivity at 1.97 FPs per scan. This shows the potentials of deep learning, in combination with cloud computing, for accurate and efficient lung nodule detection via CT imaging, which could help doctors and radiologists in treating lung cancer patients.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/JTEHM.2019.2955458DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6946021PMC
December 2019

AUNet: attention-guided dense-upsampling networks for breast mass segmentation in whole mammograms.

Phys Med Biol 2020 02 28;65(5):055005. Epub 2020 Feb 28.

Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong 518055, People's Republic of China. School of Control Science and Engineering, Shandong University, Jinan, Shandong 250100, People's Republic of China. These authors contribute equally to this paper.

Mammography is one of the most commonly applied tools for early breast cancer screening. Automatic segmentation of breast masses in mammograms is essential but challenging due to the low signal-to-noise ratio and the wide variety of mass shapes and sizes. Existing methods deal with these challenges mainly by extracting mass-centered image patches manually or automatically. However, manual patch extraction is time-consuming and automatic patch extraction brings errors that could not be compensated in the following segmentation step. In this study, we propose a novel attention-guided dense-upsampling network (AUNet) for accurate breast mass segmentation in whole mammograms directly. In AUNet, we employ an asymmetrical encoder-decoder structure and propose an effective upsampling block, attention-guided dense-upsampling block (AU block). Especially, the AU block is designed to have three merits. Firstly, it compensates the information loss of bilinear upsampling by dense upsampling. Secondly, it designs a more effective method to fuse high- and low-level features. Thirdly, it includes a channel-attention function to highlight rich-information channels. We evaluated the proposed method on two publicly available datasets, CBIS-DDSM and INbreast. Compared to three state-of-the-art fully convolutional networks, AUNet achieved the best performances with an average Dice similarity coefficient of 81.8% for CBIS-DDSM and 79.1% for INbreast.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1088/1361-6560/ab5745DOI Listing
February 2020

Graph Sequence Recurrent Neural Network for Vision-based Freezing of Gait Detection.

IEEE Trans Image Process 2019 Oct 15. Epub 2019 Oct 15.

Freezing of gait (FoG) is one of the most common symptoms of Parkinson's disease (PD), a neurodegenerative disorder which impacts millions of people around the world. Accurate assessment of FoG is critical for the management of PD and to evaluate the efficacy of treatments. Currently, the assessment of FoG requires well-trained experts to perform time-consuming annotations via vision-based observations. Thus, automatic FoG detection algorithms are needed. In this study, we formulate vision-based FoG detection, as a fine-grained graph sequence modelling task, by representing the anatomic joints in each temporal segment with a directed graph, since FoG events can be observed through the motion patterns of joints. A novel deep learning method is proposed, namely graph sequence recurrent neural network (GS-RNN), to characterize the FoG patterns by devising graph recurrent cells, which take graph sequences of dynamic structures as inputs. For the cases of which prior edge annotations are not available, a data-driven based adjacency estimation method is further proposed. To the best of our knowledge, this is one of the first studies on vision-based FoG detection using deep neural networks designed for graph sequences of dynamic structures. Experimental results on more than 150 videos collected from 45 patients demonstrated promising performance of the proposed GS-RNN for FoG detection with an AUC value of 0.90.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2019.2946469DOI Listing
October 2019

Flexible Multi-Layer Semi-Dry Electrode for Scalp EEG Measurements at Hairy Sites.

Micromachines (Basel) 2019 Aug 4;10(8). Epub 2019 Aug 4.

School of Electronic and Information Engineering, South China University of Technology, Guangzhou 510641, China.

