New York University Research Articles

Metainferentially substructural validity theories

Porter (2023) [Three essays on substructural approaches to semantic paradoxes [PhD thesis]. City University of New York, New York, USA. https://academicworks.cuny.edu/gc_etds/5144/] has presented higher-level metainferential versions of the validity Curry argument introduced by Beall and Murzi (2013) [Two flavours of curry paradox. Journal of Philosophy, 110(3), 143–65]. What is peculiar of these paradoxes is that no operational rule nor any structural rule play any role in the derivation of the unpleasant result. Thus, to block them and have a uniform solution to, at least, every validity Curry paradox, something else needs to be done. We introduce, as an answer to Porter's challenge, validity theories that are substructural at every metainferential level. In this sense, they are free from Porter's criticism. These theories count as both theories of naïve validity (because they do not give up V P or V D, nor any higher-level version of them) and as offering a uniform solution to validity curries, in the more traditional and intuitive sense: either they give up every higher-level version of Cut – as well as every higher-level version of it – or they give up Contraction – as well as every higher-level version of it. Finally, we provide philosophical justifications for each of these options.

Bryan Stanley Turner's Philosophical Thoughts: Citizenship Outline Analysis

Bryan Stanley Turner is a professor of sociology and politics who was born in Birmingham, England on January 14, 1945. Bryan S Turner is also a director of the religious committee at The City University Of New York. He is also the director of the Centre for Contemporary Islamic Studies at the University of Western Sydney. Not stopping there, Brayen is also registered as a member of the American Sociological Research Association. This study uses a literature study. Literature study research is oriented to theories relevant to research problems. In this satisfaction study, it analyzes Bryan Stanley Turner's journal of philosophical thought about the analysis of the outline of citizenship published in 1990 in the British Sociological Association and several other relevant reading sources. Turner's view of the conception of citizenship is a set of rights and obligations that give a formal legal identity to individuals. These legal and formal rights have historically been united as a set of social institutions such as the jury system, parliaments and the welfare state. Bryan Turner offered a revised model of citizenship that included legal rights (or civil rights in the Marshall model), political rights, social rights and human rights.

Creating a Broader Impacts Culture

Minority-serving institutions (MSIs), historically Black colleges and universities, and Tribal colleges and universities play a pivotal role in championing inclusivity and diversity within higher education systems across the nation. As catalysts for social change, they regularly engage in Broader Impacts (BI) work. However, these institutions often face challenges in resource allocation and a dearth of human capital to sustain vital operations, which hinders their capacity building efforts. This article explores how MSIs, exemplified by the City College of the City University of New York (CUNY), can empower faculty, staff, and trainees engaged in BI work to extend their influence beyond their institutions by adopting a community of practice and engagement (COPE) approach. By leveraging collective strengths of knowledge, expertise, and diversity, CUNY’s MSI campuses endeavor to foster a transformative ripple effect, shaping a more inclusive and equitable future through research and innovation. Beginning in Spring 2023, the City College of New York, in partnership with the Advancing Research Impact in Society (ARIS) National Science Foundation–funded center, embarked on initiatives to understand the existing BI culture, knowledge, and challenges to enhance BI and research development capacity across the CUNY system. This article discusses the pilot efforts and lessons learned from these endeavors.

NIMG-18. RAPID INTRAOPERATIVE DIFFERENTIATION OF PRIMARY CNS LYMPHOMA FROM COMMON CNS TUMORS USING STIMULATED RAMAN HISTOLOGY AND DEEP LEARNING

