Zhihong Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zhihong Zhang has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
Ali Zolnour
Columbia University Irving Medical Center, New York, NY 10032, United States
2 shared publicationsMaryam Zolnoori
Data Science Institute, Columbia University, New York, NY 10027, United States; Columbia University Irving Medical Center, New York, NY 10032, United States
2 shared publicationsMaxim Topaz
Data Science Institute, Columbia University, New York, NY 10027, United States; School of Nursing, Columbia University, New York, NY 10032, United States
2 shared publicationsMohamad Javad Momeni Nezhad
Columbia University Irving Medical Center, New York, NY 10032, United States
2 shared publicationsHans Moen
Department of Computer Science, Aalto University, Espoo, Uusimaa, Finland
1 shared publicationHossein Azadmaleki
Columbia University Irving Medical Center, New York, NY 10032, USA
1 shared publication
Latest Publications
- 22 December 2025
- Published by Elsevier in International Journal of Medical Informatics
- Vol. 208, 106231
Advancing healthcare with large language models: A scoping review of applications and future directions
- 22 December 2025
- Published by Elsevier in International Journal of Medical Informatics
- Vol. 208, 106231
Advancing healthcare with large language models: A scoping review of applications and future directions
- 2 July 2025
- Published by Elsevier in International Journal of Medical Informatics
- Vol. 203, 106035
Enhancing AI for citation screening in literature reviews: Improving accuracy with ensemble models
- 2 July 2025
- Published by Elsevier in International Journal of Medical Informatics
- Vol. 203, 106035
Enhancing AI for citation screening in literature reviews: Improving accuracy with ensemble models
- 1 July 2025
- Published by Elsevier in Nursing Outlook
- Vol. 73 (4) , 102472
Toward equitable documentation: Evaluating ChatGPT’s role in identifying and rephrasing stigmatizing language in electronic health records
- 1 July 2025
- Published by Elsevier in Nursing Outlook
- Vol. 73 (4) , 102472
Toward equitable documentation: Evaluating ChatGPT’s role in identifying and rephrasing stigmatizing language in electronic health records
- 2 June 2025
- Published by Springer Nature in Pediatric Research
- Vol. 99 (1) , 159-165
Time series analysis of impact of COVID-19 on infant and neonatal mortality in the United States
Background This study aims to conduct a time series analysis to assess the effect of COVID-19 on infant and neonatal mortality in its first three years. Methods Linked infant birth and death certificate data from the United States between 2016 and 2022 were used. Long Short-Term Memory (LSTM) and Autoregressive Integrated Moving Average (ARIMA)/ARIMA with Exogenous Variables (ARIMAX) were employed to predict monthly infant and neonatal mortality trends in 2022 with and without incorporating COVID-19 as a feature. Interrupted Time Series Analysis was conducted to measure monthly rate changes before (pre-March 2020) and during the COVID-19 period (March 2020–December 2022). Results Infant and neonatal deaths declined from 2016 to 2020/2021, with increases observed starting in May 2022. Incorporating COVID-19 as a variable into prediction models reduced the predictive performance of both LSTM and ARIMA/ARIMAX models. Compared to the pre-COVID-19 period, ITS showed no significant changes in monthly infant and neonatal death rates. However, significant increases were noted in low birth weight, low Apgar scores, and infrequent prenatal visits during 2021-2022. Conclusion COVID-19 altered the previously continuous decline in infant and neonatal deaths in 2022. Despite this late increase, the overall impact of COVID-19 on infant and neonatal mortality remains minimal. Impact Infant and neonatal deaths showed a steady decline from 2016 through 2020/2021. However, an increase in infant and neonatal deaths emerged in the later stages of the COVID-19 pandemic (2022). Rates of risk factors associated with infant and neonatal deaths, such as low birth weight and inadequate prenatal care, increased during the COVID-19 period. Understanding the shifts in infant and neonatal mortality and related risk factors during the COVID-19 pandemic provides valuable insights into its impact and enhances preparedness for future pandemics.
