Through vertex component analysis of hyperspectral imaging data in the visible spectral range, we differentiated erythematous and pigmented areas in patients with cutaneous chronic graft-versus-host disease. We explored the feasibility of hyperspectral imaging in combination with unsupervised learning algorithms to differentiate active disease from inactive post-inflammatory skin changes, a fundamental practice gap in caring for these patients. We compared erythema and pigment maps to the visual assessment by a dermatologist as the ground truth.
Through noninvasive monitoring of leukocyte motion in skin capillaries of patients after hematopoietic cell transplantation, we found increased leukocyte rolling and adhesion prior to clinical signs of disease. In this longitudinal pilot study, we explored the feasibility to detect changes in leukocyte-endothelial interactions that precede acute graft-versus-host disease in patients after hematopoietic cell transplantation. We present the pattern of change in leukocyte rolling and adhesion in three patients over the course of the first 100 days post-transplant. Our preliminary data show increased leukocyte-endothelial interactions prior to clinical signs of any organ acute graft-versus-host disease.
Inflammatory tissue response is one of the first and most common manifestations of acute graft-versus-host disease (aGVHD), a potentially deadly immune-mediated disease that occurs in 30-60% of patients after stem cell transplantation. A fundamental challenge in developing effective treatment strategies for aGVHD is the lack of tools to study disease biology in real-time in post-transplant patients. The inflammatory tissue response causes increased expression of specialized endothelial proteins on vessel walls making leukocytes to roll, adhere and eventually extravasate into the tissue at a higher rate than in normal conditions. Although the importance of leukocyte-endothelial interactions to detect and track inflammation has been well shown in murine models, there are no published clinical studies in humans. In this study, we explore the feasibility to detect presence of aGVHD in post-transplant patients through the imaging of in vivo leukocyte motion. We used a clinical confocal microscope (Vivascope 1500) to acquire videos of 5 aGVHD patients and 5 controls (no aGVHD) within 50±30 days post-transplant. The microscope is capable of real-time imaging of individual cells in the postcapillary vessels at 9 frames per second. Through video analysis, we extracted five quantitative parameters: number and velocity of rolling leukocytes, number of adherent leukocytes (stationary >30 s), blood flow velocity, and number of vessels. In a limited number of subjects, we show that parameters characteristic of the dynamic motion in skin capillaries can be observed noninvasively in post-transplant patients. Further studies are needed to test the diagnostic potential of these parameters.
Chronic graft-versus-host disease (cGVHD) is a frequent and potentially life-threatening complication of allogeneic hematopoietic stem cell transplantation (HCT) and commonly affects the skin, resulting in distressing patient morbidity. The percentage of involved body surface area (BSA) is commonly used for diagnosing and scoring the severity of cGVHD. However, the segmentation of the involved BSA from patient whole body serial photography is challenging because (1) it is difficult to design traditional segmentation method that rely on hand crafted features as the appearance of cGVHD lesions can be drastically different from patient to patient; (2) to the best of our knowledge, currently there is no publicavailable labelled image set of cGVHD skin for training deep networks to segment the involved BSA. In this preliminary study we create a small labelled image set of skin cGVHD, and we explore the possibility to use a fully convolutional neural network (FCN) to segment the skin lesion in the images. We use a commercial stereoscopic Vectra H1 camera (Canfield Scientific) to acquire ~400 3D photographs of 17 cGVHD patients aged between 22 and 72. A rotational data augmentation process is then applied, which rotates the 3D photos through 10 predefined angles, producing one 2D projection image at each position. This results in ~4000 2D images that constitute our cGVHD image set. A FCN model is trained and tested using our images. We show that our method achieves encouraging results for segmenting cGVHD skin lesion in photographic images.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.