Upload your resume. PhdFit's six research agents compare your background with faculty profiles, recent publications, lab focus, and outreach opportunities, then rank professors with evidence you can review.
Ask how her lab is extending interpretability methods into fairness audits for real-world AI systems.
University of Southern California · Management and Organization
Active 2013–2025
Shaun Harper is a University Professor of Business, Education, and Public Policy at USC Marshall School of Business, where he also holds the Clifford and Betty Allen Chair in Urban Leadership. He is the founder and Chief Research Scientist of the USC Race and Equity Center. Harper teaches MBA students and advises companies on diversity, equity, and inclusion (DEI). His extensive work includes collaboration with over 400 businesses, agencies, and institutions, and he has published 13 books and more than 100 academic papers. Harper has secured significant funding, including $26 million in foundation grants, $19 million in contracts, and $6.5 million in investments and philanthropic gifts for his center. His research has been widely cited, appearing in over 27,000 published studies, amicus briefs to the U.S. Supreme Court, and on the floor of the U.S. House of Representatives. His insights have been featured in major media outlets such as The New York Times, Wall Street Journal, Washington Post, and others, and he has been interviewed on prominent networks including CNN, MSNBC, CNBC, ESPN, PBS NewsHour, and NPR. Harper's work has influenced a broad audience, with over 3 million readers of his articles in outlets like the Washington Post, Forbes, Rolling Stone, CNN.com, Fortune, and the Los Angeles Times. He has collaborated with notable organizations such as the U.S. Air Force, Nike, Google, Microsoft, Major League Baseball, the NFL, and the City of Los Angeles. Prior to his tenure at USC, Harper was a tenured full professor at the University of Pennsylvania. He has received numerous awards, six honorary degrees, and recognition as a top diversity, equity, and inclusion visionary by the Los Angeles Times in 2023 and 2024, as well as being named one of the top 100 Hollywood entertainment industry business leaders in 2025. In 2023, the Los Angeles Business Journal awarded his center for its impact on DEI initiatives globally and locally.
Agentic AI and Ethics in Telemedicine
Telehealth and Medicine Today · 2025-06-30 · 1 citations
Topics that will be discussed are: What does "agentic AI" mean in a healthcare context, and how does it differ from traditional decision-support tools? What ethical tensions arise when AI agents begin to make or recommend autonomous decisions in patient care? What safeguards should be in place to prevent algorithmic bias or patient harm in remote care settings—especially when patients and providers are not co-located? When agentic AI goes wrong, who’s responsible, and how do we ensure accountability in cross-border virtual care? How can health systems build trust in agentic AI when the algorithms themselves may not be fully explainable—even by their creators? How are organizations operationalizing AI ethics beyond principles and into practice—especially when the tech moves faster than policy? Speakers Shanil Ebrahim is a Partner and the National Life Sciences and Healthcare Consulting Leader at Deloitte Canada. He advises clients across healthcare, pharmaceuticals, and retail pharmacy on complex strategy transformations, with a focus on AI and data-driven innovation. Shanil also leads Deloitte Canada's cross-industry AI strategy, helping organizations harness AI to drive growth, personalization, and operational excellence through tech-enabled, citizen and customer-centric solutions. Dr. Dimitrios Kalogeropoulos is a digital health pioneer and a committed advocate for the ethical and responsible use of AI in healthcare. He serves as CEO of the Global Health & Digital Innovation Foundation and as Health Executive in Residence at UCL’s Global Business School for Health. Dr. Kalogeropoulos has advised leading organizations, including the WHO, and has played a key role in shaping global policy initiatives to improve healthcare accessibility and drive sustainable, innovative solutions worldwide. Moderator Sarah Harper, MA, MBA drives digital transformation across the healthcare ecosystem and beyond. With deep expertise engaging with learners of all ages and levels, she blends clinical insight, systems thinking, and user-centered design to turn bold ideas into practical, equitable solutions. Dedicated to helping others, Sarah makes digital care smarter—and more human. Sarah leads AI, Analytics, and Automation initiatives at Mayo Clinic Health System, and holds the academic rank of Assistant Professor of Healthcare Administration. She's the co-host of Tech It to the Limit, a podcast blending wit and wisdom to explore digital health’s messiest challenges. Sarah also serves as an Advisor to Mayo Clinic Platform, supporting solution developers and health systems in tech implementation and evaluation.
