Chronic Wound Image Augmentation and Assessment Using Semi-Supervised Progressive Multi-Granularity EfficientNet.

Goal: Augment a small, imbalanced, wound dataset by using semi-supervised learning with a secondary dataset. Then utilize the augmented wound dataset for deep learning-based wound assessment. Methods: The clinically-validated Photographic Wound Assessment Tool (PWAT) scores eight wound attributes: Size, Depth, Necrotic Tissue Type, Necrotic Tissue Amount, Granulation Tissue type, Granulation Tissue Amount, Edges, Periulcer Skin Viability to comprehensively assess chronic wound images. A small corpus of 1639 wound images labeled with ground truth PWAT scores was used as reference. A Semi-Supervised learning and Progressive Multi-Granularity training mechanism were used to leverage a secondary corpus of 9870 unlabeled wound images. Wound scoring utilized the EfficientNet Convolutional Neural Network on the augmented wound corpus. Results: Our proposed Semi-Supervised PMG EfficientNet (SS-PMG-EfficientNet) approach estimated all 8 PWAT sub-scores with classification accuracies and F1 scores of about 90% on average, and outperformed a comprehensive list of baseline models and had a 7% improvement over the prior state-of-the-art (without data augmentation). We also demonstrate that synthetic wound image generation using Generative Adversarial Networks (GANs) did not improve wound assessment. Conclusions: Semi-supervised learning on unlabeled wound images in a secondary dataset achieved impressive performance for deep learning-based wound grading.

Auto-segmentation of Adult-Type Diffuse Gliomas: Comparison of Transfer Learning-Based Convolutional Neural Network Model vs. Radiologists.

Segmentation of glioma is crucial for quantitative brain tumor assessment, to guide therapeutic research and clinical management, but very time-consuming. Fully automated tools for the segmentation of multi-sequence MRI are needed. We developed and pretrained a deep learning (DL) model using publicly available datasets A (n = 210) and B (n = 369) containing FLAIR, T2WI, and contrast-enhanced (CE)-T1WI. This was then fine-tuned with our institutional dataset (n = 197) containing ADC, T2WI, and CE-T1WI, manually annotated by radiologists, and split into training (n = 100) and testing (n = 97) sets. The Dice similarity coefficient (DSC) was used to compare model outputs and manual labels. A third independent radiologist assessed segmentation quality on a semi-quantitative 5-scale score. Differences in DSC between new and recurrent gliomas, and between uni or multifocal gliomas were analyzed using the Mann-Whitney test. Semi-quantitative analyses were compared using the chi-square test. We found that there was good agreement between segmentations from the fine-tuned DL model and ground truth manual segmentations (median DSC: 0.729, std-dev: 0.134). DSC was higher for newly diagnosed (0.807) than recurrent (0.698) (p < 0.001), and higher for unifocal (0.747) than multi-focal (0.613) cases (p = 0.001). Semi-quantitative scores of DL and manual segmentation were not significantly different (mean: 3.567 vs. 3.639; 93.8% vs. 97.9% scoring ≥ 3, p = 0.107). In conclusion, the proposed transfer learning DL performed similarly to human radiologists in glioma segmentation on both structural and ADC sequences. Further improvement in segmenting challenging postoperative and multifocal glioma cases is needed.

Respiratory signal estimation for cardiac perfusion SPECT using deep learning.

