ChatGPT in Urology: Bridging Knowledge and Practice for Tomorrow’s Healthcare,a Comprehensive Review

Abstract

Background:

Among emerging AI technologies, Chat-Generative Pre-Trained Transformer (ChatGPT) emerges as a notable language model, uniquely developed through artificial intelligence research. Its proven versatility across various domains, from language translation to healthcare data processing, underscores its promise within medical documentation, diagnostics, research, and education. The current comprehensive review aimed to investigate the utility of ChatGPT in urology education and practice and to highlight its potential limitations.

Methods:

The authors conducted a comprehensive literature review of the use of ChatGPT and its applications in urology education, research, and practice. Through a systematic review of the literature, with a search strategy using databases, such as PubMed and Embase, we analyzed the advantages and limitations of using ChatGPT in urology and evaluated its potential impact.

Results:

A total of 78 records were eligible for inclusion. The benefits of ChatGPT were frequently cited across various contexts. In educational/academic benefits mentioned in 21 records (87.5%), ChatGPT showed the ability to assist urologists by offering precise information and responding to inquiries derived from patient data analysis, thereby supporting decision making; in 18 records (75%), advantages comprised personalized medicine, predictive capabilities for disease risks and outcomes, streamlining clinical workflows and improved diagnostics. Nevertheless, apprehensions were expressed regarding potential misinformation, underscoring the necessity for human supervision to guarantee patient safety and address ethical concerns.

Conclusion:

The potential applications of ChatGPT hold the capacity to bring about transformative changes in urology education, research, and practice. AI technology can serve as a useful tool to augment human intelligence; however, it is essential to use it in a responsible and ethical manner.

Introduction

Artificial intelligence (AI) is revolutionizing medical care, especially in fields such as urology where neural processing systems are evolving rapidly.¹ ChatGPT, a standout AI technology, demonstrates vast potential across multiple sectors, including healthcare. Its capacity for medical documentation, diagnostics, research, and education suggests significant advancements in urology, a field already benefiting from rapid technological progress and improved patient care through the integration of AI and natural language processing.²

Launched in November 2022, ChatGPT has marked a significant advancement in AI-based language models, trained on extensive multilingual datasets to deliver responses closely mirroring human interaction. Despite its promising capabilities, the application of ChatGPT in healthcare has sparked debate, highlighting concerns over dataset biases, accuracy, and the potential spread of misinformation, known as “hallucination.”³

This article seeks to explore the untapped potential of ChatGPT within urological practice. It focuses on uncovering its benefits, understanding the inherent challenges, and projecting its future implications in a field where research is still emerging. Through this exploration, we aim to contribute to a deeper understanding of ChatGPT’s role in enhancing urological care and navigating the complexities of its implementation in healthcare.

Materials and Methods

Search strategy and inclusion criteria

The primary objective of this comprehensive review was to evaluate the advantages and disadvantages of incorporating ChatGPT into urological practice and to ascertain its prospective influence on the discipline. To identify pertinent literature, comprehensive database searches were performed in PubMed/MEDLINE and EMBASE. The search was tailored to extract peer-reviewed journal articles encompassing primary research studies and review articles that explicitly explored the integration of AI in healthcare, with a specific focus on urology.

The inclusion criteria were designed to capture studies that assessed the implementation of ChatGPT in urological settings, examining both its benefits and limitations, as well as its potential to alter clinical practice. Conversely, exclusion criteria were established to omit non-English language publications, discussions of ChatGPT in unrelated contexts, and sources from nonscholarly platforms, including newspapers, websites, and magazines.

The search strategy for PubMed/MEDLINE was finalized on November 17, 2023, utilizing a combination of search terms designed to optimize retrieval of relevant records: “chatgpt”[All Fields] AND (“urology”[MeSH Terms] OR “urology”[All Fields] OR “urological procedures”[MeSH Terms]).

The search strategy yielded a total of 78 records from PubMed/MEDLINE and EMBASE. A rigorous title and abstract screening process led to the exclusion of 52 duplicate records, and two articles were excluded because they had no outcomes of interest. This process culminated in the selection of 24 records deemed suitable for inclusion in this review (Fig. 1). The final determination of relevant studies was conducted by two independent researchers (CS, NT), with any discrepancies resolved through consultation with a third researcher, serving as an adjudicator.

FIG. 1.

PRISMA flow diagram of study screening and selection.

To ensure a comprehensive review and mitigate bias, the reference lists of included studies were scrutinized for additional relevant articles that may have been missed during the initial search. To safeguard the integrity of the review and avoid potential conflicts of interest, a thorough examination was performed to confirm the absence of financial or other biases among the authors. Furthermore, all content was generated authentically by the review authors, with diligent citation and acknowledgment of any directly quoted or paraphrased sources.

