Attitudes and Awareness of Medical Students Toward Teleradiology and the Application of Artificial Intelligence in Diagnostic Radiology: A Cross-Sectional Study

Introduction

The field of radiology is undergoing a significant transformation driven by artificial intelligence (AI) and teleradiology. AI tools are now capable of detecting abnormalities in X-ray, computed tomography, and magnetic resonance imaging scans with high accuracy, while teleradiology platforms are expanding access to imaging services in remote areas. These innovations are reshaping diagnostic workflow, improving efficiency, and enhancing patient care, marking a pivotal shift in modern radiology.

As technology continues to revolutionize healthcare, AI and teleradiology have become indispensable in radiology. For instance, teleradiology, the use of technology and electronic communication improves access to remote healthcare services such as diagnosis, examination and medical consultaions. ¹

Understanding medical students’ perceptions of these technologies is crucial, as their attitude and knowledge will directly influence their adaptability to integrate them into future clinical practice. Without adequate awareness and training, future radiologists may struggle to adapt to an increasingly AI-driven and teleradiology.

Despite the growing role of AI in radiology, concerns persist regarding its potential to replace radiologists and alter job availability. Similarly, while telemedicine and teleradiology offer improved accessibility and efficiency, their long-term implications for radiology practice remain a topic of debate. Addressing these concerns early in medical education can help ensure that future radiologists are not only aware of technological advancements but also equipped with the skills and confidence to utilize them effectively.

AI has emerged as a critical component in modern medicine, particularly in radiology. AI can predict patient outcomes based on historical data, aiding in patient management and personalized treatment planning. Beyond diagnostics, AI contributes to administrative efficiency by automating tasks such as sorting images based on urgency, ultimately improving workflow and patient care. ² Similarly, teleradiology has revolutionized healthcare accessibility by enabling faster delivery of imaging results, leading to quicker diagnoses and timely interventions. ³

Despite these advantages, the integration of AI and teleradiology into clinical practice presents several challenges. One major hurdle is the significant financial investment required for infrastructure and training. The high cost of implementing AI-driven radiology solutions may make it difficult for numerous healthcare organizations, especially those with limited resources. Large datasets are also needed for the development and verification of AI algorithms, which raises difficulties within data security, privacy, and the potential for biases in diagnostic accuracy. Furthermore, effective interoperability with current medical systems, which might be technically challenging, is necessary for the incorporation of AI into workflow procedures. A further barrier to widespread utilization is healthcare professionals’ resistance, which stems from concerns regarding losing their jobs and their lack of trust in AI-generated diagnoses.

The utilization of AI in radiology presents a number of ethical and professional difficulties in addition to practical difficulties. The responsibility is an important concern: Who is ethically and legally responsible if an AI system misdiagnoses a patient? Concerns in transparency are also raised by the “black box” aspect of many AI algorithms, which makes it difficult for physicians to completely understand how an AI system makes a certain diagnosis. A heavy reliance on AI may also result in unemployment, which would limit radiologists’ independence in interpreting and validating medical images. Furthermore, since AI systems need huge amounts of patient data, which must be managed with strict privacy protections, patient permission and data security are both important.

The perspective that medical students see AI and teleradiology greatly influences how well-prepared they are for a highly technological healthcare industry. Students may be less likely to adopt new technologies if they believe AI could threaten their job security rather than be a tool to increase performance. Future radiologists are able to understand AI's role as a supporting instrument rather than an alternative for human expertise with the right instruction and training on AI applications in radiology. Medical schools can ensure that graduates are able to use technology effectively in their future jobs by incorporating AI-related courses and practical experience with AI-driven diagnostic tools into their educational programs.

While several studies have explored the impact of AI on radiology, research on medical students’ perceptions remains limited. Most existing literature focuses on the perspectives of practicing radiologists’ and radiology trainees’ attitudes toward AI and teleradiology, with limited emphasis on how medical students who represent the future health workers perceive these technological advancements. Additionally, while AI's technical capabilities in image interpretation are well-documented, fewer studies have examined its impact on medical education and professional training and career decision making. Barriers such as insufficient AI training, doubt about AI-driven diagnoses, and concerns over professional autonomy further complicate its integration into clinical practice. The evolving integration of AI and teleradiology into radiology raises critical questions about how these advancements shape students’ career choices and perceptions of job security. This study aims to bridge this gap by analyzing the perspectives of medical students at Imam Abdulrahman University (IAU), shedding light on their understanding, concerns, and expectations regarding AI and teleradiology in radiology.

