Abstract
In the era of big data, generative artificial intelligence (AI) models are currently in a boom. The Chatbot Generative Pre-trained Transformer (ChatGPT), a large language model (LLM) developed by OpenAI (San Francisco, CA), is a type of AI software that could generate text based on the input it receives. In this study, in order to explore how ChatGPT could give reflections and suggestions about the sudden outbreak of Mpox in 2022 from the AI dimensions, our group talked with ChatGPT with several questions about Mpox. We hope this talk could enrich our knowledge on Mpox from the new AI dimensions and also explore the possibility of human and AI fight shoulder to shoulder for prevention and containment of the potential epidemics or pandemics in future.
Keywords: Mpox, Infectious disease, Public health, Artificial intelligence, ChatGPT, Chatbots
Dear Editor,
Since May 2022, the sudden emergence and transmission of monkeypox (Mpox) in locations that did not typically have cases has frequently become the headlines of academic topics and stirred public fears around a possible new global pandemic [1]. Mpox is caused by a double-stranded DNA virus belonging to the genus Orthopoxvirus, whose more notorious member includes the causative agent of smallpox. On July 23, 2022, the World Health Organization (WHO) declared Mpox outbreak to be a Public Health Emergency of International Concern (PHEIC). As of March 17, 2023, a total of 86,601 confirmed cases including 112 deaths from 110 countries/regions have been reported to WHO. In our previous studies, through the bibliometric approach, we have reported the global research trend of Mpox-related and Orthopoxvirus-related publications since the 80 s and 90 s of last century [2, 3]. Our results showed that Mpox was the most neglected branch compared with the other three orthopoxvirus genus, such as variola virus, cowpox virus, and vaccinia virus. However, the latest studies demonstrated that the number of Mpox-related publications on epidemiology data, outbreak surveillance, treatment, prevention, vaccines, and so on is showing a blowout trend [4]. Nevertheless, the emergence of a large amount of information about Mpox may lead to information overload problems and make it difficult for public or scholars to find reliable sources of information and direction.
In the era of big data, generative artificial intelligence (AI) models are currently in a boom. The Chatbot Generative Pre-trained Transformer (ChatGPT), a large language model (LLM) developed by OpenAI (San Francisco, CA), is a type of AI software that could generate text based on the input it receives [5]. ChatGPT possesses the capacities of understanding and mimicking natural language and could solve cognitive problems by reinforcement techniques and machine learning from online resources and subsequently further improve its efficiency to correctly respond to users’ needs. Since the introduction, ChatGPT has been tested on multiple domains to confirm its functioning in dialogue and communication by different industries. In the medical domain, despite the lack of specific clinical advice training and practice, one study reported that ChatGPT was able to pass the United States Medical Licensing Exam (USMLE) has generated much discussion about the pros and cons of it [6]. Meanwhile, in recent studies, many scholars have discussed the application of ChatGPT in medical research [7, 8]. Their findings showed that despite deficits in situational awareness, inference, and consistency, ChatGPT could answer some open-ended medical queries almost the same as the specialist doctors.
In this study, in order to explore how ChatGPT could give reflections and suggestions about the sudden outbreak of Mpox in 2022 from the AI dimensions, our group talked with ChatGPT with several questions about Mpox on March 17, 2023 (Fig. 1). We hope this talk could enrich our knowledge on Mpox from the new AI dimensions and also explore the possibility of human and AI fight shoulder to shoulder for prevention and containment of the potential epidemics or pandemics in future.
Question one
On July 23, 2022, the World Health Organization (WHO) declared Mpox outbreak to be a Public Health Emergency of International Concern (PHEIC). What do you think is the cause of this sudden emergence and transmission of monkeypox (Mpox) in locations that did not typically have cases?