One of the major challenges of daily wearable electroencephalogram (EEG) monitoring is that there are rarely suitable EEG electrodes for hairy sites. Wet electrodes require conductive gels, which will dry over the acquisition time, making them unstable for long-term EEG monitoring. Additionally, the electrode-scalp impedances of most dry electrodes are not adequate for high quality EEG collection at hairy sites. In view of the above problems, a flexible multi-layer semi-dry electrode was proposed for EEG monitoring in this study. The semi-dry electrode contains a flexible electrode body layer, foam layer and reservoir layer. The probe structure of the electrode body layer enables the electrode to work effectively at hairy sites. During long-term EEG monitoring, electrolytes stored in the reservoir layer are continuously released through the foam layer to the electrode-scalp interface, ensuring a lower electrode-scalp contact impedance. The experimental results showed that the average electrode-scalp impedance of the semi-dry electrode at a hairy site was only 23.89 ± 7.44 KΩ at 10 Hz, and it was lower than 40 KΩ over a long-term use of 5 h. The electrode performed well in both static and dynamic EEG monitoring, where the temporal correlation with wet electrode signals at the hairy site could reach 94.25% and 90.65%, respectively, and specific evoked EEG signals could be collected. The flexible multi-layer semi-dry electrode can be well applied to scalp EEG monitoring at hairy sites, providing a promising solution for daily long-term monitoring of wearable EEGs.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/mi10080518DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6722968PMC
August 2019

Vision-Based Freezing of Gait Detection With Anatomic Directed Graph Representation.

IEEE J Biomed Health Inform 2020 04 17;24(4):1215-1225. Epub 2019 Jun 17.

Parkinson's disease significantly impacts the life quality of millions of people around the world. While freezing of gait (FoG) is one of the most common symptoms of the disease, it is time consuming and subjective to assess FoG for well-trained experts. Therefore, it is highly desirable to devise computer-aided FoG detection methods for the purpose of objective and time-efficient assessment. In this paper, in line with the gold standard of FoG clinical assessment, which requires video or direct observation, we propose one of the first vision-based methods for automatic FoG detection. To better characterize FoG patterns, instead of learning an overall representation of a video, we propose a novel architecture of graph convolution neural network and represent each video as a directed graph where FoG related candidate regions are the vertices. A weakly-supervised learning strategy and a weighted adjacency matrix estimation layer are proposed to eliminate the resource expensive data annotation required for fully supervised learning. As a result, the interference of visual information irrelevant to FoG, such as gait motion of supporting staff involved in clinical assessments, has been reduced to improve FoG detection performance by identifying the vertices contributing to FoG events. To further improve the performance, the global context of a clinical video is also considered and several fusion strategies with graph predictions are investigated. Experimental results on more than 100 videos collected from 45 patients during a clinical assessment demonstrated promising performance of our proposed method with an AUC of 0.887.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/JBHI.2019.2923209DOI Listing
April 2020

A web-based multidisciplinary team meeting visualisation system.

Int J Comput Assist Radiol Surg 2019 Dec 21;14(12):2221-2231. Epub 2019 May 21.

Biomedical and Multimedia Information Technology Research Group, School of Computer Science, The University of Sydney, Sydney, Australia.

Purpose: Multidisciplinary team meetings (MDTs) are the standard of care for safe, effective patient management in modern hospital-based clinical practice. Medical imaging data are often the central discussion points in many MDTs, and these data are typically visualised, by all participants, on a common large display. We propose a Web-based MDT visualisation system (WMDT-VS) to allow individual participants to view the data on their own personal computing devices with the potential to customise the imaging data, i.e. different view of the data to that of the common display, for their particular clinical perspective.

Methods: We developed the WMDT-VS by leveraging the state-of-the-art Web technologies to support four MDT visualisation features: (1) 2D and 3D visualisations for multiple imaging modality data; (2) a variety of personal computing devices, e.g. smartphone, tablets, laptops and PCs, to access and navigate medical images individually and share the visualisations; (3) customised participant visualisations; and (4) the addition of extra local image data for visualisation and discussion.

Results: We outlined these MDT visualisation features on two simulated MDT settings using different imaging data and usage scenarios. We measured compatibility and performances of various personal, consumer-level, computing devices.

Conclusions: Our WMDT-VS provides a more comprehensive visualisation experience for MDT participants.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s11548-019-01999-xDOI Listing
December 2019

Illumination-Guided Video Composition via Gradient Consistency Optimization.