Louis Goldstein
Terri L. Scott
Northeastern University
Laura Gwilliams
Stanford University
Deborah F. Levy
Neurological Surgery
Ilina Bhaya-Grossman
University of California, San Francisco
Doctor of Philosophy, Linguistics
University of Southern California
Master of Arts, Lingusitics
University of Southern California
Bachelor of Arts, Linguistics/Spanish
University of Michigan
PhdFit ranks faculty by your research interests, methods, and publications — grounded in their actual work, not templates.
Artificial Intelligence in Telemedicine
Telehealth and Medicine Today · 2025-05-28
Telehealth and Medicine Today (THMT) editors and members of the International Society for Telemedicine and eHealt (ISfTeH) for this face-paced conversation. Hear frontline experts discuss real world experiences and opinions about AI, improving patient care, and operational efficiency within the healthcare sector. Topics that will be discussed the conversation will include are: How to move beyond pilot purgatory to focus on what really matters - outcomes. Drawing on real-world examples from community-based healthcare Practical strategies for defining success, embedding remeasurement cycles, and using AI and automation to advance equity and efficiency How AI is reducing documentation burden and surfacing timely clinical insight to support virtual and remote care nurses Why co-developing with nurses is critical, and how to build safe, trusted AI tools that fit the realities of telehealth and remote clinical environments. How AI agents can manage remote follow-ups, symptom monitoring, and patient education Virtual Brain Health
Computational Architecture of Speech Comprehension in the Human Brain
Annual Review of Linguistics · 2024-10-16 · 6 citations
Understanding the computational algorithm that gives rise to human language is a shared endeavor among neuroscience, linguistics, and machine learning. We propose a conceptual framework for making measurable progress toward this goal by studying the subcomponents of the processing system: its underlying representations, operations, and information flow. We review evidence from neurophysiology, neuropsychology, linguistic theory, and computational modeling and suggest future directions to push the field forward in developing a precise characterization of spoken language understanding. Overall, we claim that representations of speech properties, and the operations that generate and manipulate those representations, exist within a highly parallel, highly redundant spatiotemporal regime.
Intensity downtrends in Embosi intonation
2024-05-13
Translanguaging and teacher authority
Language and Literacy · 2024 · 2 citations
The purpose of the paper is to understand the connection between teacher authority and children’s language use. The data presented here was pulled from two large data sources: a set of case-studies in Grades 4 to 6 classrooms with multilingual children who were new to Canada and learning to read and write for the first time and qualitative research in a teacher education program preparing teacher candidates to educate multilingual students. Findings suggest that children translanguage in liminal spaces outside of the teachers’ authority and that multilingual students that were asked to translanguage in English authority classrooms had negative experiences.
Acoustic properties of utterance-final prosodic events in Embosi
The Journal of the Acoustical Society of America · 2024-10-01
In some Bantu languages, the utterance-final prosodic juncture has been found to exhibit multiple acoustic properties, like a low f0 and final devoicing (Rialland & Aborobongui, 2016; Zerbian, 2016). However, a detailed quantitative analysis of these acoustic properties is still lacking, and the underlying causes are not well understood. To this end, we carry out an acoustic analysis of the utterance-final prosodic events in Embosi, a Bantu language spoken in the Republic of Congo. We analyze the mean f0, intensity and H1-H2* of the final three moras in an Embosi utterance. We hypothesize that the acoustic properties are correlates of the pulmonic pressure initiation dynamics, i.e., when subglottal pressure drops below the phonation pressure threshold towards the end of an utterance. Thus, f0 and intensity should be lowered and final-devoicing can also occur (Zhang, 2016). It is also likely that a glottal spreading or slackening gesture is present utterance-finally, reducing the vocal fold medial surface thickness. Then, the voice quality might become breathier, which can be characterized by a low f0 and an increase in H1-H2* (Zhang, 2016). The results show that utterance-finally both intensity and f0 are lowered but there is no evidence for an increase in H1-H2*. These findings are more consistent with the pulmonic pressure initiation hypothesis.
