Publications

@article{casas_holojig_2025,

title = {HoloJig: Interactive Spoken Prompt Specified Generative AI Environments},

author = {L. Casas and S. Hannah and K. Mitchell},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105001182100&doi=10.1109%2FMCG.2025.3553780&partnerID=40&md5=9fafa25e4b6ddc9d2fe32d813fbabb20},

doi = {10.1109/MCG.2025.3553780},

issn = {02721716 (ISSN); 15581756 (ISSN)},

year  = {2025},

date = {2025-01-01},

journal = {IEEE Computer Graphics and Applications},

volume = {45},

number = {2},

pages = {69–77},

abstract = {HoloJig offers an interactive, speech-to-virtual reality (VR), VR experience that generates diverse environments in real time based on live spoken descriptions. Unlike traditional VR systems that rely on prebuilt assets, HoloJig dynamically creates personalized and immersive virtual spaces with depth-based parallax 3-D rendering, allowing users to define the characteristics of their immersive environment through verbal prompts. This generative approach opens up new possibilities for interactive experiences, including simulations, training, collaborative workspaces, and entertainment. In addition to speech-to-VR environment generation, a key innovation of HoloJig is its progressive visual transition mechanism, which smoothly dissolves between previously generated and newly requested environments, mitigating the delay caused by neural computations. This feature ensures a seamless and continuous user experience, even as new scenes are being rendered on remote servers. © 2025 Elsevier B.V., All rights reserved.},

note = {Publisher: IEEE Computer Society},

keywords = {3-D rendering, Article, Collaborative workspace, customer experience, Economic and social effects, generative artificial intelligence, human, Immersive, Immersive environment, parallax, Real- time, simulation, Simulation training, speech, Time based, Virtual environments, Virtual Reality, Virtual reality experiences, Virtual spaces, VR systems},

pubstate = {published},

tppubtype = {article}

}

@article{almashmoum_understanding_2025,

title = {Understanding the Views of Health Care Professionals on the Usability and Utility of Virtual Reality Multidisciplinary Team Meetings: Usability and Utility Study},

author = {M. Almashmoum and A. Payton and E. Johnstone and J. Cunningham and J. Ainsworth},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105012020235&doi=10.2196%2F60651&partnerID=40&md5=55e0ddff7b6b7dbde1ea4b1191f88d96},

doi = {10.2196/60651},

issn = {28183045 (ISSN)},

year  = {2025},

date = {2025-01-01},

journal = {JMIR XR and Spatial Computing},

volume = {2},

abstract = {Background: Multidisciplinary team (MDT) meetings are one of the facilitators that enhance knowledge sharing among health care professionals. However, organizing a face-to-face MDT meeting to discuss patient treatment plans can be time-con-suming. Virtual reality software is widely used in health care nowadays to save time and protect lives. Therefore, the use of virtual reality multidisciplinary team (VRMDT) meeting software may help enhance knowledge sharing between health care professionals and make meetings more efficient. Objective: The objectives of this study were to introduce VRMDT software for enhancing knowledge sharing and to evaluate the feasibility and usability of the VRMDT for use by professionals in health care institutions. Methods: We invited participants from The University of Manchester Faculty for Biology, Medicine, and Health who had a health care background. As this was the first stage of software development, individuals who did not usually attend MDT meetings were also invited via email to participate in this study. Participants evaluated VRMDT using a Meta Quest 3 headset, and software developed using the Unity platform. The software contained an onboarding tutorial that taught the participants how to select items, load and rotate 3D Digital Imaging and Communications in Medicine files, talk to a generative artificial intelligence–supported avatar, and make notes. After the evaluation (approximately 15 min), participants received an electronic survey using the Qualtrics survey tool (Qualtrics International Inc) to score the usability and feasibility of the software by responding to the 10-item system usability scale, and 12-point heuristic evaluation questions with Neilsen severity rating. Results: A total of 12 participants, including 4 health informatics, 3 with a nursing background, 2 medical doctors, 1 radiologist, and 2 biostatisticians, participated in the study. The most common age bracket of participants was 20‐30 years (6/12, 50%). Most of the respondents had no experience with virtual reality, either in educational or entertainment settings. The VRMDT received a mean usability score of 72.7 (range between 68 and 80.3), earning an overall “good” rating grade. The mean score of single items in the heuristic evaluation questionnaires was less than 1 out of 4 (the overall mean was 0.6), which indicates that only minor problems were encountered when using this software. Overall, the participant’s feedback was good with highlighted issues including a poor internet connection and the quality of the generative artificial intelligence response. Conclusions: VRMDT software (developed by SentiraXR) was developed with several functions aimed at helping health care professionals to discuss medical conditions efficiently. Participants found that the VRMDT is a powerful, and useful tool for enhancing knowledge sharing among professionals who are involved in MDT meetings due to its functionality and multiuser interactive environments. Additionally, there may be the possibility of using it to train junior professionals to interpret medical reports. © 2025 Elsevier B.V., All rights reserved.},