BACKGROUND: Respiratory motion induces artifacts in reconstructed cardiac perfusion SPECT images. Correction for respiratory motion often relies on a respiratory signal describing the heart displacements during breathing. However, using external tracking devices to estimate respiratory signals can add cost and operational complications in a clinical setting. PURPOSE: We aim to develop a deep learning (DL) approach that uses only SPECT projection data for respiratory signal estimation. METHODS: A modified U-Net was implemented that takes temporally finely sampled SPECT sub-projection data (100 ms) as input. These sub-projections are obtained by reframing the 20-s list-mode data, resulting in 200 sub-projections, at each projection angle for each SPECT camera head. The network outputs a 200-time-point motion signal for each projection angle, which was later aggregated over all angles to give a full respiratory signal. The target signal for DL model training was from an external stereo-camera visual tracking system (VTS). In addition to comparing DL and VTS, we also included a data-driven approach based on the center-of-mass (CoM) strategy. This CoM method estimates respiratory signals by monitoring the axial changes of CoM for counts in the heart region of the sub-projections. We utilized 900 subjects with stress cardiac perfusion SPECT studies, with 302 subjects for testing and the remaining 598 subjects for training and validation. RESULTS: The Pearson's correlation coefficient between the DL respiratory signal and the reference VTS signal was 0.90, compared to 0.70 between the CoM signal and the reference. For respiratory motion correction on SPECT images, all VTS, DL, and CoM approaches partially de-blured the heart wall, resulting in a thinner wall thickness and increased recovered maximal image intensity within the wall, with VTS reducing blurring the most followed by the DL approach. Uptake quantification for the combined anterior and inferior segments of polar maps showed a mean absolute difference from the reference VTS of 1.7% for the DL method for patients with motion >12 mm, compared to 2.6% for the CoM method and 8.5% for no correction. CONCLUSION: We demonstrate the capability of a DL approach to estimate respiratory signal from SPECT projection data for cardiac perfusion imaging. Our results show that the DL based respiratory motion correction reduces artefacts and achieves similar regional quantification to that obtained using the stereo-camera VTS signals. This may enable fully automatic data-driven respiratory motion correction without relying on external motion tracking devices.

Assessment of thrombectomy procedure difficulty by neurointerventionalists based on vessel geometry parameters from carotid artery 3D reconstructions.

BACKGROUND: Diagnosing and treating acute ischemic stroke patients within a narrow timeframe is challenging. Time needed to access the occluded vessel and initiate thrombectomy is dictated by the availability of information regarding vascular anatomy and trajectory. Absence of such information potentially impacts device selection, procedure success, and stroke outcomes. While the cervical vessels allow neurointerventionalists to navigate devices to the occlusion site, procedures are often encumbered due to tortuous pathways. The purpose of this retrospective study was to determine how neurointerventionalists consider the physical nature of carotid segments when evaluating a procedure's difficulty. METHODS: Seven neurointerventionalists reviewed 3D reconstructions of CT angiograms of left and right carotid arteries from 49 subjects and rated the perceived procedural difficulty on a three-point scale (easy, medium, difficult) to reach the targeted M1. Twenty-two vessel metrics were quantified by dividing the carotids into 5 segments and measuring the radius of curvature, tortuosity, vessel radius, and vessel length of each segment. RESULTS: The tortuosity and length of the arch-cervical and cervical regions significantly impacted difficulty ratings. Additionally, two-way interaction between the radius of curvature and tortuosity on the arch-cervical region was significant (p < 0.0001) wherein, for example, at a given arch-cervical tortuosity, an increased radius of curvature reduced the perceived case difficulty. CONCLUSIONS: Examining the vessel metrics and providing detailed vascular data tailored to patient characteristics may result in better procedure preparation, facilitate faster vessel access time, and improve thrombectomy outcomes. Additionally, documenting these correlations can enhance device design to ensure they suitably function under various vessel conditions.

Observer studies of image quality of denoising reduced-count cardiac single photon emission computed tomography myocardial perfusion imaging by three-dimensional Gaussian post-reconstruction filtering and deep learning.