A Dynamical Systems Approach to F0 Hard Landing in Embosi Declarative Intonation
SSRN Electronic Journal · 2023-01-01
Segmental effects on the transmission of variability between articulation and acoustics
The Journal of the Acoustical Society of America · 2021-10-01
Previous research on articulatory-acoustic relations has generated conflicting results regarding the extent to which variability in articulation can be recovered from the acoustic signal (e.g., Guenther et al., 1999; Whalen etal., 2018). This study extends previous research probing the relationship between articulatory and acoustic variability by examining the extent to which individual differences in articulatory variability are recoverable from acoustics, and vice versa, for a set of American English consonants. A series of linear mixed effects regression models were fit to articulatory and acoustic data from tokens of /s/, /ʃ/, /l/, and /ɹ/ produced by 40 speakers in the Wisconsin X-Ray Microbeam Corpus (Westbury, 1994) to evaluate the extent to which the variability exhibited in one of these physical dimensions was recoverable from the signal in the other domain. The output of these statistical models was then compared to measurements of individual speakers’ variability along specific articulatory and acoustic dimensions to evaluate whether individual differences in variability were conveyed between articulation and acoustics. The results of the analyses suggest that the variability exhibited in one physical domain is to some extent recoverable from the other, although this capacity varies across segments and articulatory/acoustic dimensions. [Work supported by NIH.]
Tenderness Between Strangers: Intimate exchanges on banlieue wastelands
Performance paradigm · 2021-10-09
This case-study is of a socially-engaged performance project, Tendresse Radicale (2019 – 2021) in Villetaneuse, a multi-cultural suburb of Paris, where kindness does not tend to be visible, public, or demonstrated. Through gifts and unsolicited gestures the project introduces a fragile, bumbling presence into an urban wilderness perceived locally as threatening. Seeking and enjoying reciprocity, the project observes the degrees of refusal and awkwardness that unsolicited acts of kindness towards strangers provokes. Through rendering tangible an inherently invisible relational practice, the article argues the importance and potential radicality of such gestures to open up genuine spaces of encounter.
Figshare · 2021-01-01
Real-time magnetic resonance imaging (RT-MRI) of human speech production is enabling significant advances in speech science, linguistics, bio-inspired speech technology development, and clinical applications. Easy access to RT-MRI is however limited, and comprehensive datasets with broad access are needed to catalyze research across numerous domains. The imaging of the rapidly moving articulators and dynamic airway shaping during speech demands high spatio-temporal resolution and robust reconstruction methods. Further, while reconstructed images have been published, to-date there is no open dataset providing raw multi-coil RT-MRI data from an optimized speech production experimental setup. Such datasets could enable new and improved methods for dynamic image reconstruction, artifact correction, feature extraction, and direct extraction of linguistically-relevant biomarkers.<br>The present dataset offers a unique corpus of 2D sagittal-view RT-MRI videos along with synchronized audio for 75 subjects performing linguistically motivated speech tasks, alongside the corresponding first-ever public domain raw RT-MRI data. The dataset also includes 3D volumetric vocal tract MRI during sustained speech sounds and high-resolution static anatomical T2-weighted upper airway MRI for each subject.<br><br>Included are the following files:<i>dataset.zip</i>: contains the entire dataset of 75 subjects<i>dataset_2drt_video_only.zip</i>: contains only files of 2D RT-MRI videos of 75 subjects<i>dataset_t2w_only.zip</i>: contains only files of T2-weighted images of 75 subjects<i>dataset_3d_only.zip</i>: contains only files of 3D volumetric images of 75 subjects <br><i>example_for_sub001.zip</i>: contains all files of 1 subject (sub001)<br><i>metafile_public_20210129.json</i>: contains meta information<i>Subjects.xlsx</i>: contains demographic information for each subject<i>Stimuli.pptx</i>: contains the experimental stimuli including scripts and pictures used for the visualization to the subjects<br>
Yizhen Zhang
University of California, San Francisco