note = {Publisher: JMIR Publications Inc.},

keywords = {Artificial intelligence, digital environments, heuristic evaluation, knowledge sharing, multidisciplinary team meetings, simulation, Usability, Virtual environments, Virtual Reality, VR},

pubstate = {published},

tppubtype = {article}

}

@article{salinas_-silico_2025,

title = {In-silico CT simulations of deep learning generated heterogeneous phantoms},

author = {C. S. Salinas and K. Magudia and A. Sangal and L. Ren and W. P. Segars},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105010297226&doi=10.1088%2F2057-1976%2Fade9c9&partnerID=40&md5=47f211fd93f80e407dcd7e4c490976c2},

doi = {10.1088/2057-1976/ade9c9},

issn = {20571976 (ISSN)},

year  = {2025},

date = {2025-01-01},

journal = {Biomedical Physics and Engineering Express},

volume = {11},

number = {4},

abstract = {Current virtual imaging phantoms primarily emphasize geometric accuracy of anatomical structures. However, to enhance realism, it is also important to incorporate intra-organ detail. Because biological tissues are heterogeneous in composition, virtual phantoms should reflect this by including realistic intra-organ texture and material variation. We propose training two 3D Double U-Net conditional generative adversarial networks (3D DUC-GAN) to generate sixteen unique textures that encompass organs found within the torso. The model was trained on 378 CT image-segmentation pairs taken from a publicly available dataset with 18 additional pairs reserved for testing. Textured phantoms were generated and imaged using DukeSim, a virtual CT simulation platform. Results showed that the deep learning model was able to synthesize realistic heterogeneous phantoms from a set of homogeneous phantoms. These phantoms were compared with original CT scans and had a mean absolute difference of 46.15 ± 1.06 HU. The structural similarity index (SSIM) and peak signal-to-noise ratio (PSNR) were 0.86 ± 0.004 and 28.62 ± 0.14, respectively. The maximum mean discrepancy between the generated and actual distribution was 0.0016. These metrics marked an improvement of 27%, 5.9%, 6.2%, and 28% respectively, compared to current homogeneous texture methods. The generated phantoms that underwent a virtual CT scan had a closer visual resemblance to the true CT scan compared to the previous method. The resulting heterogeneous phantoms offer a significant step toward more realistic in silico trials, enabling enhanced simulation of imaging procedures with greater fidelity to true anatomical variation. © 2025 Elsevier B.V., All rights reserved.},

note = {Publisher: Institute of Physics},

keywords = {adult, algorithm, Algorithms, anatomical concepts, anatomical location, anatomical variation, Article, Biological organs, bladder, Bone, bone marrow, CGAN, colon, comparative study, computer assisted tomography, Computer graphics, computer model, Computer Simulation, Computer-Assisted, Computerized tomography, CT organ texture, CT organ textures, CT scanners, CT synthesis, CT-scan, Deep learning, fluorodeoxyglucose f 18, Generative Adversarial Network, Generative AI, histogram, human, human tissue, Humans, III-V semiconductors, image analysis, Image processing, Image segmentation, Image texture, Imaging, imaging phantom, intra-abdominal fat, kidney blood vessel, Learning systems, liver, lung, major clinical study, male, mean absolute error, Medical Imaging, neoplasm, Phantoms, procedures, prostate muscle, radiological parameters, signal noise ratio, Signal to noise ratio, Signal-To-Noise Ratio, simulation, Simulation platform, small intestine, Statistical tests, stomach, structural similarity index, subcutaneous fat, Textures, three dimensional double u net conditional generative adversarial network, Three-Dimensional, three-dimensional imaging, Tomography, Virtual CT scanner, Virtual Reality, Virtual trial, virtual trials, whole body CT, X-Ray Computed, x-ray computed tomography},

pubstate = {published},

tppubtype = {article}

}

@article{stuart_developing_2024,

title = {Developing augmented reality filters to display visual cues on diverse skin tones},

author = {J. Stuart and A. Stephen and K. Aul and M. D. Bumbach and S. Huffman and B. Russo and B. Lok},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85198640398&doi=10.3389%2Ffrvir.2024.1363193&partnerID=40&md5=800d1a5cb574fa6297235cb1bbd35bb1},