IEEE Trans Image Process 2019 May 20. Epub 2019 May 20.

Video composition aims at cloning a patch from the source video into the target scene to create a seamless and harmonious blending frame sequence. Previous work in video composition usually suffer from artifacts around the blending region and spatial-temporal consistency when illumination intensity varies in the input source and target video. We propose an illumination-guided video composition method via a unified spatial and temporal optimization framework. Our method can produce globally consistent composition results and maintain the temporal coherency. We first compute a spatial-temporal blending boundary iteratively. For each frame, the gradient field of the target and source frames are mixed adaptively based on gradients and inter-frame color difference. The temporal consistency is further obtained by optimizing luminance gradients throughout all the composition frames. Moreover, we extend the mean-value cloning by smoothing discrepancies between the source and target frames, then eliminate the color distribution overflow exponentially to reduce falsely blending pixels. Various experiments have shown the effectiveness and high-quality performance of our illumination-guided composition.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2019.2916769DOI Listing
May 2019

Non-Contact Sleep Stage Detection Using Canonical Correlation Analysis of Respiratory Sound.

IEEE J Biomed Health Inform 2020 02 11;24(2):614-625. Epub 2019 Apr 11.

Respiratory sound is able to differentiate sleep stages and provide a non-contact and cost-effective solution for the diagnosis and treatment monitoring of sleep-related diseases. While most of the existing respiratory sound-based methods focus on a limited number of sleep stages such as sleep/wake and wake/rapid eye movement (REM)/non-REM, it is essential to detect sleep stages at a finer level for sleep quality evaluation. In this paper, we for the first time study a sleep stage detection method aiming at classifying sleep states into four sleep stages: wake, REM, light sleep, and deep sleep from the respiratory sound. In addition to extracting time-domain features, frequency-domain features of respiratory sound, non-linear features of snoring sound are devised to better characterize snoring-related signals of respiratory sound. To effectively fuse the three sets of features, a novel feature fusion technique combining the generalized canonical correlation analysis with the ReliefF algorithm is proposed for discriminative feature selection. Final stage detection is achieved with popular classifiers including decision tree, support vector machines, K-nearest neighbor, and the ensemble classifier. To evaluate our proposed method, we built an in-house dataset, which is comprised of 13 nights of sleep audio data from a sleep laboratory. Experimental results indicate that our proposed method outperforms the existing related ones and is promising for large-scale non-contact sleep monitoring.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/JBHI.2019.2910566DOI Listing
February 2020

Liver Extraction Using Residual Convolution Neural Networks From Low-Dose CT Images.

IEEE Trans Biomed Eng 2019 09 21;66(9):2641-2650. Epub 2019 Jan 21.

An efficient and precise liver extraction from computed tomography (CT) images is a crucial step for computer-aided hepatic diseases diagnosis and treatment. Considering the possible risk to patient's health due to X-ray radiation of repetitive CT examination, low-dose CT (LDCT) is an effective solution for medical imaging. However, inhomogeneous appearances and indistinct boundaries due to additional noise and streaks artifacts in LDCT images often make it a challenging task. This study aims to extract a liver model from LDCT images for facilitating medical expert in surgical planning and post-operative assessment along with low radiation risk to the patient. Our method carried out liver extraction by employing residual convolutional neural networks (LER-CN), which is further refined by noise removal and structure preservation components. After patch-based training, our LER-CN shows a competitive performance relative to state-of-the-art methods for both clinical and publicly available MICCAI Sliver07 datasets. We have proposed training and learning algorithms for LER-CN based on back propagation gradient descent. We have evaluated our method on 150 abdominal CT scans for liver extraction. LER-CN achieves dice similarity coefficient up to 96.5[Formula: see text], decreased volumetric overlap error up to 4.30[Formula: see text], and average symmetric surface distance less than 1.4 [Formula: see text]. These findings have shown that LER-CN is a favorable method for medical applications with high efficiency allowing low radiation risk to patients.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TBME.2019.2894123DOI Listing
September 2019

A direct volume rendering visualization approach for serial PET-CT scans that preserves anatomical consistency.