BACKGROUND: The aim of this research was to asses perfusion-defect detection-accuracy by human observers as a function of reduced-counts for 3D Gaussian post-reconstruction filtering vs deep learning (DL) denoising to determine if there was improved performance with DL. METHODS: SPECT projection data of 156 normally interpreted patients were used for these studies. Half were altered to include hybrid perfusion defects with defect presence and location known. Ordered-subset expectation-maximization (OSEM) reconstruction was employed with the optional correction of attenuation (AC) and scatter (SC) in addition to distance-dependent resolution (RC). Count levels varied from full-counts (100%) to 6.25% of full-counts. The denoising strategies were previously optimized for defect detection using total perfusion deficit (TPD). Four medical physicist (PhD) and six physician (MD) observers rated the slices using a graphical user interface. Observer ratings were analyzed using the LABMRMC multi-reader, multi-case receiver-operating-characteristic (ROC) software to calculate and compare statistically the area-under-the-ROC-curves (AUCs). RESULTS: For the same count-level no statistically significant increase in AUCs for DL over Gaussian denoising was determined when counts were reduced to either the 25% or 12.5% of full-counts. The average AUC for full-count OSEM with solely RC and Gaussian filtering was lower than for the strategies with AC and SC, except for a reduction to 6.25% of full-counts, thus verifying the utility of employing AC and SC with RC. CONCLUSION: We did not find any indication that at the dose levels investigated and with the DL network employed, that DL denoising was superior in AUC to optimized 3D post-reconstruction Gaussian filtering.

Mesh modeling of system geometry and anatomy phantoms for realistic GATE simulations and their inclusion in SPECT reconstruction.

Objective.Monte-Carlo simulation studies have been essential for advancing various developments in single photon emission computed tomography (SPECT) imaging, such as system design and accurate image reconstruction. Among the simulation software available, Geant4 application for tomographic emission (GATE) is one of the most used simulation toolkits in nuclear medicine, which allows building systems and attenuation phantom geometries based on the combination of idealized volumes. However, these idealized volumes are inadequate for modeling free-form shape components of such geometries. Recent GATE versions alleviate these major limitations by allowing users to import triangulated surface meshes.Approach.In this study, we describe our mesh-based simulations of a next-generation multi-pinhole SPECT system dedicated to clinical brain imaging, called AdaptiSPECT-C. To simulate realistic imaging data, we incorporated in our simulation the XCAT phantom, which provides an advanced anatomical description of the human body. An additional challenge with the AdaptiSPECT-C geometry is that the default voxelized XCAT attenuation phantom was not usable in our simulation due to intersection of objects of dissimilar materials caused by overlap of the air containing regions of the XCAT beyond the surface of the phantom and the components of the imaging system.Main results.We validated our mesh-based modeling against the one constructed by idealized volumes for a simplified single vertex configuration of AdaptiSPECT-C through simulated projection data of123I-activity distributions. We resolved the overlap conflict by creating and incorporating a mesh-based attenuation phantom following a volume hierarchy. We then evaluated our reconstructions with attenuation and scatter correction for projections obtained from simulation consisting of mesh-based modeling of the system and the attenuation phantom for brain imaging. Our approach demonstrated similar performance as the reference scheme simulated in air for uniform and clinical-like123I-IMP brain perfusion source distributions.Significance.This work enables the simulation of complex SPECT acquisitions and reconstructions for emulating realistic imaging data close to those of actual patients.

A Smartphone Physical Activity App for Patients in Alcohol Treatment: Single-Arm Feasibility Trial.