Abstract Accurate intraoperative diagnosis is crucial for decision-making during tumor surgery and stereotactic-guided biopsies. Differentiating between primary CNS lymphoma (PCNSL) and non-PCNSL entities such as gliomas and metastatic cancers presents significant challenges due to overlapping histomorphological features, time constraints, and the differing further therapy strategies required. Our study utilized deep learning and stimulated Raman histology (SRH) to address this challenge. We imaged unprocessed, label-free tissue samples intraoperatively using a portable Raman scattering microscope, generating virtual H&E-like images within less than five minutes. Our training set comprised over 54,000 SRH tumor patch images, including various brain tumors. We developed a deep learning pipeline utilizing a ResNet-50 feature extractor and the BYOL (Bootstrap Your Own Latent) self-supervised learning algorithm. We trained a binary linear classifier to distinguish between PCNSL and non-PCNSL entities. Data were collected from New York University (NYU), the University of Cologne (UKK), and the University of Michigan (UM), including 263 SRH whole slide images from 177 patients sourced from surgical resections and stereotactic-guided biopsies. In an international bicentric test cohort (N = 100, NYU and UKK), the model achieved an overall accuracy of 91.1%, an area under the receiver operating characteristic curve (AUROC) of 0.983, a sensitivity of 89.3%, and a specificity of 100% compared to the final formalin-fixed, paraffin-embedded-based diagnosis. Our algorithm can accurately extract typical histomorphological features to help differentiate between PCNSL and non-PCNSL tumors. These results underscore the potential of deep learning models to enhance the accuracy of intraoperative diagnoses for PCNSL and non-PCNSL entities, thereby improving intraoperative decision-making and further treatment strategies by supporting neuropathologists and surgeons.

Seventy-five years of impactful environmental and occupational health research at the Nelson Institute of Environmental Medicine at New York University.

Founded in 1947 as the Institute of Industrial Medicine, the Nelson Institute and Department of Environmental Medicine at New York University (NYU) Grossman School of Medicine (NYUGSOM) was supported by a National Institute of Environmental Health Science (NIEHS) Center Grant for over 56 years. Nelson Institute researchers generated 75 years of impactful research in environmental and occupational health, radiation effects, toxicology, and cancer. Environmental health research is continuing at NYUGSOM in its departments of medicine and population health. The objective of this historical commentary is to highlight the major achievements of the Nelson Institute and the department in the context of its history at facilities in Sterling Forest, Tuxedo, NY and Manhattan, NY. Aspects of our discussion include leadership, physical facilities, and research in many areas, including air pollution, health effects of environmental radiation exposures, inhalation toxicology methodology, carcinogenesis by chemicals, metals, and hormones, cancer chemoprevention, human microbiome, ecotoxicology, epidemiology, biostatistics, and community health concerns. The research of the institute and department benefited from unique facilities, strong leadership focused on team-based science, and outstanding investigators, students, and staff. A major lasting contribution has been the training of hundreds of graduate students and postdoctoral fellows, many of whom have been and are training the next generation of environmental and occupational health researchers at various institutions.

Maintaining and Improving Virtual Call Centers for Public Health: The CUNY Recovery Corps Experience.

In the early stages of the COVID-19 pandemic in New York City, individuals who tested positive, exposed contacts, and their families faced significant challenges in accessing essential resources for successful quarantine and isolation. These resources included alternative housing, food access, grocery delivery, missed wage assistance, transportation for household members, medical care, and more. The City University of New York (CUNY) Recovery Corps emerged as an innovative virtual resource navigation service to connect New Yorkers to essential resources and later to vaccinations, treatment, and services for Long COVID. The program prioritized reaching people living in underserved and under-resourced neighborhoods. Over 3years, it made a substantial impact, reaching 374728 New Yorkers and connecting 135147 to crucial resources. Implementing a successful virtual navigation center amidst a global pandemic posed unique challenges. It required well-established systems to provide services, manage staff, build community, and both assure and improve service quality. Continuous quality assurance and process improvement efforts resulted in a nearly 10-point increase in successful client-resource connections program-wide, with individual navigators showing improvements of up to 20%. Staff-led initiatives, such as population-specific resource guides and incentive awards, boosted morale and cohesion. Providing continuous professional development opportunities and requested training including trauma-informed care, self-care, and strategies for managing grief contributed to staff retention and improved client interactions. The CUNY Recovery Corps provides a blueprint for setting up navigation services for emergency response and supporting health and social services. The current paper delves into the intensive quality improvement efforts, program management infrastructure, and staff support that made this program a success. By applying these lessons on establishing and maintaining a virtual call center prioritizing the client's well-being and staff strengths, policymakers and social service leaders can effectively address various health and social service needs proactively rather than rebuilding after each disaster.