- 2 June 2025
- Published by Springer Nature in Pediatric Research
- Vol. 99 (1) , 159-165
Time series analysis of impact of COVID-19 on infant and neonatal mortality in the United States
Background This study aims to conduct a time series analysis to assess the effect of COVID-19 on infant and neonatal mortality in its first three years. Methods Linked infant birth and death certificate data from the United States between 2016 and 2022 were used. Long Short-Term Memory (LSTM) and Autoregressive Integrated Moving Average (ARIMA)/ARIMA with Exogenous Variables (ARIMAX) were employed to predict monthly infant and neonatal mortality trends in 2022 with and without incorporating COVID-19 as a feature. Interrupted Time Series Analysis was conducted to measure monthly rate changes before (pre-March 2020) and during the COVID-19 period (March 2020–December 2022). Results Infant and neonatal deaths declined from 2016 to 2020/2021, with increases observed starting in May 2022. Incorporating COVID-19 as a variable into prediction models reduced the predictive performance of both LSTM and ARIMA/ARIMAX models. Compared to the pre-COVID-19 period, ITS showed no significant changes in monthly infant and neonatal death rates. However, significant increases were noted in low birth weight, low Apgar scores, and infrequent prenatal visits during 2021-2022. Conclusion COVID-19 altered the previously continuous decline in infant and neonatal deaths in 2022. Despite this late increase, the overall impact of COVID-19 on infant and neonatal mortality remains minimal. Impact Infant and neonatal deaths showed a steady decline from 2016 through 2020/2021. However, an increase in infant and neonatal deaths emerged in the later stages of the COVID-19 pandemic (2022). Rates of risk factors associated with infant and neonatal deaths, such as low birth weight and inadequate prenatal care, increased during the COVID-19 period. Understanding the shifts in infant and neonatal mortality and related risk factors during the COVID-19 pandemic provides valuable insights into its impact and enhances preparedness for future pandemics.
- 1 May 2025
- Published by Springer Nature in Supportive Care in Cancer
- Vol. 33 (5) , 1-9
A scoping review of resilience in adults with cancer based on the NIH resilience concept model
Purpose The NIH Resilience Concept Model (NIH-RCM) is a comprehensive model that conceptualizes resilience through three components: challenge or stressor, system, and the system’s responses, providing a clear definition and measurement approach for resilience research. This scoping review uses the NIH-RCM as a framework to organize and summarize research on resilience in adults with cancer, especially those aged 65 + , and identify research gaps. Methods A systematic search of PubMed, CINAHL, PsycINFO, and Web of Science was conducted to identify studies that examined resilience in the individual domain of adults (including those ≥ 65 years) with cancer from inception to July 2022. Included studies were organized into the components (stressor, system, and responses), definition, and measures of resilience. Results Eighty-nine studies were included. Most (88%) studies included adults of all ages, with only ten focusing exclusively on older adults. Of the 89 studies, 78% identified cancer-related events as stressors, and 94% of studies identified the individual psychosocial system as the system affected by the stressor and measured the system’s responses within the psychosocial subdomain. Several resilience definitions were identified from 60% of studies, with 44% conceptualizing resilience as an ability. Five types of resilience measures were identified in 99% of studies, with 15% measuring resilience by comparing pre- and post-stressor changes in the system’s responses. Conclusions There was a range of definitions and measures for resilience. Gaps in research include a minimal focus on older adults with cancer and individuals’ physiological and molecular/genetic systems and their responses within these subdomains.
- 1 May 2025
- Published by Springer Nature in Supportive Care in Cancer
- Vol. 33 (5) , 1-9
A scoping review of resilience in adults with cancer based on the NIH resilience concept model
Purpose The NIH Resilience Concept Model (NIH-RCM) is a comprehensive model that conceptualizes resilience through three components: challenge or stressor, system, and the system’s responses, providing a clear definition and measurement approach for resilience research. This scoping review uses the NIH-RCM as a framework to organize and summarize research on resilience in adults with cancer, especially those aged 65 + , and identify research gaps. Methods A systematic search of PubMed, CINAHL, PsycINFO, and Web of Science was conducted to identify studies that examined resilience in the individual domain of adults (including those ≥ 65 years) with cancer from inception to July 2022. Included studies were organized into the components (stressor, system, and responses), definition, and measures of resilience. Results Eighty-nine studies were included. Most (88%) studies included adults of all ages, with only ten focusing exclusively on older adults. Of the 89 studies, 78% identified cancer-related events as stressors, and 94% of studies identified the individual psychosocial system as the system affected by the stressor and measured the system’s responses within the psychosocial subdomain. Several resilience definitions were identified from 60% of studies, with 44% conceptualizing resilience as an ability. Five types of resilience measures were identified in 99% of studies, with 15% measuring resilience by comparing pre- and post-stressor changes in the system’s responses. Conclusions There was a range of definitions and measures for resilience. Gaps in research include a minimal focus on older adults with cancer and individuals’ physiological and molecular/genetic systems and their responses within these subdomains.