BACKGROUND: Alcohol use disorder (AUD) is a significant public health concern worldwide. Alcohol consumption is a leading cause of death in the United States and has a significant negative impact on individuals and society. Relapse following treatment is common, and adjunct intervention approaches to improve alcohol outcomes during early recovery continue to be critical. Interventions focused on increasing physical activity (PA) may improve AUD treatment outcomes. Given the ubiquity of smartphones and activity trackers, integrating this technology into a mobile app may be a feasible, acceptable, and scalable approach for increasing PA in individuals with AUD. OBJECTIVE: This study aims to test the Fit&Sober app developed for patients with AUD. The goals of the app were to facilitate self-monitoring of PA engagement and daily mood and alcohol cravings, increase awareness of immediate benefits of PA on mood and cravings, encourage setting and adjusting PA goals, provide resources and increase knowledge for increasing PA, and serve as a resource for alcohol relapse prevention strategies. METHODS: To preliminarily test the Fit&Sober app, we conducted an open pilot trial of patients with AUD in early recovery (N=22; 13/22, 59% women; mean age 43.6, SD 11.6 years). At the time of hospital admission, participants drank 72% of the days in the last 3 months, averaging 9 drinks per drinking day. The extent to which the Fit&Sober app was feasible and acceptable among patients with AUD during early recovery was examined. Changes in alcohol consumption, PA, anxiety, depression, alcohol craving, and quality of life were also examined after 12 weeks of app use. RESULTS: Participants reported high levels of satisfaction with the Fit&Sober app. App metadata suggested that participants were still using the app approximately 2.5 days per week by the end of the intervention. Pre-post analyses revealed small-to-moderate effects on increase in PA, from a mean of 5784 (SD 2511) steps per day at baseline to 7236 (SD 3130) steps per day at 12 weeks (Cohen d=0.35). Moderate-to-large effects were observed for increases in percentage of abstinent days (Cohen d=2.17) and quality of life (Cohen d=0.58) as well as decreases in anxiety (Cohen d=-0.71) and depression symptoms (Cohen d=-0.58). CONCLUSIONS: The Fit&Sober app is an acceptable and feasible approach for increasing PA in patients with AUD during early recovery. A future randomized controlled trial is necessary to determine the efficacy of the Fit&Sober app for long-term maintenance of PA, ancillary mental health, and alcohol outcomes. If the efficacy of the Fit&Sober app could be established, patients with AUD would have a valuable adjunct to traditional alcohol treatment that can be delivered in any setting and at any time, thereby improving the overall health and well-being of this population. TRIAL REGISTRATION: ClinicalTrials.gov NCT02958280; https://www.clinicaltrials.gov/ct2/show/NCT02958280.

Cumulative effect of simvastatin, L-arginine, and tetrahydrobiopterin on cerebral blood flow and cognitive function in Alzheimer's disease.

BACKGROUND AND OBJECTIVES: Vascular disease is a known risk factor for Alzheimer's disease (AD). Endothelial dysfunction has been linked to reduced cerebral blood flow. Endothelial nitric oxide synthase pathway (eNOS) upregulation is known to support endothelial health. This single-center, proof-of-concept study tested whether the use of three medications known to augment the eNOS pathway activity improves cognition and cerebral blood flow (CBF). METHODS: Subjects with mild AD or mild cognitive impairment (MCI) were sequentially treated with the HMG-CoA reductase synthesis inhibitor simvastatin (weeks 0-16), L-arginine (weeks 4-16), and tetrahydrobiopterin (weeks 8-16). The primary outcome of interest was the change in CBF as measured by MRI from baseline to week 16. Secondary outcomes included standard assessments of cognition. RESULTS: A total of 11 subjects were deemed eligible and enrolled. One subject withdrew from the study after enrollment, leaving 10 subjects for data analysis. There was a significant increase in CBF from baseline to week 8 by ~13% in the limbic and ~15% in the cerebral cortex. Secondary outcomes indicated a modest but significant increase in the MMSE from baseline (24.2±3.2) to week 16 (26.0±2.7). Exploratory analysis indicated that subjects with cognitive improvement (reduction of the ADAS-cog 13) had a significant increase in their respective limbic and cortical CBF. CONCLUSIONS: Treatment of mild AD/MCI subjects with medications shown to augment the eNOS pathway was well tolerated and associated with modestly increased cerebral blood flow and cognitive improvement. TRIAL REGISTRATION: This study is registered in https://www. CLINICALTRIALS: gov ; registration identifier: NCT01439555; date of registration submitted to registry: 09/23/2011; date of first subject enrollment: 11/2011.

Exploring Factors Associated With Mobile Phone Behaviors and Attitudes Toward Technology Among Adults With Alcohol Use Disorder and Implications for mHealth Interventions: Exploratory Study.

