|Year : 2021 | Volume
| Issue : 2 | Page : 77-81
Awareness and education of medical students toward artificial intelligence and radiology: A cross-sectional multicenter survey at Kanpur, Uttar Pradesh
Rohan Sachdev1, Kriti Garg2, Akash Srivastava2
1 Department of Public Health, UWA School of Population and Global Health, Nedlands, Australia
2 Department of Oral Medicine and Radiology, Rama Dental College, Kanpur, Uttar Pradesh, India
|Date of Submission||31-May-2021|
|Date of Decision||07-Sep-2021|
|Date of Acceptance||07-Sep-2021|
|Date of Web Publication||30-Nov-2021|
Department of Public Health, UWA School of Population and Global Health, Nedlands, WA 6009
Source of Support: None, Conflict of Interest: None
Context: Artificial intelligence (AI) is focused on understanding the essence of human intelligence and developing smart artifacts that can perform the tasks that intelligence is said to entail when performed by humans. Aims: The aim of the study was to understand medical students views, understanding and level of confidence in working on AI and explore if AI influences their career intentions with specific regard to radiology. Settings and Design: 401 medical students of two medical colleges were included. Subjects and Methods: Four and one medical students of two medical colleges were included to complete an anonymous electronic survey consisting of Likert and dichotomous questions. Statistical Analysis Used: Statistical analysis was performed with simple descriptive statistics in frequency and percentages. Results: In this study, 89.11% was the response rate of participants and majority (30.9%) of medical students strongly agreed about the awareness of AI. Nearly 36.3% of the medical students disagreed that they were less likely to consider a career in radiology due to AI. Around 29% of medical students strongly agreed for the scope of integration of AI in medical education of India. Conclusions: Medical students recognize the significance of AI and are eager to get involved. AI medical college curriculum should be broadened and upgraded. Students must be provided with practical use cases and drawbacks of AI so that they may not feel discouraged from pursuing radiology.
Keywords: Artificial intelligence, education, medical students, radiology
|How to cite this article:|
Sachdev R, Garg K, Srivastava A. Awareness and education of medical students toward artificial intelligence and radiology: A cross-sectional multicenter survey at Kanpur, Uttar Pradesh. Dent Med Res 2021;9:77-81
|How to cite this URL:|
Sachdev R, Garg K, Srivastava A. Awareness and education of medical students toward artificial intelligence and radiology: A cross-sectional multicenter survey at Kanpur, Uttar Pradesh. Dent Med Res [serial online] 2021 [cited 2022 Oct 5];9:77-81. Available from: https://www.dmrjournal.org/text.asp?2021/9/2/77/331391
| Introduction|| |
In this research, we conducted a web-based survey with a aim to understand the opinions, understanding, and level of confidence of medical students in working on AI and to explore whether AI affects their career intentions in Kanpur, Uttar Pradesh, and particularly concerning radiology artificial intelligence (AI) term was coined by John McCarthy in 1956, and he defined it as “the science and engineering of making intelligent machines.” AI is the branch of computer science which deals with the study and design of intelligent agents that perceive its environment and take actions which maximize its chances of success., Due to advances in computing resources, learning algorithms, and the accessibility of vast databases from medical records and wearable health monitors, AI is likely to play an increasingly prominent role in medicine and health care. In the year 2016, it was found in the literature that the health-care sector for AI is increasing at a rate of 40% and is expected to reach $6.6 billion by 2021.
In health-care sector, the application of AI continues to generate significant attention. In the year 2017, Mesko in his review suggested AI as “The Stethoscope of the 21st Century” that implies it is an essential tool for the medical fraternity. In developed countries, multiple studies have demonstrated the potential utility of such AI algorithms in some medical specialties including ophthalmology, dermatology, and pathology.,, Within medical radiology, there are now AI tools that employ deep learning methods such as convolutional neural networks which are effective at performing narrow classification tasks where large training datasets are available. High profile examples include the classification of chest radiographs based on abnormality for triage or pathological processes.,, Although there is currently insufficient evidence to conclude the regular use of AI algorithms in emerging economies such as India in the field of clinical radiology, validated use cases for AI equipment in radiology are expected to evolve rapidly with growing academic and industry interest. In this research, we conducted a web-based survey with a aim to understand the opinions, understanding, and level of confidence of medical students in working on AI and to explore whether AI affects their career intentions in Kanpur, Uttar Pradesh, and particularly concerning radiology.
| Subjects and Methods|| |
A total of 450 medical students from two medical universities in Kanpur City were invited to complete an electronic survey (Google Forms and Google LLC) jointly developed by medical clinicians engaged in AI and/or medical education. With a request to distribute the survey, contact was made with medical students through the medical college student coordinator department. Medical students were also directly invited through social media to complete the survey. At the start of the survey, all participants were asked to enter their university email address; all answers were subsequently anonymized. At the start of the study, it was made specifically clear to participants that their answers were anonymous. Only the research included medical students who all returned the duly filled form. The institutional ethical committee received limited risk ethical approval, and informed consent was verified at the start of the survey with individual participants.