Int J Comput Assist Radiol Surg 2019 May 19;14(5):733-744. Epub 2019 Jan 19.

Sydney Medical School, The University of Sydney, Sydney, Australia.

Purpose: Our aim was to develop an interactive 3D direct volume rendering (DVR) visualization solution to interpret and analyze complex, serial multi-modality imaging datasets from positron emission tomography-computed tomography (PET-CT).

Methods: Our approach uses: (i) a serial transfer function (TF) optimization to automatically depict particular regions of interest (ROIs) over serial datasets with consistent anatomical structures; (ii) integration of a serial segmentation algorithm to interactively identify and track ROIs on PET; and (iii) parallel graphics processing unit (GPU) implementation for interactive visualization.

Results: Our DVR visualization more easily identifies changes in ROIs in serial scans in an automated fashion and parallel GPU computation which enables interactive visualization.

Conclusions: Our approach provides a rapid 3D visualization of relevant ROIs over multiple scans, and we suggest that it can be used as an adjunct to conventional 2D viewing software from scanner vendors.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s11548-019-01916-2DOI Listing
May 2019

Image-Aligned Dynamic Liver Reconstruction Using Intra-Operative Field of Views for Minimal Invasive Surgery.

IEEE Trans Biomed Eng 2019 08 30;66(8):2163-2173. Epub 2018 Nov 30.

During hepatic minimal invasive surgery (MIS), 3-D reconstruction of a liver surface by interpreting the geometry of its soft tissues is achieving attractions. One of the major issues to be addressed in MIS is liver deformation. Moreover, it severely inhibits free sight and dexterity of tissue manipulation, which causes its intra-operative morphology and soft tissue motion altered as compared to its pre-operative shape. While many applications focus on 3-D reconstruction of rigid or semi-rigid scenes, the techniques applied in hepatic MIS must be able to cope with a dynamic and deformable environment. We propose an efficient technique for liver surface reconstruction based on the structure from motion to handle liver deformation. The reconstructed liver will assist surgeons to visualize liver surface more efficiently with better depth perception. We use the intra-operative field of views to generate 3-D template mesh from a dense keypoint cloud. We estimate liver deformation by finding best correspondence between 3-D templates and reconstruct a liver image to calculate translation and rotational motions. Our technique then finely tunes deformed surface by adding smoothness using shading cues. Up till now, this technique is not used for solving the human liver deformation problem. Our approach is tested and validated with synthetic as well as real in vivo data, which reveal that the reconstruction accuracy can be enhanced using our approach even in challenging laparoscopic environments.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TBME.2018.2884319DOI Listing
August 2019

Key Marker Selection for the Detection of Early Parkinson' s Disease using Importance-Driven Models.

Annu Int Conf IEEE Eng Med Biol Soc 2018 Jul;2018:6100-6103

The detection of early Parkinson' s disease (PD) is crucial for PD management. Most of previous efforts on PD diagnosis focus more on improving PD detection accuracies by trying using features from more modalities, which results in a common question: is it true that the more features available, the better the performance of the diagnosis system? This paper proposes an importance-driven approach for the detection of PD. The importance of features based on gradient boosting is firstly learned. The ranked features based on feature importance are input to a progressive learning pipeline to find key features of PD. The experiment results show that a comparable PD classification performance can be obtained with much less key features and therefore fewer modalities of tests are required. Such findings have critical socioeconomic values.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/EMBC.2018.8513564DOI Listing
July 2018

Ischemic stroke clinical outcome prediction based on image signature selection from multimodality data.