Categorization of benefits and risks of ChatGPT

The benefits and potential applications of ChatGPT were systematically categorized into distinct domains to comprehensively assess its utility within the healthcare landscape. These categories included:

Educational benefits in urology education: This encompassed the generation of diverse clinical vignettes, customization of clinical cases tailored to individual student needs with immediate feedback, and improvements in communication skills within healthcare education settings.

Benefits in academic/scientific research incorporating text generation, summarization, translation, and literature review capabilities, particularly relevant in facilitating scientific research endeavors.

Benefits in urology practice: Contributions toward personalized medicine, diagnosis, treatment strategies, lifestyle recommendations based on personalized traits, and streamlined documentation/report generation within urology practices.

Conversely, the risks and concerns associated with ChatGPT were meticulously categorized to ascertain potential challenges and ethical considerations. These categories encompassed:

Ethical issues: Including risks of bias, discrimination based on training data quality, and concerns regarding plagiarism.

Hallucination: Referring to the generation of scientifically inaccurate yet plausible content.

Content-related concerns: Reflecting issues of over-detailing, redundancy, and excessiveness within the generated content.

Legal, interpretability, and referencing issues: Encompassing challenges related to copyright, authorship, interpretability, referencing accuracy, and potential academic fraud.

Incorrect content: Focusing on the risks associated with generating inaccurate or misleading content, contributing to the spread of misinformation.

This categorization aimed to thoroughly outline both the potential benefits and inherent risks associated with the utilization of ChatGPT in the context of urology, serving as a comprehensive framework for analysis within this systematic review.

Results

Our review identified and analyzed 24 studies that detailed the advantages and challenges associated with the integration of ChatGPT in urology, using a diverse array of research designs from observational studies to comparative analyses and narrative reviews (Table 1).

Table 1.

Summary of Available Literature on the Role of ChatGPT in Urology

Study/year	Study design	Primary aim	Results
Adhikari⁴ (2023)	Literature review	The ethical implications of ChatGPT in urology.	ChatGPT in urology has potential to improve patient care. Ethical and legal concerns regarding privacy and security.
Caglar⁵ (2023)	Preprint	Evaluates precision and consistency of ChatGPT responses in pediatric urology.	ChatGPT’s performance in pediatric urology questions assessed for reliability and accuracy.
Caglar⁶ (2023)	Observational study	ChatGPT’s performance in pediatric urology questions assessed for reliability and accuracy.	ChatGPT provided satisfactory responses to inquiries related to pediatric urology.
Cakir⁷ (2023)	Observational study	Analyzing ChatGPT’s knowledge about urolithiasis.	Applying ChatGPT in urology clinics can help patients understand urolithiasis diagnosis and treatment.
Cocci⁸ (2023)	Comprehensive comparative study	Evaluating ChatGPT responses on urologic cases compared with a urologist.	52% of ChatGPT responses were deemed appropriate. ChatGPT had more appropriate responses for nononcology conditions.
Coskun⁹ (2023)	Case Report	ChatGPT’s performance in providing prostate cancer information.	ChatGPT’s performance in providing patient information on prostate cancer was suboptimal.
Davis¹⁰ (2023)	Observational study	Evaluate the appropriateness and readability of natural language processor-generated responses to urology-related medical inquiries.	77.8% of responses from ChatGPT were deemed appropriate. Missing information was the most common reason for low scores.
Deebel¹¹ (2023)	Observational study	The study aimed to see how well ChatGPT could answer test questions from the AUA study program and to check if it could explain its answers simply.	ChatGPT performed better on the 2021 AUA SASP question set compared to the 2022 set. ChatGPT answered many high-level questions correctly.
Eppler¹² (2023)	Prospective, cross-sectional survey	Potential use of ChatGPT in medicine.	Recognition of ChatGPT’s potential in research, education, and clinical care.
Gabriel¹³ (2023)	Observational study	Examining the use of ChatGPT as a patient education tool in robot-assisted radical prostatectomy.	ChatGPT assessed as patient education tool in robot-assisted radical prostatectomy. ChatGPT provides accurate and helpful information for patients.
Huynh¹⁴ (2023)	Observational study	Evaluated its performance on the Self-assessment Study Program for urology.	ChatGPT performs poorly on the 2022 Self-assessment Study Program for Urology. Model undertraining and lack of urological content may contribute to low accuracy.
Javid¹⁵ (2023)	Narrative review	Potential impact of AGI and LLMs on urology, including diagnosis and education.	AGI and LLMs have the potential to enhance patient care and clinical outcomes.
Kianian¹⁶ (2023)	Comparative study	The study aimed to find out if ChatGPT can create easier-to-understand health information about preventing and treating kidney stones compared with what is already on the internet.	ChatGPT-generated health information is easier to understand.
Kim¹⁷ (2023)	Observational study	Evaluated changes in patient perceptions regarding AI before and after reading a ChatGPT-written explanation.	The negative perception change among urolithiasis patients after receiving the explanatory note provided by the AI chatbot program was observed.
Manolitsis¹⁸ (2023)	Comparative study	Investigating the use of a custom ChatGPT model. Comparing the custom model with the general model.	ChatGPT assists urologists in obtaining accurate medical information. Custom-trained model provides precise answers about urologic subjects.
Musheyev¹⁹ (2023)	Brief Correspondence	Characterize the quality of responses to top urological cancer from AI chatbots.	AI chatbot responses had moderate-to-high information quality. Understandability and actionability of responses were moderate.
Nedbal²⁰ (2023)	Narrative review	Explores ChatGPT’s impact, benefits, challenges, and future scenarios in urology.	ChatGPT has the potential to contribute to precision medicine and tailored practice.
Okui²¹ (2023)	Observational study	Investigated factors predicting post-onabotulinum A injection PVR volume.	An SVM model incorporating daytime frequency, frailty, and voiding efficiency improved PVR volume prediction.
Pan²² (2023)	Cross-sectional study design	Quality of consumer health information. Understandability and actionability of information.	AI chatbots provide high-quality, accurate information about common cancers. ChatGPT is reliable in answering medical questions accurately.
Schuppe²³ (2023)	Case Report	Rare case of atypical Nelson syndrome reported.	This case report’s outlining, drafting, and conclusions were written in part by or with the support of ChatGPT.
Shah²⁴ (2023)	Comprehensive comparative study	ChatGPT vs. UCF for men’s health education materials comparison.	ChatGPT provided longer responses with more complex words. ChatGPT had poorer readability compared with UCF.
Szczesniewski²⁵ (2023)	Observational study	Assessing quality of information provided by ChatGPT.	ChatGPT provides well-balanced information on urological diseases. The quality of treatment information is moderate.
Talyshinskii²⁶ (2023)	Literature review	The use of ChatGPT in urology.	ChatGPT can save healthcare providers time through optimized automation. ChatGPT improves communication between physicians and patients.
Whiles²⁷ (2023)	Observational study	Evaluation conducted on ChatGPT responses in urology healthcare context.	60% of ChatGPT responses were deemed appropriate. Only 7 out of 13 topics met the cutoff score for good-quality content.