Answer from ChatGPT
As an AI language model, I cannot speculate or provide a definitive answer as to the cause of the sudden emergence and transmission of monkeypox (Mpox) in locations that did not typically have cases. However, there are several factors that may contribute to the emergence of infectious diseases in general, including (1) Environmental factors: Changes in the environment, such as deforestation, urbanization, and climate change, can alter the habitat of animals and increase the likelihood of human–animal interactions, leading to the emergence of zoonotic diseases. (2) Human behavior: Human activities such as encroachment on wildlife habitats, consumption of bushmeat, and international travel can facilitate the spread of infectious diseases. (3) Pathogen evolution: Pathogens can evolve and adapt to new environments or host populations, leading to the emergence of new strains or outbreaks. (4) It is possible that a combination of these factors or others may be contributing to the sudden emergence and transmission of monkeypox in new locations. Further investigation and research are needed to better understand the underlying causes of this outbreak.
Question two
As of March 17, 2023, a total of 86,601 confirmed cases including 112 deaths from 110 countries/regions have been reported to WHO. Please analyze the trends of confirmed cases of Mpox in future.
Answer from ChatGPT
As an AI language model, I am not able to predict future trends in confirmed cases of monkeypox (Mpox) with certainty. However, I can provide some insights based on the available data and the general trends observed in infectious disease outbreaks. The trajectory of future Mpox cases will depend on several factors, including (1) Control measures: The effectiveness of control measures, such as vaccination campaigns, isolation and quarantine of infected individuals, and public health education, will play a critical role in containing the outbreak. (2) Surveillance and reporting: The accuracy and completeness of surveillance and reporting systems will influence the reliability of case counts and trends. (3) Human behavior: The adherence to prevention and control measures by the general public will also impact the spread of the virus. (4) Given the current status of the outbreak, it is difficult to predict the exact trajectory of Mpox cases in future. However, it is possible that the number of cases may continue to rise in some areas, particularly where control measures are not well implemented or where surveillance and reporting systems are not robust. On the other hand, if control measures are effective and surveillance and reporting systems are strengthened, the number of cases could potentially decrease over time. It is important to note that infectious disease outbreaks can be unpredictable, and the trajectory of the Mpox outbreak may change in response to various factors. Continual monitoring and response efforts will be necessary to mitigate the impact of the outbreak and prevent its further spread.
Question three
What are the implications for the future?
Answer from ChatGPT
The implications for the future of the monkeypox (Mpox) outbreak are significant and multifaceted. Some potential implications include (1) Public health impact: The Mpox outbreak has already resulted in thousands of cases and dozens of deaths, and it has the potential to continue to spread and cause significant morbidity and mortality in affected communities. (2) Economic impact: The outbreak has also had economic consequences, such as loss of productivity and income due to illness, as well as costs associated with control and response efforts. (3) Global health security: The emergence and spread of Mpox in new areas highlight the need for robust global health security measures, such as surveillance and response systems, to detect and respond to emerging infectious disease threats. (4) Vaccine development: The outbreak underscores the need for continued research and development of effective vaccines and treatments for Mpox and other zoonotic diseases. (5) Social and cultural impact: The outbreak has also had social and cultural implications, such as increased stigma and fear associated with Mpox, and the potential impact on traditional practices, such as the consumption of bushmeat. In summary, the Mpox outbreak has significant implications for public health, economics, global health security, vaccine development, and social and cultural norms. Continued monitoring, control efforts, and research are needed to mitigate the impact of the outbreak and prevent its further spread.
Question four
Give me five novel systematic review ideas that have not been published related to monkeypox (Mpox).