Annu Int Conf IEEE Eng Med Biol Soc 2018 Jul;2018:722-725

Quantitative models are essential in precision medicine that can be used to predict health status and prevent disease and disability. Current radiomics models for clinical outcome prediction often depend on huge amount of image features and may include redundant information and ignore individual feature importance. In this work, we propose a prognostic discrimination ranking strategy to select the most relevant image features for image assisted clinical outcome prediction. Firstly, a redundancy and prognostic discrimination evaluation method is proposed to evaluate and rank a large number of features extracted from images. Secondly, forward sequential feature selection is performed to select the top ranked relevant features in each discriminate quantization. Finally, representative vectors are generated by the fusion of pivotal clinical parameters and selected image signatures to be fed into a classification model. The proposed model was trained and tested over 70 patient studies with six MR sequences and four clinical parameters from ISLES challenges. The evaluations using ROC curves demonstrated the improved performance over five other feature selection models where the proposed model achieved AUCs of 0.821, 0.968, 0.983, 0.896 and 1 when predicting five clinical outcome scores respectively.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/EMBC.2018.8512291DOI Listing
July 2018

Cancer type prediction based on copy number aberration and chromatin 3D structure with convolutional neural networks.

BMC Genomics 2018 Aug 13;19(Suppl 6):565. Epub 2018 Aug 13.

Key Laboratory of Systems Biomedicine, Shanghai Center for Systems Biomedicine, Shanghai Jiaotong University, Shanghai, 200240, China.

Background: With the developments of DNA sequencing technology, large amounts of sequencing data have been produced that provides unprecedented opportunities for advanced association studies between somatic mutations and cancer types/subtypes which further contributes to more accurate somatic mutation based cancer typing (SMCT). In existing SMCT methods however, the absence of high-level feature extraction is a major obstacle in improving the classification performance.

Results: We propose DeepCNA, an advanced convolutional neural network (CNN) based classifier, which utilizes copy number aberrations (CNAs) and HiC data, to address this issue. DeepCNA first pre-process the CNA data by clipping, zero padding and reshaping. Then, the processed data is fed into a CNN classifier, which extracts high-level features for accurate classification. Experimental results on the COSMIC CNA dataset indicate that 2D CNN with both cell lines of HiC data lead to the best performance. We further compare DeepCNA with three widely adopted classifiers, and demonstrate that DeepCNA has at least 78% improvement of performance.

Conclusions: This paper demonstrates the advantages and potential of the proposed DeepCNA model for processing of somatic point mutation based gene data, and proposes that its usage may be extended to other complex genotype-phenotype association studies.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12864-018-4919-zDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6101087PMC
August 2018

Deep Color Guided Coarse-to-Fine Convolutional Network Cascade for Depth Image Super-Resolution.

IEEE Trans Image Process 2018 Oct 8. Epub 2018 Oct 8.

Depth image super-resolution is a significant yet challenging task. In this paper, we introduce a novel deep color guided coarse-to-fine convolutional neural network (CNN) framework to address this problem. First, we present a datadriven filter method to approximate the ideal filter for depth image super-resolution instead of hand-designed filters. Based on large data samples, the filter learned is more accurate and stable for upsampling depth image. Second, we introduce a coarse-to-fine CNN to learn different sizes of filter kernels. In coarse stage, larger filter kernels are learned by CNN to achieve crude high-resolution depth image. As to fine stage, the crude high-resolution depth image is used as the input so that smaller filter kernels are learned to gain more accurate results. Benefit from this network, we can progressively recover the high frequency details. Third, we construct a color guidance strategy that fuses color difference and spatial distance for depth image upsampling. We revise the interpolated high-resolution depth image according to the corresponding pixels in highresolution color maps. Guided by color information, the depth of high-resolution image obtained can alleviate texture copying artifacts and preserve edge details effectively. Quantitative and qualitative experimental results demonstrate our state-of-the-art performance for depth map super-resolution.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2018.2874285DOI Listing
October 2018

Retinal Vessel Segmentation Using Minimum Spanning Superpixel Tree Detector.

IEEE Trans Cybern 2019 Jul 22;49(7):2707-2719. Epub 2018 May 22.