AGI, Artificial generative intelligence; LLM, Large language models; OAB: overactive bladder; PVR, post-void residual; SVM, support vector machine.

Educational and academic insights

The educational utility of ChatGPT was widely recognized. For instance, studies by Okui et al. and Kianian et al. demonstrated its capacity for enhancing data analysis and education in urology,^21,29 whereas Schuppe et al. and Eppler et al. acknowledged its benefits in expediting medical literature review processes.^5,16 Similarly, Adhikari et al. and Talyshinskii et al. shed light on ChatGPT’s role in synthesizing research, aiding in academic writing, and generating literature summaries.^12,23 Notably, these studies collectively suggest that ChatGPT has significant potential in enriching academic discourse and streamlining the research methodology (Table 2).

Table 2.

Current Scenarios to Use chatGPT in Urology, Their Potential Advantages and Shortcomings

Study/year	Study design	Educational/academic benefits	Benefits in clinical practice	Ethical/legal concerns	Hallucination	Interpretability issues /incorrect content
Adhikari⁴ (2023)	Literature review	Supports education, offering insights and reasoning in medical exams. Helps students with research questions and data cleaning. Synthesizes research, assists in writing academic articles. Generates scientific literature summaries, saving time for academics. Aids patient education for informed decisions and anxiety reduction.	Automates preoperative and postoperative care for efficiency. Assists in diagnostics by analyzing patient history and symptoms. Streamlines daily tasks by managing low complexity activities. Provides dosage recommendations, drug interaction info, treatment options, and potential complications. Potential to reduce clinician burnout by streamlining workflow.	Ethical and legal concerns include privacy, security, and accountability in patient data. Addressing bias in AI algorithms and handling liability and intellectual property rights are crucial.	Ensuring human oversight is crucial to mitigate the risk of ChatGPT providing unreliable medical information.	ChatGPT’s reliance on pattern recognition may lead to unreliable responses in unique medical scenarios, requiring human supervision to avoid legal issues.
Caglar⁵ (2023)	Preprint	N/A	N/A	ChatGPT might have limitations in understanding ethics, confidentiality, and generating sensitive responses.	N/A	ChatGPT’s accuracy >80% for andrology queries but faces limitations due to the evolving nature of andrology literature, hindering access to the latest information.
Caglar⁶ (2023)	Observational study	Provides medical information with a high accuracy rate, aiding in patient education and support.	Can serve as a supplementary tool for healthcare professionals.	N/A	No question was answered completely wrong.	Reproducibility rates ranging from 93.75% to 96.67% for different topics ChatGPT provides satisfactory responses in pediatric urology.
Cakir⁷ (2023)	Observational study	ChatGPT’s high accuracy in answering medical questions can be a valuable educational tool for both patients and healthcare professionals.	Implementing ChatGPT in urology practice under urologist supervision can assist patients and their families in better understanding the diagnosis and treatment of urolithiasis.	N/A	N/A	ChatGPT accurately and satisfactorily responded to over 95% of the urolithiasis-related questions.
Cocci⁸ (2023)	Comprehensive comparative study	Healthcare professionals prefer ChatGPT for its quality and empathetic responses. ChatGPT enhances health literacy with accessible information.	ChatGPT’s interactive medical information delivery might boost patient satisfaction. It assists in initial diagnosis and educating patients in nononcology urology cases. ChatGPT aids healthcare professionals by providing workload-reducing information.	N/A	N/A	Only 52% of ChatGPT responses were deemed appropriate. Patients sharing incorrect or insufficient information pose a risk of receiving potentially harmful inaccurate information.
Coskun⁹ (2023)	Case Report	ChatGPT can assist in the dissemination of prostate cancer information, potentially improving patient understanding.	Interface is user-friendly and easily accessible for health queries, adaptable to education levels. ChatGPT’s responses are not a replacement for professional advice but provide initial information for patient discussions.	Ethical and legal concerns, especially about privacy and data security, should be noted. Interactions with ChatGPT could be recorded and stored.	N/A	ChatGPT showed limitations in providing accurate and high-quality patient information on prostate cancer. Lack of personalization can lead to generalized or potentially inaccurate information.