BACKGROUND: Alcohol use disorder (AUD) is associated with severe chronic medical conditions and premature mortality. Expanding the reach or access to effective evidence-based treatments to help persons with AUD is a public health objective. Mobile phone or smartphone technology has the potential to increase the dissemination of clinical and behavioral interventions (mobile health interventions) that increase the initiation and maintenance of sobriety among individuals with AUD. Studies about how this group uses their mobile phone and their attitudes toward technology may have meaningful implications for participant engagement with these interventions. OBJECTIVE: This exploratory study examined the potential relationships among demographic characteristics (race, gender, age, marital status, and income), substance use characteristics (frequency of alcohol and cannabis use), and clinical variables (anxiety and depression symptoms) with indicators of mobile phone use behaviors and attitudes toward technology. METHODS: A sample of 71 adults with AUD (mean age 42.9, SD 10.9 years) engaged in an alcohol partial hospitalization program completed 4 subscales from the Media Technology Usage and Attitudes assessment: Smartphone Usage measures various mobile phone behaviors and activities, Positive Attitudes and Negative Attitudes measure attitudes toward technology, and the Technological Anxiety/Dependence measure assesses level of anxiety when individuals are separated from their phone and dependence on this device. Participants also provided demographic information and completed the Epidemiologic Studies Depression Scale (CES-D) and the Generalized Anxiety Disorder (GAD-7) scale. Lastly, participants reported their frequency of alcohol use over the past 3 months using the Drug Use Frequency Scale. RESULTS: Results for the demographic factors showed a significant main effect for age, Smartphone Usage (P=.003; ηp2=0.14), and Positive Attitudes (P=.01; ηp2=0.07). Marital status (P=.03; ηp2=0.13) and income (P=.03; ηp2=0.14) were associated only with the Technological Anxiety and Dependence subscale. Moreover, a significant trend was found for alcohol use and the Technological Anxiety/Dependence subscale (P=.06; R2=0.02). Lastly, CES-D scores (P=.03; R2=0.08) and GAD symptoms (P=.004; R2=0.13) were significant predictors only of the Technological Anxiety/Dependence subscale. CONCLUSIONS: Findings indicate differences in mobile phone use patterns and attitudes toward technology across demographic, substance use, and clinical measures among patients with AUD. These results may help inform the development of future mHealth interventions among this population.

Comprehensive Assessment of Fine-Grained Wound Images Using a Patch-Based CNN With Context-Preserving Attention.

GOAL: Chronic wounds affect 6.5 million Americans. Wound assessment via algorithmic analysis of smartphone images has emerged as a viable option for remote assessment. METHODS: We comprehensively score wounds based on the clinically-validated Photographic Wound Assessment Tool (PWAT), which comprehensively assesses clinically important ranges of eight wound attributes: Size, Depth, Necrotic Tissue Type, Necrotic Tissue Amount, Granulation Tissue type, Granulation Tissue Amount, Edges, Periulcer Skin Viability. We proposed a DenseNet Convolutional Neural Network (CNN) framework with patch-based context-preserving attention to assess the 8 PWAT attributes of four wound types: diabetic ulcers, pressure ulcers, vascular ulcers and surgical wounds. RESULTS: In an evaluation on our dataset of 1639 wound images, our model estimated all 8 PWAT sub-scores with classification accuracies and F1 scores of over 80%. CONCLUSIONS: Our work is the first intelligent system that autonomously grades wounds comprehensively based on criteria in the PWAT rubric, alleviating the significant burden that manual wound grading imposes on wound care nurses.

Machine learning models for synthesizing actionable care decisions on lower extremity wounds.

Lower extremity chronic wounds affect 4.5 million Americans annually. Due to inadequate access to wound experts in underserved areas, many patients receive non-uniform, non-standard wound care, resulting in increased costs and lower quality of life. We explored machine learning classifiers to generate actionable wound care decisions about four chronic wound types (diabetic foot, pressure, venous, and arterial ulcers). These decisions (target classes) were: (1) Continue current treatment, (2) Request non-urgent change in treatment from a wound specialist, (3) Refer patient to a wound specialist. We compare classification methods (single classifiers, bagged & boosted ensembles, and a deep learning network) to investigate (1) whether visual wound features are sufficient for generating a decision and (2) whether adding unstructured text from wound experts increases classifier accuracy. Using 205 wound images, the Gradient Boosted Machine (XGBoost) outperformed other methods when using both visual and textual wound features, achieving 81% accuracy. Using only visual features decreased the accuracy to 76%, achieved by a Support Vector Machine classifier. We conclude that machine learning classifiers can generate accurate wound care decisions on lower extremity chronic wounds, an important step toward objective, standardized wound care. Higher decision-making accuracy was achieved by leveraging clinical comments from wound experts.