Hurricane Katrina survivor now helps NYC students deal with emergencies

Dr. Rhonda Mouton, Director of LaGuardia College Access for Retention and Economic Success (CARES) at LaGuardia Community College (part of the City University of New York), knows what it feels like to have housing uncertainty. In 2005, she and her family fled New Orleans just prior to Hurricane Katrina. She lost her home and her daycare business and had to begin anew not just for her and her daughter but for her extended family.

Hurricane Katrina survivor now helps NYC students deal with emergencies

Dr. Rhonda Mouton, Director of LaGuardia College Access for Retention and Economic Success (CARES) at LaGuardia Community College (part of the City University of New York), knows what it feels like to have housing uncertainty. In 2005, she and her family fled New Orleans just prior to Hurricane Katrina. She lost her home and her daycare business and had to begin anew not just for her and her daughter but for her extended family.

Teaching the art of compassionate inquiry: involving survivors from 9/11 in social work education

ABSTRACT This paper reports findings from pedagogic research evaluating the impact of the involvement of survivors from the World Trade Center attacks in New York City in 2001 in trauma-specific social work education. A pedagogic approach to mental health education is discussed which aims to prepare students to develop trauma-informed assessment and intervention skills concurrent with their encounters with trauma survivors in field practice placements. The small-scale research involved surveying students’ evaluations at a university in New York, following exposure to first-hand accounts of survivors’ experiences. Across the three areas—confidence in knowledge of trauma, impact on learning, and preparation for field practice—the evaluation findings indicate that the students’ knowledge, gained from the involvement of 9/11 survivors, improved over time. This paper presents the background to this project, preparations involved, and findings from research evaluations with the students. The findings suggest that the involvement of those with direct and lived trauma experience in classroom teaching, whilst challenging, can yield positive impacts for students. The 9/11 survivors poignantly shared with students that their lives were changed forever in the aftermath of these events. The findings have potential global educational impact and resonance.

Does OER Research-Based Learning Improve Performance: A Case Study from Students Enrolled in a Community College at City University of New York (CUNY)

This study investigates the effectiveness of Open Education Resources (OER) and research-based learning. A quantitative analysis was conducted on students enrolled in Macroeconomic courses at a City University of New York (CUNY) community college in Spring 2018 and Spring 2019. Prior studies have shown the positive effects of conducting research in natural sciences. Yet, there is a lot to be inquired about regarding the impact of research-based learning on social sciences and the use of OER materials as a cheaper and more equitable alternative to commercial textbooks. Additionally, this study evaluates the academic effectiveness of research-based learning through assessment scores through a new indicator, the growth mindset profile. Findings should be used to assess critically research-based learning applications incorporated with OER materials and open pedagogy, as well as to consider the duration, wide application throughout the program rather than isolated courses, and design of course assignments and research projects to engage students actively and effectively.

The Use of fMRI Regional Analysis to Automatically Detect ADHD Through a 3D CNN-Based Approach.

Attention deficit hyperactivity disorder (ADHD) is a neurodevelopmental disorder characterized by a reduced attention span, hyperactivity, and impulsive behaviors, which typically manifest during childhood. This study employs functional magnetic resonance imaging (fMRI) to use spontaneous brain activity for classifying individuals with ADHD, focusing on a 3D convolutional neural network (CNN) architecture to facilitate the design of decision support systems. We developed a novel deep learning model based on 3D CNNs using the ADHD-200 database, which comprises datasets from NeuroImage (NI), New York University (NYU), and Peking University (PU). We used fractional amplitude of low-frequency fluctuations (fALFF) and regional homogeneity (ReHo) data in three dimensions and performed a fivefold cross-validation to address the dataset imbalance. We aimed to verify the efficacy of our proposed 3D CNN by contrasting it with a fully connected neural network (FCNN) architecture. The 3D CNN achieved accuracy rates of 76.19% (NI), 69.92% (NYU), and 70.77% (PU) for fALFF data. The FCNN model yielded lower accuracy rates across all datasets. For generalizability, we trained on NI and NYU datasets and tested on PU. The 3D CNN achieved 69.48% accuracy on fALFF outperforming the FCNN. Our results demonstrate that using 3D CNNs for classifying fALFF data is an effective approach for diagnosing ADHD. Also, FCNN confirmed the efficiency of the designed model.