Davis¹⁰ (2023)	Observational study	N/A	N/A	Patients and doctors using these tools should proceed cautiously. The empathetic nature of chatbot responses, while convincing, might pose risks if not entirely accurate.	N/A	Emergency and oncology-related questions frequently had inappropriate responses. Lack of essential information was the main reason for low suitability scores. Responses were at a college-level readability.
Deebel¹¹ (2023)	Observational study	ChatGPT could be valuable in educating medical students and residents. Learners can simulate small group learning by using this tool to ask questions and discuss diagnosis or treatment concepts.	ChatGPT offers logical and coherent justifications for answer choices, assisting in medical education and decision making.	N/A	N/A	The model’s ability to discern question order was high, yet it still made errors in content accuracy.
Eppler¹² (2023)	Prospective, cross-sectional survey.	81.6% of participants agree that ChatGPT has a potential role in academic brainstorming. ChatGPT are utilized by 47.7% in academic practice Perceived benefits in urology: - Medical education (66.8%). - Research (70.2%).	N/A	The most frequent ethical concerns were plagiarism (51.9%).	Artificial hallucination (46.1%).	N/A
Gabriel¹³ (2023)	Observational study	ChatGPT’s accurate patient responses aid in clinical research by highlighting common concerns and knowledge gaps, enhancing understanding.	ChatGPT offers precise information on urological procedures and complications in response to patient inquiries.	N/A	N/A	78.6% of ChatGPT’s quoted figures matched the BAUS information leaflet. 92.9% of ChatGPT’s responses were globally accurate. ChatGPT, mostly accurate, had notable errors, like misrepresenting infertility risks postrobotic radical prostatectomy.
Huynh¹⁴ (2023)	Observational study	ChatGPT aids urology trainees and physicians in education, displaying promise in licensing exams with high concordance and insightful explanations.	ChatGPT can support clinical decision making.	The study did not focus on legal issues but emphasized the risk of depending on AI for medical decisions.	N/A	Provides consistent justifications, aiding learning. ChatGPT responses showed lengthy, sometimes redundant explanations. Text complexity suits graduate-level understanding.
Javid¹⁵ (2023)	Narrative review	Personalized patient educational materials based on individual factors like age, medical history, and information needs. Provides accurate, easy-to-understand information to improve patient adherence to treatment plans. Enhances communication between patients and healthcare providers.	ChatGPT demonstrated a high ability to provide responses consistent with strong evidence-based recommendations in most cases.	Current GPT models lack HIPAA compliance, cautioning clinicians against using them for patient information until more secure versions are accessible.	Despite the model’s confidence, potential misinformation can occur Addressing hallucinations necessitates human interaction for validation and feedback.	Varying accuracy in ChatGPT responses might risk misleading education without proper authorization and guidance.
Kianian¹⁶ (2023)	Comparative study	ChatGPT is easy to use and accessible to everyone. ChatGPT can generate more comprehensible information.	ChatGPT, which supports many different dialects and languages, could benefit health institutions worldwide.	N/A	N/A	ChatGPT scored 62 (good) for stone prevention and 52 (good) for treatment on the DISCERN Instrument, surpassing other sources.
Kim¹⁷ (2023)	Observational study	Chatbots serve as high-quality medical databases, offering evidence-based recommendations.	ChatGPT aid in information flow, summarizing clinical data, managing patient education, and scheduling. ChatGPT offers detailed lifestyle advice for disease prevention, improving patient comprehension.	Upon checking references, some were invalid or unavailable, possibly impacting decision making, and causing ethical and legal concerns.	N/A	Patients with lower education levels expressed more negative responses. Patients might doubt AI-generated notes despite understanding the content, showing interpretability issues.
Manolitsis¹⁸ (2023)	Comparative study	ChatGPT improves medical training with interactive, up-to-date content. Training with specific medical datasets ensures credible education. AI enhances accuracy in urological research data extraction.	ChatGPT offered urologists accurate information in line with the latest EAU guidelines, enriching clinical learning practice.	N/A	The study does not discuss hallucinations using ChatGPT models in urological practice or research.	N/A