doi = {10.3389/frvir.2024.1363193},

issn = {26734192 (ISSN)},

year  = {2024},

date = {2024-01-01},

journal = {Frontiers in Virtual Reality},

volume = {5},

abstract = {Introduction: Variations in skin tone can significantly alter the appearance of symptoms such as rashes or bruises. Unfortunately, previous works utilizing Augmented Reality (AR) in simulating visual symptoms have often failed to consider this critical aspect, potentially leading to inadequate training and education. This study seeks to address this gap by integrating generative artificial intelligence (AI) into the AR filter design process. Methods: We conducted a 2 × 5 within-subjects study with second-year nursing students (N = 117) from the University of Florida. The study manipulated two factors: symptom generation style and skin tone. Symptom generation style was manipulated using a filter based on a real symptom image or a filter based on a computer-generated symptom image. Skin tone variations were created by applying AR filters to computer-generated images of faces with five skin tones ranging from light to dark. To control for factors like lighting or 3D tracking, 101 pre-generated images were created for each condition, representing a range of filter transparency levels (0–100). Participants used visual analog scales on a computer screen to adjust the symptom transparency in the images until they observed image changes and distinct symptom patterns. Participants also rated the realism of each condition and provided feedback on how the symptom style and skin tone impacted their perceptions. Results: Students rated the symptoms displayed by the computer-generated AR filters as marginally more realistic than those displayed by the real image AR filters. However, students identified symptoms earlier with the real-image filters. Additionally, SET-M and Theory of Planned Behavior questions indicate that the activity increased students’ feelings of confidence and self-efficacy. Finally, we found that similar to the real world, where symptoms on dark skin tones are identified at later stages of development, students identified symptoms at later stages as skin tone darkened regardless of cue type. Conclusion: This work implemented a novel approach to develop AR filters that display time-based visual cues on diverse skin tones. Additionally, this work provides evidence-based recommendations on how and when generative AI-based AR filters can be effectively used in healthcare education. © 2024 Elsevier B.V., All rights reserved.},

note = {Publisher: Frontiers Media SA},

keywords = {Augmented Reality, fidelity, Healthcare, realism, simulation, symptoms, visual cue training},

pubstate = {published},

tppubtype = {article}

}

@article{donekal_chandrashekar_understanding_2024,

title = {Understanding User Behavior for Enhancing Cybersecurity Training with Immersive Gamified Platforms},

author = {N. Donekal Chandrashekar and A. Lee and M. Azab and D. Grǎcanin},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85213435167&doi=10.3390%2Finfo15120814&partnerID=40&md5=9e015d2d280ef872e30ad46db0781ce8},

doi = {10.3390/info15120814},

issn = {20782489 (ISSN)},

year  = {2024},

date = {2024-01-01},

journal = {Information (Switzerland)},

volume = {15},

number = {12},

abstract = {In modern digital infrastructure, cyber systems are foundational, making resilience against sophisticated attacks essential. Traditional cybersecurity defenses primarily address technical vulnerabilities; however, the human element, particularly decision-making during cyber attacks, adds complexities that current behavioral studies fail to capture adequately. Existing approaches, including theoretical models, game theory, and simulators, rely on retrospective data and static scenarios. These methods often miss the real-time, context-specific nature of user responses during cyber threats. To address these limitations, this work introduces a framework that combines Extended Reality (XR) and Generative Artificial Intelligence (Gen-AI) within a gamified platform. This framework enables continuous, high-fidelity data collection on user behavior in dynamic attack scenarios. It includes three core modules: the Player Behavior Module (PBM), Gamification Module (GM), and Simulation Module (SM). Together, these modules create an immersive, responsive environment for studying user interactions. A case study in a simulated critical infrastructure environment demonstrates the framework’s effectiveness in capturing realistic user behaviors under cyber attack, with potential applications for improving response strategies and resilience across critical sectors. This work lays the foundation for adaptive cybersecurity training and user-centered development across critical infrastructure. © 2024 Elsevier B.V., All rights reserved.},

note = {Publisher: Multidisciplinary Digital Publishing Institute (MDPI)},

keywords = {Artificial intelligence, Critical infrastructures, Cyber attacks, Cyber security, Cyber systems, Cyber-attacks, Cybersecurity, Decisions makings, Digital infrastructures, digital twin, Extended reality, Gamification, Immersive, Network Security, simulation, Technical vulnerabilities, Training, user behavior, User behaviors},

pubstate = {published},

tppubtype = {article}

}