Semantic Segmentation of Smartphone Wound Images: Comparative Analysis of AHRF and CNN-Based Approaches.

Smartphone wound image analysis has recently emerged as a viable way to assess healing progress and provide actionable feedback to patients and caregivers between hospital appointments. Segmentation is a key image analysis step, after which attributes of the wound segment (e.g. wound area and tissue composition) can be analyzed. The Associated Hierarchical Random Field (AHRF) formulates the image segmentation problem as a graph optimization problem. Handcrafted features are extracted, which are then classified using machine learning classifiers. More recently deep learning approaches have emerged and demonstrated superior performance for a wide range of image analysis tasks. FCN, U-Net and DeepLabV3 are Convolutional Neural Networks used for semantic segmentation. While in separate experiments each of these methods have shown promising results, no prior work has comprehensively and systematically compared the approaches on the same large wound image dataset, or more generally compared deep learning vs non-deep learning wound image segmentation approaches. In this paper, we compare the segmentation performance of AHRF and CNN approaches (FCN, U-Net, DeepLabV3) using various metrics including segmentation accuracy (dice score), inference time, amount of training data required and performance on diverse wound sizes and tissue types. Improvements possible using various image pre- and post-processing techniques are also explored. As access to adequate medical images/data is a common constraint, we explore the sensitivity of the approaches to the size of the wound dataset. We found that for small datasets (< 300 images), AHRF is more accurate than U-Net but not as accurate as FCN and DeepLabV3. AHRF is also over 1000x slower. For larger datasets (> 300 images), AHRF saturates quickly, and all CNN approaches (FCN, U-Net and DeepLabV3) are significantly more accurate than AHRF.

Do Novice and Expert Users of Clinical Decision Support Tools Need Different Explanations?

A key requirement for the successful adoption of clinical decision support systems (CDSS) is their ability to provide users with reliable explanations for any given recommendation which can be challenging for some tasks such as wound management decisions. Despite the abundance of decision guidelines, wound non-expert (novice hereafter) clinicians who usually provide most of the treatments still have decision uncertainties. Our goal is to evaluate the use of a Wound CDSS smartphone App that provides explanations for recommendations it produces. The App utilizes wound images taken by the novice clinician using smartphone camera. This study experiments with two proposed variations of rule-tracing explanations called verbose-based and gist-based. Deriving upon theories of decision making, and unlike prior literature that says rule-tracing explanations are only preferred by novices, we hypothesize that, rule-tracing explanations are preferred by both clinicians but in different forms: novices prefer verbose-based rule-tracing and experts prefer gist-based rule-tracing.

Dense motion propagation from sparse samples.

There are many applications for which sparse, or partial sampling of dynamic image data can be used for articulating or estimating motion within the medical imaging area. In this new work, we propose a generalized framework for dense motion propagation from sparse samples which represents an example of transfer learning and manifold alignment, allowing the transfer of knowledge across imaging sources of different domains which exhibit different features. Many such examples exist in medical imaging, from mapping 2D ultrasound or fluoroscopy to 3D or 4D data or monitoring dynamic dose delivery from partial imaging data in radiotherapy. To illustrate this approach we animate, or articulate, a high resolution static MR image with 4D free breathing respiratory motion derived from low resolution sparse planar samples of motion. In this work we demonstrate that sparse sampling of dynamic MRI may be used as a viable approach to successfully build models of free- breathing respiratory motion by constrained articulation. Such models demonstrate high contrast, and high temporal and spatial resolution that have so far been hitherto unavailable with conventional imaging methods. The articulation is based on using a propagation model, in the eigen domain, to estimate complete 4D motion vector fields from sparsely sampled free-breathing dynamic MRI data. We demonstrate that this approach can provide equivalent motion vector fields compared to fully sampled 4D dynamic data, whilst preserving the corresponding high resolution/high contrast inherent in the original static volume. Validation is performed on three 4D MRI datasets using eight extracted slices from a fast 4D acquisition (0.7 s per volume). The estimated deformation fields from sparse sampling are compared to the fully sampled equivalents, resulting in an rms error of the order of 2 mm across the entire image volume. We also present exemplar 4D high contrast, high resolution articulated volunteer datasets using this methodology. This approach facilitates greater freedom in the acquisition of free breathing respiratory motion sequences which may be used to inform motion modelling methods in a range of imaging modalities and demonstrates that sparse sampling of free breathing data may be used within a manifold alignment and transfer learning paradigm to estimate fully sampled motion. The method may also be applied to other examples of sparse sampling to produce dense motion propagation.