Musheyev¹⁹ (2023)	Brief Correspondence	Chatbots have a minimal risk of misinformation, boosting their educational reliability.	N/A	Ethical use demands transparency on AI chatbots’ limitations in medical advice.	N/A	Chatbots offer accurate, high-quality information on urological m alignancies and consumer health. Information’s understandability and actionability scored moderately.
Nedbal²⁰ (2023)	Narrative review	ChatGPT can reach a larger population with different languages. It can provide access to medical information for underserved populations.	Reducing administrative workload: Patient inquiries, administrative paperwork Automating routine tasks for better focus on critical patient care aspects. Assisting in decision making for urologists: - Aiding in generating differential diagnoses and suggesting treatment options. - Providing assistance in consultations, managing drug interactions, and noting side effects.	Concerns about patient privacy and data security need to be addressed for successful integration.	To reduce artificial hallucinations in chatbots, comprehensive training with diverse datasets is crucial. Implementing methods like human assessment or anomaly detection can help monitor and detect such occurrences.	- Challenges in validation and regulation to prevent misinformation in ChatGPT’s responses. - Limitations in accuracy of ChatGPT’s medical advice, especially in emergency urological cases according to some studies.
Okui²¹ (2023)	Observational study	ChatGPT in research, providing a new educational perspective on data analysis.	By employing ChatGPT, the research underscores the importance of considering multiple intersecting factors to forecast PVR volume accurately. These findings enrich our comprehension of onabotulinum A injection treatment for OAB and easily accessible information.	N/A	N/A	N/A
Pan²² (2023)	Cross-sectional study design	N/A	Chatbots may enhance clinical outcomes more effectively than other online sources by encouraging users to seek personalized medical attention.	N/A	N/A	ChatGPT generating the longest text. chatGPT 70 % Understandability and actionability of information.
Schuppe²³ (2023)	Case Report	ChatGPT in drafting and editing can save researchers time and offer new insights. ChatGPT-assisted case reports can help in identifying discussion points and clarifying language for medical literature.	N/A	Potential for ChatGPT to fabricate references, necessitating verification to maintain research integrity.	N/A	ChatGPT struggled with providing accurate references and citations.
Shah²⁴ (2023)	Comprehensive comparative study	ChatGPT-a was more comprehensive than UCF, indicating extra educational value.	ChatGPT helps personalize education and saves clinicians’ time on patient messages.	N/A	N/A	ChatGPT offered longer, more complex responses. ChatGPT had lower readability than UCF. Adjusted ChatGPT responses (ChatGPT-a) improved readability. ChatGPT’s stochastic nature leads to varied results over time.
Szczesniewski²⁵ (2023)	Observational study	ChatGPT’s user-friendly information delivery helps in understanding urological diseases and treatments.	ChatGPT could help streamline healthcare professionals’ daily tasks and decision making.	Unvalidated testing against clinical guidelines might lead to legal repercussions if AI advice proves incorrect or harmful.	Did not report hallucinations.	The study noticed bias in the information concerning BPH pathology.
Talyshinskii²⁶ (2023)	Literature review	ChatGPT enables examining text and identifying promising research ideas by clarifying problem-solving issues needing scientific analysis.	Encouraged seeking expert medical opinions for further clarification. Advised on postoperative relief by identifying alarming symptoms. Offers robust clinical suggestions and concise reports based on its comprehensive database. Enables real-time documentation assistance for physicians due to its extensive expertise.	ChatGPT face accidental plagiarism risks and biases, emphasizing the need for their responsible ethical conduct.	ChatGPT may display signs of this phenomenon due to its training with vast amounts of unsupervised data.	ChatGPT provided nonexistent or unrelated references to the study.
Whiles²⁷ (2023)	Observational study	Encourages interactive learning by simulating one-on-one tutoring sessions. Can generate diverse educational content, including explanations, examples, and quizzes. ChatGPT’s ability to generate human-like responses may help in simplifying complex medical information for patient understanding.	ChatGPT’s information was considered suitable in 59.0%.	Need for caution when using ChatGPT or similar chatbots for healthcare-related information by patients or healthcare providers.	N/A	In 25% of assessments, ChatGPT provided notably varied answers for the same query, with at least one response differing significantly in appropriateness from the rest.