Answer from ChatGPT
As an AI language model, I can suggest some potential systematic review ideas related to monkeypox (Mpox) that have not been published, based on the current knowledge and available literature on the subject. Here are five possible ideas: (1) A systematic review of the epidemiology and clinical characteristics of Mpox in immunocompromised individuals. This review could explore the incidence, prevalence, and clinical features of Mpox in individuals with various forms of immunosuppression, such as HIV/AIDS, cancer, or organ transplantation. (2) A systematic review of the public health response to Mpox outbreaks in different settings. This review could examine the effectiveness of various control measures, such as vaccination campaigns, contact tracing, and isolation and quarantine measures, in different contexts, such as rural or urban areas, or low- and middle-income countries. (3) A systematic review of Mpox transmission dynamics and risk factors in animal–human transmission. This review could explore the factors that contribute to the emergence of Mpox and other zoonotic diseases, such as the impact of habitat fragmentation, the role of animal trade and consumption, and the impact of climate change on animal populations. (4) A systematic review of the economic impact of Mpox outbreaks on affected communities. This review could examine the costs associated with control and response efforts, as well as the indirect economic consequences of the outbreak, such as loss of income and productivity due to illness or quarantine measures. (5) A systematic review of Mpox diagnosis and treatment options. This review could explore the current diagnostic methods and their accuracy, as well as the efficacy and safety of various treatment options, such as antiviral drugs or immunomodulatory therapies, in Mpox patients. These are just a few examples of potential systematic review topics related to Mpox. The actual research questions and methods would need to be further refined and tailored based on the available literature and research gaps in the field.
Prospect
Based on the answers of these four questions, it is not difficult to see that ChatGPT really could give certain suggestions and reflections according to our language context. And also, the answers are usually not restricted to one point. However, as stated by ChatGPT itself, as an AI language model, it cannot speculate or provide a definitive answer as to the cause of the sudden emergence and transmission of Mpox or the future trends in confirmed cases. The answers for these questions are quite generic and do not critically analyze differences among studies. Moreover, another concern is that ChatGPT could falsify data. There are a lot of inaccurate information online about Mpox, and it could use that information to produce inaccurate results. Howard et al. asked ChatGPT for antimicrobial advice in eight hypothetical infection scenario-based questions. They have found that ChatGPT could recognize natural language appropriately without spelling and grammar mistakes. However, ChatGPT sometimes could provide dangerous advice about the antimicrobial contraindications and often miss clinical patient safety cues [8]. Therefore, we believe that although ChatGPT has the potential to significantly improve the work of infectious disease doctors, it cannot fully replace the knowledge and expertise of them. Meanwhile, it is worth noting that OpenAI recently reported the further development of GPT-4, a large-scale, multimodal model that could accept image and text inputs and produce text outputs. As rash is almost always present in Mpox cases, GPT-4 may serve as an important adjuvant tool to help the differential diagnosis of Mpox rash in future.
In addition, we also found that ChatGPT is capable of conceiving innovative systematic review ideas in the field of Mpox. After checking all the five potential directions it provided based on Pubmed database, it was surprisingly found that these research directions were really important topics that require further summarization. Take the third recommendation as an example, ChatGPT suggested a systematic review of Mpox transmission dynamics and risk factors in animal–human transmission. Indeed, the mode of Mpox transmission is still controversial. Apart from the well-recognized sexual transmission, a mounting number of studies suggest the transmission route through the saliva and air. Moreover, Gupta and colleagues also investigated the application of ChatGPT to aid the development of unpublished systematic review topics on cosmetic plastic surgery. Their findings showed that ChatGPT had a 65% accuracy in conceiving new systematic review topics [8]. In fact, ChatGPT is a powerful tool for language translation, knowledge summarization, and even draft generation. These functions are important for academic activities [9]. Nevertheless, many journals and institutions, especially educational institutions believe that the use of ChatGPT in scientific writing could trigger a series of ethical concerns and research misconduct. In view of this, many institutions have decided to ban the use of ChatGPT in scientific writing [10]. However, in our opinion, for the field of scientific research, it is not a question of refusal to use ChatGPT, but a question of how to avoid the wrong use. Since this system is not mature enough, it is necessary to insist on manual confirmation of information when using it. And also in future, various technologies may be used to solve this problem.
All in all, in recent years, AI has begun to permeate and reform the field of medicine and is constantly expanding. The appearance of ChatGPT is only represented as a beginning, not an end. ChatGPT is a double-edged sword, as it improves knowledge summarization and innovation efficiency on the one hand and trigger ethical concerns on the other hand. Therefore, production of practice guidelines and consensus statements for the use of ChatGPT in scientific activities is of paramount importance for the medical community.