Hospital charges for laparoscopic sleeve gastrectomy compared to robotic sleeve gastrectomy: a multicenter study.

Sleeve gastrectomy has become a gold standard in addressing medically refractory obesity. Robotic platforms are becoming more utilized, however, data on its cost-effectiveness compared to laparoscopy remain controversial (1-3). At NYU Langone Health, many of the bariatric surgeons adopted robotic surgery as part of their practices starting in 2021. We present a retrospective cost analysis of laparoscopic sleeve gastrectomy (LSG) vs. robotic sleeve gastrectomy (RSG) at New York University (NYU) Langone Health campuses. All adult patients ages 18-65 who underwent LSG or RSG from 202 to 2023 at NYU Langone Health campuses (Manhattan, Long Island, and Brooklyn) were evaluated via electronic medical records and MBSAQIP 30-day follow-up data. Patients with prior bariatric surgery were excluded. Complication-related ICD-10/CPT codes are collected and readmission costs will be estimated from ICD codes using the lower limit of CMS transparent NYU standard charges (3). Direct charge data for surgery and length of stay cost data were also obtained. Statistical T-test and chi-squared analysis were used to compare groups. Direct operating cost data at NYU Health Campuses demonstrated RSG was associated with 4% higher total charges, due to higher OR charges, robotic-specific supplies, and more post-op ED visits. RSG was associated with higher overall hospital charges compared to LSG, though there are multiple contributing factors. More research is needed to identify cost saving measures. This study is retrospective in nature, and does not include indirect costs nor reimbursement. Direct operating costs, per contractual agreement with suppliers, are only given as percentages. Data are limited to 30-day follow-up.

Wrestling with the deepfakes: Detection and beyond

Wrestling with the deepfakes: Detection and beyond Siwei Lyu, SUNY Empire Innovation Professor from the University at Buffalo, State University of New York, delves into detection and beyond in the realm of DeepFakes, starting with a look at what they are. Since 2017, the term “DeepFake” has become widely known, often appearing in news and media. It combines “deep learning” (a type of artificial intelligence [AI] model) and “fake” to describe synthetic media – like text, audio, images, and videos – created with advanced AI technologies. The concept gained notoriety when a Reddit user began sharing AI-generated pornographic videos that superimposed celebrity faces onto other figures.

Multicenter privacy-preserving model training for deep learning brain metastases autosegmentation

Objectives:This work aims to explore the impact of multicenter data heterogeneity on deep learning brain metastases (BM) autosegmentation performance, and assess the efficacy of an incremental transfer learning technique, namely learning without forgetting (LWF), to improve model generalizability without sharing raw data. Materials and methods:A total of six BM datasets from University Hospital Erlangen (UKER), University Hospital Zurich (USZ), Stanford, UCSF, New York University (NYU), and BraTS Challenge 2023 were used. First, the performance of the DeepMedic network for BM autosegmentation was established for exclusive single-center training and mixed multicenter training, respectively. Subsequently privacy-preserving bilateral collaboration was evaluated, where a pretrained model is shared to another center for further training using transfer learning (TL) either with or without LWF. Results:For single-center training, average F1 scores of BM detection range from 0.625 (NYU) to 0.876 (UKER) on respective single-center test data. Mixed multicenter training notably improves F1 scores at Stanford and NYU, with negligible improvement at other centers. When the UKER pretrained model is applied to USZ, LWF achieves a higher average F1 score (0.839) than naive TL (0.570) and single-center training (0.688) on combined UKER and USZ test data. Naive TL improves sensitivity and contouring accuracy, but compromises precision. Conversely, LWF demonstrates commendable sensitivity, precision and contouring accuracy. When applied to Stanford, similar performance was observed. Conclusion:Data heterogeneity (e.g., variations in metastases density, spatial distribution, and image spatial resolution across centers) results in varying performance in BM autosegmentation, posing challenges to model generalizability. LWF is a promising approach to peer-to-peer privacy-preserving model training.