Clinical practice applications

Regarding clinical practice, a set of studies, including those by Coskun et al. and Kim et al., indicated that ChatGPT serves as a robust source of information, which can be instrumental in assisting healthcare professionals with patient inquiries and medical education.^9,27 Manolitsis et al. also observed that ChatGPT provides information consistent with EAU guidelines, further supporting its role in enhancing clinical learning practices.¹⁷ These findings suggest that ChatGPT could be a valuable asset in both patient and clinician education, provided the content is verified for accuracy (Table 2).

Potential risks and ethical concerns

However, the integration of ChatGPT in urology is not without its risks. Caglar et al. raised concerns about ethical and privacy issues; ChatGPT might have limitations in understanding ethics, confidentiality, and generating sensitive responses,²⁵ and Szczesniewski et al. noted the potential for misinformation, highlighting the importance of human oversight.²⁰ Moreover, Schuppe and colleagues identified challenges related to content verification, such as the accuracy of references.¹⁶ These studies collectively underscore the imperative for rigorous content validation to mitigate the dissemination of misinformation and ensure ethical usage (Table 2).

In summary, the benefits of ChatGPT in urology are multifaceted, enhancing both education and clinical practice. However, these advantages are tempered by potential risks, particularly in the realm of misinformation, which necessitates a cautious and supervised approach to integration. The comparative analysis of the studies reveals a consensus on the utility of ChatGPT, yet also aligns in cautioning against its unverified use.

Discussion

The incorporation of AI, particularly ChatGPT, within urology presents substantial promise and transformative potential. The collective evidence from these studies underscores ChatGPT’s adeptness in furnishing precise information and addressing inquiries arising from patient data analysis, thereby augmenting decision-making processes within the realm of urology. Significantly, it has contributed to the realms of personalized medicine, predictive abilities concerning disease risks and outcomes, the streamlining of clinical workflows, improvements in diagnostics, and advancements in documentation processes.^28
–30 Remarkably, this proficiency spans across various subdomains within urology, particularly in oncology, functional urology, and general urology.

Noteworthy is its demonstrated potential in determining prognoses and guiding decisions related to adjuvant therapy, indicating its applicability in surgical planning, diagnostic methodologies, diverse treatment modalities, and subsequent patient outcomes.

Educational enrichment through AI

ChatGPT’s role in urological education resonates with a call for innovation in medical learning.^{5,12,16,21,23,29} The findings from the review advocate for its implementation as a tool that can amplify the quality of medical education. It can act as a bridge, connecting intricate medical concepts to the practical necessities of clinical education, thereby enhancing the comprehension of complex urological diseases for both trainees and seasoned practitioners. The literature presents a consensus on the beneficial use of ChatGPT for educational purposes, yet it also calls for a strategic approach to its integration, ensuring it complements rather than replaces traditional learning methods.

Ethical imperatives in the age of AI

As with any profound technological advancement, the adoption of ChatGPT carries with it ethical imperatives that must be addressed.^{5,9,11

–16,20,23,25
–27} Studies underscore the risks associated with potential misinformation and privacy concerns, suggesting that although ChatGPT has the potential to be a powerful supporter in urological practice, it also poses risks that warrant careful consideration. The discussion extends beyond the technical capabilities of ChatGPT to the heart of ethical medical practice. Ensuring the ethical deployment of ChatGPT involves maintaining vigilance over its outputs, safeguarding patient privacy, and consistently validating the veracity of the information provided.