Fine-grained diabetic wound depth and granulation tissue amount assessment using bilinear convolutional neural network.

Diabetes mellitus is a serious chronic disease that affects millions of people worldwide. In patients with diabetes, ulcers occur frequently and heal slowly. Grading and staging of diabetic ulcers is the first step of effective treatment and wound depth and granulation tissue amount are two important indicators of wound healing progress. However, wound depths and granulation tissue amount of different severities can visually appear quite similar, making accurate machine learning classification challenging. In this paper, we innovatively adopted the fine-grained classification idea for diabetic wound grading by using a Bilinear CNN (Bi-CNN) architecture to deal with highly similar images of five grades. Wound area extraction, sharpening, resizing and augmentation were used to pre-process images before being input to the Bi-CNN. Innovative modifications of the generic Bi-CNN network architecture are explored to improve its performance. Our research generated a valuable wound dataset. In collaboration with wound experts from University of Massachusetts Medical School, we collected a diabetic wound dataset of 1639 images and annotated them with wound depth and granulation tissue grades as labels for classification. Deep learning experiments were conducted using holdout validation on this diabetic wound dataset. Comparisons with widely used CNN classification architectures demonstrated that our Bi-CNN fine-grained classification approach outperformed prior work for the task of grading diabetic wounds.

Adaptation of the modified Bouc-Wen model to compensate for hysteresis in respiratory motion for the list-mode binning of cardiac SPECT and PET acquisitions: testing using MRI.

PURPOSE: Binning list-mode acquisitions as a function of a surrogate signal related to respiration has been employed to reduce the impact of respiratory motion on image quality in cardiac emission tomography (SPECT and PET). Inherent in amplitude binning is the assumption that there is a monotonic relationship between the amplitude of the surrogate signal and respiratory motion of the heart. This assumption is not valid in the presence of hysteresis when heart motion exhibits a different relationship with the surrogate during inspiration and expiration. The purpose of this study was to investigate the novel approach of using the Bouc-Wen (BW) model to provide a signal accounting for hysteresis when binning list-mode data with the goal of thereby improving motion correction. The study is based on the authors' previous observations that hysteresis between chest and abdomen markers was indicative of hysteresis between abdomen markers and the internal motion of the heart. METHODS: In 19 healthy volunteers, they determined the internal motion of the heart and diaphragm in the superior-inferior direction during free breathing using MRI navigators. A visual tracking system (vts) synchronized with MRI acquisition tracked the anterior-posterior motions of external markers placed on the chest and abdomen. These data were employed to develop and test the Bouc-Wen model by inputting the vts derived chest and abdomen motions into it and using the resulting output signals as surrogates for cardiac motion. The data of the volunteers were divided into training and testing sets. The training set was used to obtain initial values for the model parameters for all of the volunteers in the set, and for set members based on whether they were or were not classified as exhibiting hysteresis using a metric derived from the markers. These initial parameters were then employed with the testing set to estimate output signals. Pearson's linear correlation coefficient between the abdomen, chest, average of chest and abdomen markers, and Bouc-Wen derived signals versus the true internal motion of the heart from MRI was used to judge the signals match to the heart motion. RESULTS: The results show that the Bouc-Wen model generated signals demonstrated strong correlation with the heart motion. This correlation was slightly larger on average than that of the external surrogate signals derived from the abdomen marker, and average of the abdomen and chest markers, but was not statistically significantly different from them. CONCLUSIONS: The results suggest that the proposed model has the potential to be a unified framework for modeling hysteresis in respiratory motion in cardiac perfusion studies and beyond.