It’s About Time: The Inequitable Distribution of Time as a Resource for College, by Gender and Race/Ethnicity

Existing qualitative research in higher education on students’ work and family commitments already suggests that time as a resource for college is likely not distributed equitably by race/ethnicity or gender. However, the relationship between race/ethnicity, gender, and time as a resource for college has yet to be quantitatively measured in large-scale higher education research. This study explored whether gender or race/ethnicity correlated with differences in time as a resource for college; and further, the extent to which differences in time as a resource for college may be explained by other factors such as age, number of children, and access to childcare. Retrospective survey responses (n = 41,579) on self-reported time use were merged with institutional data records from students at the City University of New York (CUNY), a large diverse public university in the U.S. Women, Black, and Hispanic students were all significantly more time poor than male, White, or Asian students. Age accounted for significant portions of these differences, perhaps because it correlates with increased work and family responsibilities. Having children as well as a student’s access to childcare also explained a significant portion of inequitable distributions of time as a resource for college.

Meta-Analysis of Children's Moral Education in the Digital Era: A Bibliometric Analysis (1977-2021)

This study uses bibliometric methods to conduct a meta-analysis of research on children's moral education in the digital era, published between 1977 and 2021. The study analyzes literature to address questions related to children's moral education in a digitalized society, where digitalization affects various aspects of social life and changes ways of thinking and living. The bibliometric analysis excludes the year 2022. Data was collected to examine the contributions of various authors, research institutions, and countries in this field. The analysis identifies key authors contributing to the field, with Cooney, M., Firth, A., and Herman-Pillath, C. being the most prolific. Leading research institutions include Delft University of Technology, City University of New York, and Cornell University. The United States, the United Kingdom, and China are the top contributing countries. The study also visualizes the types of documents published, with articles being the most common (48.6%), followed by book chapters (27.6%), books (10.0%), reviews (9.0%), conference papers (2.1%), editorials (2.1%), and notes (0.7%). This study highlights the importance of moral education in the digital era, emphasizing that effective methods—such as best practices, socialization, counseling, mindfulness, humor, and supervision—are crucial for children to navigate digital media responsibly. The findings provide valuable insights for families, educators, and policymakers to enhance moral education strategies in a rapidly evolving digital landscape.

Project Impact: A Public Community College Campus–Based Support and Reentry Program for Justice-Impacted Students in New York City

Background: Between 70 million and 100 million Americans have a record of interaction with the criminal legal system, a group facing stigma and discrimination in civic participation, housing, employment, and education. Justice-impacted people face collateral consequences in the community, making reentry programs essential to success at reintegration. Focus of Study: Project Impact (PI) at the Borough of Manhattan Community College/City University of New York (BMCC/CUNY) supports students impacted by the criminal legal system. PI provides direct services to justice-impacted students through an extensive peer mentoring program and a high school equivalency program, and by offering college credit–bearing courses to students presently incarcerated in partnership with John Jay College’s Prison-to-College Pipeline. PI defines “justice impacted” expansively as students who have been formerly incarcerated, have family who have been incarcerated or arrested, have had issues in family court, or in any other way have been significantly affected by the criminal legal system. Research Design/Conclusions: This evaluation provides unequivocal evidence of the strong, positive impact that PI is making, serving as a rare campus-based reentry program within a community college in New York City.

Life Stressors as Predictors of Community College Students’ Course Outcomes: The COVID-19 Pandemic as a Focusing Event

Background or Context: The COVID-19 pandemic may be considered a focusing event that has drawn greater attention to the physical and mental health of community college students. Purpose, Objective, Research Question, or Focus of Study: This study examined community college students’ self-reported life stressors pre- and post-pandemic-onset to identify any reported differences, both in occurrence and in impact on course completion. Research Design: Surveys were collected after the COVID-19 pandemic onset at the City University of New York’s largest community college. The 529 students enrolled in STEM subjects were asked about potentially stressful life events experienced during spring 2020 and how strongly these impacted the time or energy they had for their studies (response rate 88.2%). Conclusions or Recommendations: Students’ reporting of the presence of life stressors, as well as their rating of the impact on their studies, were significantly predictive of subsequent course outcomes. However, stressors reported as beginning prior to the pandemic had a larger and more significant relationship with course outcomes than those reported as beginning after pandemic onset. Further, stressors were already highly prevalent prepandemic, with over three-quarters of students reporting them. Although the pandemic may have drawn attention to the existence and impact of stressors, it may simply be a focusing event that drew heightened attention to a pre-existing issue. Results from this study indicate that life stress may significantly impact the outcomes of community college students even outside of pandemic conditions, suggesting that this is a critical area for community colleges to address.