Materials and Methods

Results

The participants expressed a range of opinions regarding the role of AI and teleradiology in radiology. Figure 1 illustrates the selection process of study participants. Out of 448 eligible medical students, all were invited to participate. A total of 292 students completed the survey, resulting in a response rate of 65.2%. Nonparticipation (34.8%) was attributed to factors such as lack of interest, academic workload, or technical barriers. A significant proportion, 56.1%, believed that AI would enhance radiologists’ capabilities and improve their efficiency. In contrast, 23.4% felt that AI could completely replace radiologists in the current era, while 42.5% thought AI alone could reduce the number of radiologists needed. Despite these concerns, 65.3% of the students agreed that radiologists should embrace AI and collaborate with the IT industry to facilitate its implementation, as shown in Figure 2.

OPEN IN VIEWER

Figure 1.

Flowchart of Participant Selection and Response Rate.

OPEN IN VIEWER

Figure 2.

Medical Students’ Perceptions of AI in Radiology.

Regarding telemedicine and teleradiology, 15.5% expressed a personal interest in these technologies, yet 45.5% mentioned that the introduction of telemedicine/teleradiology influenced their decision to pursue radiology as a specialty. Furthermore, 84.9% believed that telemedicine/teleradiology could improve patient care. When asked about the potential impact on the workplace, 46.2% agreed that teleradiology could replace hospital-based work, while a larger proportion, 62.2%, indicated they planned to integrate telemedicine/teleradiology into their future careers, as shown in Figure 3.

OPEN IN VIEWER

Figure 3.

Medical Students’ Perception of Telemedicine and Teleradiology.

Although efforts were made to recruit participants using social media and direct communication, approximately 34% of eligible students did not complete the survey. While the response rate was acceptable, the perceptions of nonrespondents may differ from those who participated, which could influence the results.

A total of 292 responses were analyzed. Missing data were minimal, with listwise deletion applied to maintain data integrity. The proportion of missing responses for individual questions did not exceed 5%, ensuring no significant impact on statistical conclusions. Table 1 presents a summary of participants’ responses regarding AI and teleradiology.

OPEN IN VIEWER

Table 1.

Role of Artificial Intelligence and Teleradiology in Radiology as Reported by Medical Students.

ITEMS		NO	PERCENT
Sample size		292	65.2
Artificial intelligence will augment radiologist's capability and make radiologists more efficient	Strongly disagree	33	7.7
	Disagree	35	8.1
	Neutral	121	28.1
	Somewhat agree	73	16.9
	Agree	99	23.0
	Strongly agree	70	16.2
Artificial intelligence will replace radiologists in the current time	Strongly disagree	80	18.6
	Disagree	133	30.9
	Neutral	117	27.1
	Somewhat agree	65	15.1
	Agree	28	6.5
	Strongly agree	8	1.9
The impact of artificial intelligence alone will reduce the number of radiologists that are needed	Strongly disagree	42	9.7
	Disagree	75	17.4
	Neutral	131	30.4
	Somewhat agree	106	24.6
	Agree	53	12.3
	Strongly agree	24	5.6
Radiologists should embrace artificial intelligence and work with the IT industry for its application	Strongly disagree	18	4.2
	Disagree	12	2.8
	Neutral	120	27.8
	Somewhat agree	87	20.2
	Agree	108	25.1
	Strongly agree	86	20.0
How is your interest in telemedicine/teleradiology?	Uninterested	148	34.3
	Neutral	216	50.1
	Interested	67	15.5
Would the introduction of telemedicine/teleradiology to the field of radiology influence your choice?	Yes	196	45.5
	No	235	54.5
The use of telemedicine/teleradiology will improve patients’ care	Agree	366	84.9
	Disagree	65	15.1
Teleradiology can replace working at the hospital	Agree	199	46.2
	Disagree	232	53.8
I plan to use telemedicine/teleradiology in my future career	Yes	268	62.2
	No	163	37.8

This flow illustrates the selection process of study participants. Out of 448 eligible medical students, all were invited to participate. A total of 292 students completed the survey, resulting in a response rate of 65.2%. Nonparticipation (34.8%) was attributed to factors such as lack of interest, academic workload, or technical barriers. No responses were excluded from the final analysis.

This figure illustrates the distribution of medical students’ opinions on the role of AI in radiology. The responses are categorized into 3 groups: disagree (strongly disagree/disagree) in red, neutral in gray, and agree (somewhat agree/agree/strongly agree) in blue.