A comparative lens reveals a nuanced understanding of ChatGPT’s applications. Although studies such as those by Whiles et al. and Coskun et al. concur on the utility of ChatGPT’s interface in patient information dissemination, they also caution against a one-size-fits-all approach, given the individual needs of patients.^9,11 Similarly, the concerns of content accuracy and reliability echo across studies, underscoring the need for a standardized framework of oversight to ensure the responsible use of ChatGPT in urological settings.

The future of AI in urology must be navigated with a responsible outlook, balancing the benefits against potential inaccuracies and ethical concerns. The rapidly evolving AI landscape calls for a dynamic approach to research and implementation. As such, ongoing vigilance and adaptation are essential, ensuring that AI tools such as ChatGPT are utilized to their full potential while upholding the highest standards of patient care and safety.

Strengths and limitations

The present review serves as an initial and concise overview of ChatGPT’s applicability within urology education, research, and practice. Nonetheless, it is imperative to consider several limitations when interpreting the findings. First, the variability in the quality of the included records might pose challenges to the generalizability of the outcomes. Second, the exclusion of non-English records could introduce selection bias, potentially limiting the comprehensiveness of the review. These limitations collectively urge careful consideration and temperance in extrapolating the conclusions drawn from this review.

Conclusion

In conclusion, although ChatGPT stands at the frontier of technological advancement in urology, it is imperative that its integration is guided by a commitment to patient safety, data security, and ethical responsibility. Moving forward, the medical community should embrace the benefits of AI with a cautious optimism, ensuring that advancements in AI such as ChatGPT are aligned with the core values of medicine. Continuous evaluation, research, and development are crucial to navigate the evolving interface of AI and healthcare, ensuring that the potential of ChatGPT is fully realized in a manner that is both innovative and conscientious.

Footnotes

Authors’ Contributions

All authors contributed to the study conception and design. Conceptualization and the idea for the article: N.T.; Methodology: C.S. and N.T.; Formal analysis and investigation: C.S., G.P.A., S.R., A.A.M., and V.M.P.; Writing––original draft preparation: C.S.; Writing––review and editing: N.T. and O.T.; Supervision: O.T.

Declarations of Conflicts of Interest

The others authors declare that they have no conflicts of interest. ChatGPT was not used in the production of the article.

Data Availability Statement

Data supporting this comprehensive review are available in the original publications and preprints that were cited in the reference section. In addition, the analyzed data that were used during the current review are available from the author on reasonable request.

Ethics Statement

Not applicable.

Consent to Participate

Not applicable.

Author Disclosure Statement

Prof. O.T. is a consultant for Coloplast, Rocamed, Karl Storz, IPG Medical, AMBU, Quanta System.

Funding Information

This is an independent study and is not funded by any external body.

Abbreviations Used

References

Lee

, Bubeck

, Petro

. Benefits, limits, and risks of GPT-4 as an AI chatbot for medicine. N Engl J Med, 2023; 388(13):1233–1239; doi: 10.1056/NEJMsr2214184

Elali

, Rachid

. AI-generated research paper fabrication and plagiarism in the scientific community. Patterns, 2023; 4(3):100706.

Deng

, Lin

. The benefits and challenges of ChatGPT: An overview. FCIS, 2023; 2(2):81–83. Available from https://drpress.org/ojs/index.php/fcis/article/view/4465

Adhikari

, Naik

, Hameed

, Raghunath

, Somani

. Exploring the ethical, legal, and social implications of ChatGPT in urology. Curr Urol Rep, 2023; 25(1):1–8.

Caglar

, Yildiz

, Ozervarli

, et al. Assessing the performance of Chat Generative Pretrained Transformer (ChatGPT) in answering andrology-related questions. Urol Res Pract, 2023; 49(6):365–369.

Caglar

, Yildiz

, Meric

, Ayranci

, Gelmis

, Sarilar

, et al. Evaluating the performance of ChatGPT in answering questions related to pediatric urology. J Pediatr Urol, 2023.

Cakir

, Caglar

, Yildiz

, Meric

, Ayranci

, Ozgor

. Evaluating the performance of ChatGPT in answering questions related to urolithiasis. Int Urol Nephrol, 2023; 56(1):17–21.

Cocci

, Pezzoli

, Lo Re

, et al. Quality of information and appropriateness of ChatGPT outputs for urology patients. Prostate Cancer Prostatic Dis, 2023; 27(1):159–160.

Coskun

, Ocakoglu

, Yetemen

, Kaygisiz

. Can ChatGPT, an artificial intelligence language model, provide accurate and high-quality patient information on prostate cancer? Urology, 2023; 180:35–58.

10.

Davis

, Eppler

, Ayo-Ajibola

, et al. Evaluating the effectiveness of artificial intelligence-powered large language models application in disseminating appropriate and readable health information in urology. J Urol, 2023; 210(4):688–694.