Telemedicine in Cardiology: Enhancing Access to Care and Improving Patient Outcomes.

Telemedicine has gained significant recognition, particularly since the COVID-19 pandemic. However, its roots date back to its significant role during major epidemic outbreaks such as severe acute respiratory syndrome (SARS), H1N1 and H7N9 influenza, and Middle East respiratory syndrome (MERS), where alternate means of accessing healthcare were adopted to combat the outbreak while limiting the spread of the virus.In Sub-Saharan Africa, telemedicine has supported healthcare delivery, patient and professional health education,disease prevention, and surveillance, starting with its first adoption in Ethiopia in 1980. In the United States, telemedicine has significantly impacted cardiology, particularly at-home monitoring programs, which have proven highly effective for patients with abnormal heart rhythms. Devices such as Holter monitors, blood pressure monitors, and implantable cardioverter-defibrillators have reduced mortality rates and hospital readmissions while improving healthcare efficiency by saving healthcare costs. However, the COVID-19 pandemic accelerated the adoption of telemedicine, as evidenced by a dramatic increase in telemedicine visits at institutions like New York University (NYU) Langone Health during and post-COVID-19 pandemic. In addition, telemedicine has also facilitated cardiac rehabilitation and improved access to specialized cardiology care in rural and underserved areas, reducing disparities in cardiovascular health outcomes. As technology advances, telemedicine is poised to play an increasingly significant role in cardiology and healthcare at large, enhancing patient management, healthcare efficiency, and cost reduction. This review underscores the significance of telemedicine in cardiology, its challenges, and future directions.

Impact of Transfer Learning Using Local Data on Performance of a Deep Learning Model for Screening Mammography.

Purpose To investigate the issues of generalizability and replication of deep learning models by assessing performance of a screening mammography deep learning system developed at New York University (NYU) on a local Australian dataset. Materials and Methods In this retrospective study, all individuals with biopsy or surgical pathology-proven lesions and age-matched controls were identified from a South Australian public mammography screening program (January 2010 to December 2016). The primary outcome was deep learning system performance-measured with area under the receiver operating characteristic curve (AUC)-in classifying invasive breast cancer or ductal carcinoma in situ (n = 425) versus no malignancy (n = 490) or benign lesions (n = 44). The NYU system, including models without (NYU1) and with (NYU2) heatmaps, was tested in its original form, after training from scratch (without transfer learning), and after retraining with transfer learning. Results The local test set comprised 959 individuals (mean age, 62.5 years ± 8.5 [SD]; all female). The original AUCs for the NYU1 and NYU2 models were 0.83 (95% CI: 0.82, 0.84) and 0.89 (95% CI: 0.88, 0.89), respectively. When NYU1 and NYU2 were applied in their original form to the local test set, the AUCs were 0.76 (95% CI: 0.73, 0.79) and 0.84 (95% CI: 0.82, 0.87), respectively. After local training without transfer learning, the AUCs were 0.66 (95% CI: 0.62, 0.69) and 0.86 (95% CI: 0.84, 0.88). After retraining with transfer learning, the AUCs were 0.82 (95% CI: 0.80, 0.85) and 0.86 (95% CI: 0.84, 0.88). Conclusion A deep learning system developed using a U.S. dataset showed reduced performance when applied "out of the box" to an Australian dataset. Local retraining with transfer learning using available model weights improved model performance. Keywords: Screening Mammography, Convolutional Neural Network (CNN), Deep Learning Algorithms, Breast Cancer Supplemental material is available for this article. © RSNA, 2024 See also commentary by Cadrin-Chênevert in this issue.