The figure demonstrates medical students’ perspectives on telemedicine and teleradiology. The majority agree (blue) that teleradiology can enhance patient care and plan to incorporate it into their future careers.

However, opinions are divided on whether it can fully replace in-hospital work. While many acknowledge its potential influence on career choices, a notable portion remains neutral or disagrees (red) with embracing telemedicine in their practice.

Discussion

In our study, 56.1% of medical students agreed that AI will augment radiologists’ capabilities and improve efficiency. This aligns with a meta-analysis by Hassankhani et al, which reported that 79.48% of respondents, including medical students, radiology trainees, and radiologists, believed AI will revolutionize radiology in the future. ⁵ Similarly, Klenam et al found that 76.9% of students were influenced to consider radiology as a future career due to the application of technology and AI in the field. ⁶

Conversely, 42.5% of medical students in our study believed that AI would reduce the number of radiologists needed, and 23.4% felt that AI would replace radiologists. This is consistent with findings from Bin Dahmash et al, where 44.8% of medical students believed AI would minimize the need for radiologists in the future, and 31% believed AI would replace radiologists within their lifetime. ⁷ Cherry sit also concluded that medical students saw AI as a potential “threat,” making them less inclined to specialize in radiology. ⁸ Reeder and Lee reinforced these concerns, finding that approximately one-third of medical students thought AI would fully replace radiologists, and about half of aspiring radiologists expressed anxiety toward AI. ⁹

Moreover, Hassankhani et al noted that medical students had a higher rate of belief (42.66%) that AI poses a threat to the radiology job market compared to radiology trainees and radiologists (6.25%). ⁵ As the field shifts toward teleradiology, our study found that 45.5% of medical students’ choice of radiology as a specialty was influenced by the introduction of telemedicine/teleradiology. This trend may impact the level of interest in diagnostic radiology, as medical students often prefer specialties with more direct patient contact, as noted by Hammond et al. ¹⁰

A systematic review by Santomartino found that medical students were almost evenly divided in their belief that AI poses a threat to the radiologist's role, with 36.6% agreeing, 25.7% neutral, and 37.6% disagreeing. This division highlights the uncertainty among medical students regarding the future impact of AI on radiology. ¹¹

Furthermore, a study by Grayev revealed that approximately one-third of Canadian medical students thought AI would fully replace radiologists, and about half of aspiring radiologists expressed anxiety toward AI. ¹² This anxiety may deter some students from pursuing radiology as a specialty.

These studies collectively indicate that while many medical students recognize the potential benefits of AI in enhancing radiological practice, there is also significant concern about job displacement and the future role of radiologists. This ambivalence underscores the need for educational initiatives to address misconceptions about AI and to prepare future radiologists for a landscape increasingly influenced by technological advancements.

Our study, conducted at a single institution focusing on medical students at IAU in Saudi Arabia's Eastern region, has limitations that should be acknowledged. While these findings provide valuable insights, they may not be generalizable to other institutions or regions. Future research should include multicenter studies to capture a more comprehensive perspective. Additionally, the study's participant diversity was somewhat limited, and despite efforts to include a broad sample, the perspectives of underrepresented groups may not be fully reflected. Furthermore, a formal power analysis was not conducted, which limits the ability to determine whether the sample size was sufficient to detect meaningful statistical differences. This should be considered in the interpretation of the results, and future studies should conduct power analyses to ensure adequate statistical power.

The cross-sectional design offers only a snapshot of participants’ perceptions at a specific point in time. Future studies should incorporate a longitudinal approach to track changes in attitudes over time, providing deeper insights into evolving perceptions regarding AI and teleradiology. Moreover, the absence of a comparison group, such as practicing radiologists or other healthcare professionals, restricts the broader context of these findings. Including such groups in future research could offer valuable comparative insights.

To strengthen future research, studies should employ more rigorous designs, expand sample diversity, and integrate advanced statistical analyses. Such approaches would enhance the understanding of medical students’ attitudes toward AI and teleradiology.

Conclusion

Medical students at IAU had mixed views on AI in radiology; some saw it as a tool for improving accuracy and efficiency, while others worried about job displacement. Similarly, many viewed teleradiology as a way to enhance patient care, but concerns persisted about its ability to fully replace hospital-based radiology.

These findings emphasize the need for targeted education to help future radiologists navigate AI and telemedicine advancements. Integrating these topics into medical training will ensure a workforce that embraces innovation while maintaining patient-centered care.

Supplemental Material