11.

Deebel

, Terlecki

. ChatGPT performance on the American Urological Association Self-assessment Study Program and the potential influence of artificial intelligence in urologic training. Urology, 2023; 177:29–33.

12.

Eppler

, Ganjavi

, Ramacciotti

, Piazza

, Rodler

, Checcucci

, et al. Awareness and use of ChatGPT and large language models: A Prospective Cross-Sectional Global Survey in Urology. Eur Urol, 2023.

13.

Gabriel

, Shafik

, Alanbuki

, Larner

. The utility of the ChatGPT artificial intelligence tool for patient education and enquiry in robotic radical prostatectomy. Int Urol Nephrol, 2023; 55(11):2717–2732.

14.

Huynh

, Bonebrake

, Schultis

, Quach

, Deibert

. New artificial intelligence ChatGPT performs poorly on the 2022 Self-assessment Study Program for Urology. Urol Pract, 2023; 10(4):409–415.

15.

Javid

, Reddiboina

, Bhandari

. Emergence of artificial generative intelligence and its potential impact on urology. Can J Urol, 2023; 30(4):11588–11598.

16.

Kianian

, Carter

, Finkelshtein

, Eleswarapu

, Kachroo

. Application of Artificial Intelligence to patient-targeted health information on kidney stone disease. J Ren Nutr, 2023.

17.

Kim

, Tae

, Chang

, et al. Changes in patient perceptions regarding ChatGPT-written explanations on lifestyle modifications for preventing urolithiasis recurrence. Digit Health, 2023; 9:20552076231203940.

18.

Manolitsis

, Feretzakis

, Tzelves

, et al. Training ChatGPT models in assisting urologists in daily practice. Stud Health Technol Inform, 2023; 305:576–579.

19.

Musheyev

, Pan

, Loeb

, Kabarriti

. How well do artificial intelligence chatbots respond to the top search queries about urological malignancies? Eur Urol, 2023.

20.

Nedbal

, Naik

, Castellani

, Gahuar

, Geraghty

, Somani

. ChatGPT in urology practice: Revolutionizing efficiency and patient care with generative artificial intelligence. Curr Opin Urol, 2023; 34(2):98–104.

21.

Okui

, Ikegami

, Hashimoto

, Kouno

, Nakano

, Okui

. Predictive factors for high post-void residual volume in older females after onabotulinuma treatment for severe overactive bladder using a machine learning model. Cureus, 2023; 15(7):e42668.

22.

Pan

, Musheyev

, Bockelman

, Loeb

, Kabarriti

. Assessment of artificial intelligence chatbot responses to top searched queries about cancer. JAMA Oncol, 2023; 9(10):1437–1440.

23.

Schuppe

, Burke

, Cohoe

, Chang

, Lance

, Mroch

. Atypical Nelson Syndrome following right partial and left total nephrectomy with incidental bilateral total adrenalectomy of renal cell carcinoma: A Chat Generative Pre-Trained Transformer (ChatGPT)—Assisted Case Report and Literature Review. Cureus, 2023; 15(3):e36042.

24.

Shah

, Ghosh

, Hochberg

, et al. Comparison of ChatGPT and traditional patient education materials for men’s health. Urol Pract, 2023; 11(1):87–94.

25.

Szczesniewski

, Tellez Fouz

, Ramos Alba

, Diaz Goizueta

, García Tello

, Llanes González

. ChatGPT and most frequent urological diseases: Analysing the quality of information and potential risks for patients. World J Urol, 2023; 41(11):3149–3153.

26.

Talyshinskii

, Naik

, Hameed

BMZ

, et al. Expanding horizons and navigating challenges for enhanced clinical workflows: ChatGPT in urology. Front Surg, 2023; 10:1257191.

27.

Whiles

, Bird

, Canales

, DiBianco

, Terry

. Caution! AI Bot has entered the patient chat: ChatGPT has limitations in providing accurate urologic healthcare advice. Urology, 2023; 180:278–284.

28.

Gabrielson

, Odisho

, Canes

. Harnessing generative artificial intelligence to improve efficiency among urologists: Welcome ChatGPT. J Urol, 2023; 209(5):827–829. Available from https://www.embase.com/search/results?subaction=viewrecord&id=L2023909441&from=export

29.

Moher

, Liberati

, Tetzlaff

, Altman

, PRISMA Group . Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Med, 2009; 6(7):e1000097.

30.

Ang

, Choolani

, See

, Poh

. The rise of artificial intelligence: Addressing the impact of large language models such as ChatGPT on scientific publications. Singapore Med J, 2023; 64(4):219–221.