Rejection sensitivity (RS) is a core component of social anxiety that may impact symptom severity and treatment outcomes. There may be substantial differences in both neural and behavioral responses to fear as a function of trait RS in socially anxious individuals, as reflected in altered regional brain activation patterns underlying fear and self-referential processing. Individuals with the highest levels of RS may struggle to extinguish learned social fears but it is unclear if they learn safety differently by watching another person doing so, a process called vicarious extinction learning. Recent work has highlighted the advantage of vicarious extinction learning in fear regulation over traditional extinction learning, but this has not yet been tested in socially anxious populations. Given the relevance of social learning processes in social anxiety, vicarious extinction learning may enhance the effectiveness of exposure-based therapy in this population. In this study, individuals will undergo a social fear task in which they learn to pair images of angry faces with a mild electric shock. During Phase II, participants will watch a video of another person undergoing the same task and learning safety from one of the photos previously associated with a shock. The final phase will test if participants can apply the model's safety learning to themselves. I hypothesize that high trait RS may moderate vicarious extinction learning, as seen by differential activation in the ventromedial prefrontal cortex (vmPFC). It is possible that those with higher RS might show increased vmPFC activation during extinction and might benefit more from vicarious safety learning because social signals are more salient in this population and they have more to gain from learning safety, but it is also possible that they may not benefit, or even worsen, if they are unable to empathize with the learning model and remain vigilant to cues of social rejection, as evident by decreased vmPFC activation during extinction. Regardless of the direction of the effect, trait RS may be an important variable to consider when treating socially anxious patients using extinction-based principles and may influence the development of novel interventions to target this population. 

Juvenile Idiopathic Arthritis (JIA) is the most common rheumatic disease in children. It causes joint swelling and stiffness and can last for months to years. Current methods to screen for JIA include Magnetic Resonance Imaging (MRI) and Musculoskeletal Ultrasound, both of which are often time-consuming and expensive. Using our developed thermal imaging algorithm, the ability to screen for JIA would become more accessible, affordable, and less time-consuming to kids and their families. The goal of our study was to determine if using a smartphone-attached thermal camera was reliable for image detection of arthritis in children. We also wanted to test the effect that physical activities, such as walking, would have on lower extremity temperature data within our imaging algorithm. Using an industrial-grade thermal imaging camera, we took thermal images of the lower extremities from the anterior, posterior, medial, and lateral views. We also repeated this using the smartphone-based thermal imaging camera. The temperature data was extracted from the thermal images and analyzed for temperature fluctuations in the regions of interest. Even though the smartphone-attached camera had lower resolution and less precision than the industrial-grade camera, both performed well in their sensitivity and specificity to detect the inflamed joints compared to the common standard of doing a physical joint exam. In addition, the cohort which performed physical activity demonstrated significant temperature changes which were consistent over time and did not return to pre-activity levels. The results of this study show potential for faster and more accessible JIA imaging platforms in the future.

Chronic non-bacterial osteomyelitis (CNO), also known in its severist form as Chronic Recurrent Multifocal Osteomyelitis (CRMO), is a rare, auto-inflammatory disease with no present cure. The disease involves the chronic inflammation of normal, healthy bone without the presence of infection. Currently, no medications have been approved by the US Food and Drug Administration specifically for CNO. Consequently, many different types of medications, including disease modifying anti-rheumatic drugs (DMARDs) and tumor necrosis factor (TNF) inhibitors, and Bisphosphonates are being prescribed off label. However, patients can have significant side effects after taking these medications and consistent reports on the prevalence of these serious adverse events (SAE) among CNO patients are lacking. We will be examining instances of COVID-19 infection, hospitalizations, and psoriasis while taking CNO medication. We aim to investigate the association between taking various medications with the prevalence of SAEs among patients under 21 years old. For our research, we are drawing information from one of the largest CNO clinical research databases, Seattle Children’s Hospital’s database from January 2014 - present day. We hypothesize that all medications will be well tolerated by CNO patients under 21 years old. Through self-reported patient data and physician examination, information on SAE prevalence was collected. The patient population includes 351 patients treated with DMARDs, 294 treated with TNFs, and 89 treated with Bisphosphonates. General statistical methods were used to summarize the data and determine correlation; descriptive statistics was used to report the incidence rate per 100 patient years.As there is minimal knowledge about effective treatments for CNO, we expect that the results of this study will shed light on the reliability of various medications, improving patient disease management and possibly lead to a cure.

Juvenile idiopathic arthritis (JIA) is the most common rheumatic disease in children and frequently presents in knees, followed by ankles, wrists and elbows. JIA is typically evaluated by pediatric rheumatologists using joint exams. However, musculoskeletal ultrasound or MRI with contrast can be used for greater sensitivity but are more expensive and require extensive operator training. In contrast, infrared thermal imaging is a noninvasive tool that is quick, economical, and precise in detecting temperatures of different body parts. Thermal cameras are becoming increasingly accessible and can be used with smartphones, like the FLIR ONE Pro camera by Teledyne FLIR. Recent efforts with temperature after within limb calibration (TAWiC) algorithms have made use of thermal imaging and demonstrated promising results in detecting arthritis in knees and ankles. However, the current TAWiC algorithm implementation has limited scalability due to its dependency on trained technicians to identify key anatomical points. We leveraged existing computer vision libraries like OpenCV and human pose estimation models like OpenPose to automate the TAWiC algorithm. We designed a pipeline that first entails extracting thermal and visible images from a radiometric JPEG generated by a FLIR ONE Pro. The extracted images are subsequently co-registered. Key anatomical points are then labeled in the visible image using OpenPose and contours drawn around the regions of interest. The contours coupled with the labeled key points are used to segment and retrieve the temperatures of specific anatomical regions. We collected preliminary data that demonstrated the feasibility of this workflow. We anticipate that similar TAWiC measurements will be generated when using the automated approach to re-analyze participant thermal images. Automating this algorithm will increase the scalability of this approach and allows for extending this algorithm to other joints.