The retinal vessel is one of the determining factors in an ophthalmic examination. Automatic extraction of retinal vessels from low-quality retinal images still remains a challenging problem. In this paper, we propose a robust and effective approach that qualitatively improves the detection of low-contrast and narrow vessels. Rather than using the pixel grid, we use a superpixel as the elementary unit of our vessel segmentation scheme. We regularize this scheme by combining the geometrical structure, texture, color, and space information in the superpixel graph. And the segmentation results are then refined by employing the efficient minimum spanning superpixel tree to detect and capture both global and local structure of the retinal images. Such an effective and structure-aware tree detector significantly improves the detection around the pathologic area. Experimental results have shown that the proposed technique achieves advantageous connectivity-area-length (CAL) scores of 80.92% and 69.06% on two public datasets, namely, DRIVE and STARE, thereby outperforming state-of-the-art segmentation methods. In addition, the tests on the challenging retinal image database have further demonstrated the effectiveness of our method. Our approach achieves satisfactory segmentation performance in comparison with state-of-the-art methods. Our technique provides an automated method for effectively extracting the vessel from fundus images.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TCYB.2018.2833963DOI Listing
July 2019

Classification of Medical Images in the Biomedical Literature by Jointly Using Deep and Handcrafted Visual Features.

IEEE J Biomed Health Inform 2018 09 20;22(5):1521-1530. Epub 2017 Nov 20.

The classification of medical images and illustrations from the biomedical literature is important for automated literature review, retrieval, and mining. Although deep learning is effective for large-scale image classification, it may not be the optimal choice for this task as there is only a small training dataset. We propose a combined deep and handcrafted visual feature (CDHVF) based algorithm that uses features learned by three fine-tuned and pretrained deep convolutional neural networks (DCNNs) and two handcrafted descriptors in a joint approach. We evaluated the CDHVF algorithm on the ImageCLEF 2016 Subfigure Classification dataset and it achieved an accuracy of 85.47%, which is higher than the best performance of other purely visual approaches listed in the challenge leaderboard. Our results indicate that handcrafted features complement the image representation learned by DCNNs on small training datasets and improve accuracy in certain medical image classification problems.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/JBHI.2017.2775662DOI Listing
September 2018

Study of an Oxygen Supply and Oxygen Saturation Monitoring System for Radiation Therapy Associated with the Active Breathing Coordinator.

Sci Rep 2018 01 19;8(1):1254. Epub 2018 Jan 19.

Biomedical And Multimedia Information Technology (BMIT) Research Group, School Of Information Technologies (SIT), The University Of Sydney, Sydney, Nsw, 2008, Australia.

In this study, we designed an oxygen supply and oxygen saturation monitoring (OSOSM) system. This OSOSM system can provide a continuous supply of oxygen and monitor the peripheral capillary oxygen saturation (SpO2) of patients who accept radiotherapy and use an active breathing coordinator (ABC). A clinical test with 27 volunteers was conducted. The volunteers were divided into two groups based on the tendency of SpO2 decline in breath-holding without the OSOSM system: group A (12 cases) showed a decline in SpO2 of less than 2%, whereas the decline in SpO2 in group B (15 cases) was greater than 2% and reached up to 6% in some cases. The SpO2 of most volunteers declined during rest. The breath-holding time of group A without the OSOSM system was significantly longer than that of group B (p < 0.05) and was extended with the OSOSM system by 26.6% and 27.85% in groups A and B, respectively. The SpO2 recovery time was reduced by 36.1%, and the total rest time was reduced by 27.6% for all volunteers using the OSOSM system. In summary, SpO2 declines during breath-holding and rest time cannot be ignored while applying an ABC. This OSOSM system offers a simple and effective way to monitor SpO2 variation and overcome SpO2 decline, thereby lengthening breath-holding time and shortening rest time.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41598-018-19576-8DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5775202PMC
January 2018

Multi-view collaborative segmentation for prostate MRI images.