Acknowledgements
We thank Dr Noor from Duke University for her linguistic assistance during the preparation of this manuscript.
Author contributions
KC participated in the Conceptualization, Methodology, Data Curation, Formal analysis, Resources, Investigation, Writing of the Original Draft, and Writing, Reviewing, & Editing of the manuscript; YH participated in the Conceptualization, Methodology, Data Curation, Formal analysis, Resources, and Investigation; CL participated in the Conceptualization, Methodology, Data Curation, Formal analysis, and Resources; RX participated in the Conceptualization, Methodology, Data Curation, Formal analysis, and Resources; YL participated in the Conceptualization, Methodology, Data Curation, Formal analysis, and Resources; SG participated in the Conceptualization, Methodology, Data Curation, Formal analysis, Resources, Investigation, Writing of the Original Draft, and Writing, Reviewing, & Editing of the manuscript; HW participated in the Conceptualization, Methodology, Data Curation, Formal analysis, Resources, Investigation, Writing of the Original Draft, and Writing, Reviewing, & Editing of the manuscript.
Data Availability
Not applicable.
Declarations
Competing interest
The authors declare no conflict of interest.
Ethical Approval
This study does not include any individual-level data and thus does not require any ethical approval.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Kunming Cheng, Yongbin He, Cheng Li, and Ruijie Xie have contributed equally to this work and share the first authorship.
Contributor Information
Shuqin Gu, Email: shuqin.gu@duke.edu.
Haiyang Wu, Email: wuhaiyang2021@tmu.edu.cn.
References
- 1.Lancet The. Monkeypox: a global wake-up call. Lancet. 2022;400(10349):337. doi: 10.1016/S0140-6736(22)01422-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cheng K, Zhou Y, Wu H. Bibliometric analysis of global research trends on monkeypox: Are we ready to face this challenge? J Med Virol. 2023;95(1):e27892. doi: 10.1002/jmv.27892. [DOI] [PubMed] [Google Scholar]
- 3.Cheng K, Guo Q, Shen Z, et al. Global research trends on four orthopoxviruses threatening human health: Monkeypox is a neglected branch which deserves more attention. Int J Surg. 2022;105:106846. doi: 10.1016/j.ijsu.2022.106846. [DOI] [PubMed] [Google Scholar]
- 4.Zeeshan HM, Rubab A, Dhlakama H, et al. Global Research Trends on Monkeypox Virus: A Bibliometric and Visualized Study. Trop Med Infect Dis. 2022;7(12):402. doi: 10.3390/tropicalmed7120402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Castelvecchi D. Are ChatGPT and AlphaCode going to replace programmers? Nature. 2022. Epub ahead of print. [DOI] [PubMed]
- 6.Mbakwe AB, Lourentzou I, Celi LA, et al. ChatGPT passing USMLE shines a spotlight on the flaws of medical education. PLOS Digit Health. 2023;2(2):e0000205. doi: 10.1371/journal.pdig.0000205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Howard A, Hope W, Gerada A. ChatGPT and antimicrobial advice: the end of the consulting infection doctor? Lancet Infect Dis. 2023;S1473–3099(23):00113–115. doi: 10.1016/S1473-3099(23)00113-5. [DOI] [PubMed] [Google Scholar]
- 8.Gupta R, Pande P, Herzog I, et al. Application of ChatGPT in Cosmetic Plastic Surgery: Ally or Antagonist. Aesthet Surg J. 2023:sjad042. [DOI] [PubMed]
- 9.Graf A, Bernardi RE. ChatGPT in Research: Balancing Ethics, Transparency and Advancement. Neuroscience. 2023;S0306–4522(23):00079–89. doi: 10.1016/j.neuroscience.2023.02.008. [DOI] [PubMed] [Google Scholar]
- 10.Macdonald C, Adeloye D, Sheikh A, Rudan I. Can ChatGPT draft a research article? An example of population-level vaccine effectiveness analysis. J Glob Health. 2023;13:01003. doi: 10.7189/jogh.13.01003. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Not applicable.