Digital anthropomorphic phantoms of non-rigid human respiratory and voluntary body motion for investigating motion correction in emission imaging.

The development of methods for correcting patient motion in emission tomography has been receiving increased attention. Often the performance of these methods is evaluated through simulations using digital anthropomorphic phantoms, such as the commonly used extended cardiac torso (XCAT) phantom, which models both respiratory and cardiac motion based on human studies. However, non-rigid body motion, which is frequently seen in clinical studies, is not present in the standard XCAT phantom. In addition, respiratory motion in the standard phantom is limited to a single generic trend. In this work, to obtain a more realistic representation of motion, we developed a series of individual-specific XCAT phantoms, modeling non-rigid respiratory and non-rigid body motions derived from the magnetic resonance imaging (MRI) acquisitions of volunteers. Acquisitions were performed in the sagittal orientation using the Navigator methodology. Baseline (no motion) acquisitions at end-expiration were obtained at the beginning of each imaging session for each volunteer. For the body motion studies, MRI was again acquired only at end-expiration for five body motion poses (shoulder stretch, shoulder twist, lateral bend, side roll, and axial slide). For the respiratory motion studies, an MRI was acquired during free/regular breathing. The magnetic resonance slices were then retrospectively sorted into 14 amplitude-binned respiratory states, end-expiration, end-inspiration, six intermediary states during inspiration, and six during expiration using the recorded Navigator signal. XCAT phantoms were then generated based on these MRI data by interactive alignment of the organ contours of the XCAT with the MRI slices using a graphical user interface. Thus far we have created five body motion and five respiratory motion XCAT phantoms from the MRI acquisitions of six healthy volunteers (three males and three females). Non-rigid motion exhibited by the volunteers was reflected in both respiratory and body motion phantoms with a varying extent and character for each individual. In addition to these phantoms, we recorded the position of markers placed on the chest of the volunteers for the body motion studies, which could be used as external motion measurement. Using these phantoms and external motion data, investigators will be able to test their motion correction approaches for realistic motion obtained from different individuals. The non-uniform rational B-spline data and the parameter files for these phantoms are freely available for downloading and can be used with the XCAT license.

MRI Investigation of the Linkage Between Respiratory Motion of the Heart and Markers on Patient's Abdomen and Chest: Implications for Respiratory Amplitude Binning List-Mode PET and SPECT Studies.

Respiratory motion of the heart impacts the diagnostic accuracy of myocardial-perfusion emission-imaging studies. Amplitude binning has come to be the method of choice for binning list-mode based acquisitions for correction of respiratory motion in PET and SPECT. In some subjects respiratory motion exhibits hysteretic behavior similar to damped non-linear cyclic systems. The detection and correction of hysteresis between the signals from surface movement of the patient's body used in binning and the motion of the heart within the chest remains an open area for investigation. This study reports our investigation in nine volunteers of the combined MRI tracking of the internal respiratory motion of the heart using Navigators with stereo-tracking of markers on the volunteer's chest and abdomen by a visual-tracking system (VTS). The respiratory motion signals from the internal organs and the external markers were evaluated for hysteretic behavior analyzing the temporal correspondence of the signals. In general, a strong, positive correlation between the external marker motion (AP direction) and the internal heart motion (SI direction) during respiration was observed. The average ± standard deviation in the Spearman's ranked correlation coefficient (ρ) over the nine volunteer studied was 0.92 ± 0.1 between the external abdomen marker and the internal heart, and 0.87 ± 0.2 between the external chest marker and the internal heart. However despite the good correlation on average for the nine volunteers, in three studies a poor correlation was observed due to hysteretic behavior between inspiration and expiration for either the chest marker and the internal motion of the heart, or the abdominal marker and the motion of the heart. In all cases we observed a good correlation of at least either the abdomen or the chest with the heart. Based on this result, we propose the use of marker motion from both the chest and abdomen regions when estimating the internal heart motion to detect and address hysteresis when binning list-mode emission data.