Annu Int Conf IEEE Eng Med Biol Soc 2017 Jul;2017:3529-3532

Prostate delineation from MRI images is a prolonged challenging issue partially due to appearance variations across patients and disease progression. To address these challenges, our proposed collaborative method takes into account the computed multiple label-relevance maps as multiple views for learning the optimal boundary delineation. In our method, we firstly extracted multiple label-relevance maps to represent the affinities between each unlabeled pixel to the pre-defined labels to avoid the selection of handcrafted features. Then these maps were incorporated in a collaborative clustering to learn the adaptive weights for an optimal segmentation which overcomes the seeds selection sensitivity problems. The segmentation results were evaluated over 22 prostate MRI patient studies with respect to dice similarity coefficient (DSC), absolute relative volume difference (ARVD) and average symmetric surface distance (ASSD) (mm). The results and t-Test demonstrated that the proposed method improved the segmentation accuracy and robustness and the improvement was statistically significant.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/EMBC.2017.8037618DOI Listing
July 2017

Structure and location preserving topological representation with applications on CT segmentation.

Annu Int Conf IEEE Eng Med Biol Soc 2017 Jul;2017:548-551

Contour tree represents the nesting relations of the individual components in the image; however, it neglects the geometric structure of the terrain. In this paper, we propose a new topological representation that provides the nesting and spatial relations of regions for CT image interpretation. The tree is constructed based on the signed distance transformation of binary CT image, and combines intensity based contour tree and a new gradient based topology tree. We also investigate the application of the topological representation as a constraint in target object segmentation. Ten non-small cell lung tumor CT studies were used for segmentation accuracy evaluation. The image representation results showed that the proposed tree structure retained the nesting and spatial relations of the tissues or objects in the CT image. The segmentation results demonstrated its usability in the separation of adjacent objects with similar intensity distributions.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/EMBC.2017.8036883DOI Listing
July 2017

Reversion Correction and Regularized Random Walk Ranking for Saliency Detection.

IEEE Trans Image Process 2018 Mar 12;27(3):1311-1322. Epub 2017 Oct 12.

In recent saliency detection research, many graph-based algorithms have applied boundary priors as background queries, which may generate completely "reversed" saliency maps if the salient objects are on the image boundaries. Moreover, these algorithms usually depend heavily on pre-processed superpixel segmentation, which may lead to notable degradation in image detail features. In this paper, a novel saliency detection method is proposed to overcome the above issues. First, we propose a saliency reversion correction process, which locates and removes the boundary-adjacent foreground superpixels, and thereby increases the accuracy and robustness of the boundary prior-based saliency estimations. Second, we propose a regularized random walk ranking model, which introduces prior saliency estimation to every pixel in the image by taking both region and pixel image features into account, thus leading to pixel-detailed and superpixel-independent saliency maps. Experiments are conducted on four well-recognized data sets; the results indicate the superiority of our proposed method against 14 state-of-the-art methods, and demonstrate its general extensibility as a saliency optimization algorithm. We further evaluate our method on a new data set comprised of images that we define as boundary adjacent object saliency, on which our method performs better than the comparison methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2017.2762422DOI Listing
March 2018

Foreground Detection With Simultaneous Dictionary Learning and Historical Pixel Maintenance.

IEEE Trans Image Process 2016 Nov;25(11):5035-5049

Foreground detection is fundamental in surveillance video analysis and meaningful toward object tracking and higher level tasks, such as anomaly detection and activity analysis. Nevertheless, existing methods are still limited in accurately detecting the foreground due to the complex scene settings. To robustly handle the diverse background variations and foreground challenges, this paper proposes a Background REpresentation approach With Dictionary Learning and Historical Pixel Maintenance (BREW-DLHPM). Specifically, a dictionary learning problem is formulated at the frame level to adaptively represent the background signals with the varied structure information captured, while a pixel-level maintenance is exploited to grasp the dynamic nature of historical information under the help of the learned background. The simultaneous utilization of dictionary learning and historical pixel maintenance facilitates the accurate description of the background and thus guides a wise foreground detection decision. The proposed BREW-DLHPM has been evaluated on the prestigious change detection challenge data set against 11 state-of-the-art foreground detection approaches and encouraging performances have been achieved by our method.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2016.2598680DOI Listing
November 2016
-->