The survey design underwent several rounds of iteration, and final validation was performed with a group of forty medical students from who were not included in the final survey. The final survey consisted of 11, 5-point Likert questions, whereby participants rated their agreement toward a presented statement relating to their current attitudes toward AI, their career intentions toward radiology, their current understanding of AI, and their confidence in using AI tools in a routine and critical manner following graduation. Dichotomous questioning was used to determine if participants received teaching on AI and if this teaching formed a compulsory part of their curriculum.
The data were collected, compiled, arranged in a systematic manner, and analyzed using SPSS Version 17.0 (SPSS Inc., Chicago, IL, USA). Statistical analysis was performed with simple descriptive statistics are presented in frequency and percentage to compare the responses relating to knowledge of AI, career choice, and perceived competence in postqualification use of AI tools.
| Results|| |
A total of 401 responses were received out of 450 medical students from two medical colleges of Kanpur city, currently awarding medical degrees recognized by the Medical Council of India. The response rate of participants was 89.11%. The majority of medical students (n = 124, 30.9%) strongly agreed about the awareness of AI [Graph 1]. The majority of respondents were neutral (n = 103, 25.69%) about their believe that AI will play an important role in health care in future when compared with others and rating their responses as either strongly agree (24.67%) or agree (24.17%). A similar number of students also believed that some specialties will be replaced by AI within their lifetime and presented as strongly agree/agree, 27.8%/26.5% [Graph 2].
With regard to the questions relating to the current understanding of AI, nearly half the respondents selected responses indicating they have an understanding of the basic computational principles that underpin AI; 9.97%/15.71% of respondents selecting strongly agree or agree, while 21.6%/28.6% selected disagree or strongly disagree, and the remaining 23.6% were neutral. With regard to the current limitations of AI, more students reported they had an understanding on this than those who did not; 14.3%/21%, respectively, selected strongly agree or agree to this question, while 20.3% selecting strongly disagree or 18.3% disagree and 25.8% selected neutral. More students reported that they did not feel comfortable with the nomenclature associated with AI (strongly disagree, 20.1%), than those that reported otherwise (strongly agree, 18.4%) [Graph 3]. An overwhelming majority of students believed that education in AI would be beneficial for their careers with 42.5% of students responding either strongly agree or agree 39.7%, while 7.8% of students submitted neutral responses on this question. Nearly 4.3% of medical students disagreed with this statement. Along this line of questioning, a significant majority of students 56.2% also strongly agreed with the statement that all medical students should receive education in AI as a part of their medical degree, while 41.3% of students selected agree. Neutral responses were recorded by 1% of students [Graph 4].
Nearly 36.3% and 28.9% of the medical students disagreed and strongly disagreed, respectively, that they were less likely to consider a career in radiology due to AI, compared to those who feel they are likely to consider a career in radiology due to AI (strongly agree 11.2%) [Graph 5]. Around 20.7% of medical students strongly agreed that any computer language knowledge will help in medical education. Nearly 47.2% strongly agreed and 31.7% agreed to incorporation of AI in the Indian medical education syllabus, while only 5% strongly disagreed. Around 29% of medical students strongly agreed and 22% agreed for the scope of integration of AI in medical education of India [Graph 6].
| Discussion|| |
In our study, majority of medical students were aware about the of AI through the Internet search engines which was found similar (62.5%) to the study done by Jindal and Bansal. Sit et al. in their study found that a majority of students with 44.2% strongly agreeing that AI will play an important role in health-care sector, while 48.3% of medical students believed that some specialties will be replaced by AI within their lifetime which was similar to the results of our study.
Due to improvements in computing resources, learning algorithms, and the accessibility of vast databases from medical records and wearable health monitors, AI is likely to play a significant prominent role in medicine and health care. In the study done by Sit et al., 44.6% of medical students were strongly agreed to the understanding of the basic components of principles of AI which was not similar to our study results where only a few students were agreed to the understanding of the principles of AI. While regarding the comfort level with the nomenclature associated with AI, our study results came in accordance with the study done by Sit et al.
Pinto dos Santos et al. in the year 2019 identified an overall low level of information of medical students about AI, with students stating that they acquired this from mainstream media rather than university teaching. Pinto dos Santos et al. also highlighted that students who were more knowledgeable about AI were less afraid of working with technology. In the study done by Sit et al. among the UK medical students, majority of medical students gave an overwhelming response of 53.7% regarding the education of AI would be beneficial for their careers. Whereas 50% agreed with the fact that all medical students should receive the training in AI as a part of their medical degree and these results were found to be similar where majority of Indian students agreed to the statement.
It is now accepted that AI will likely have a profound impact in future practice of radiology and other medical fields. Despite convincing arguments that AI will not replace radiologists, majority of the medical students in the study done by Sit et al. and Langlotz are less likely to consider a career in radiology due to the perceived success of AI, which was found to similar to our study results., Associated with a serious shortage of radiologists in developing countries such as India, it is concerning that a significant number of students in our research overlook radiology as a future occupation. The implications for the workforce are truly devastating, specifically with a growing demand for imaging investigations. We assume that the narrative favored by some adamant supporters of AI, endorsed by mass media, greatly affects students opinions on this. AI has also been considered a potential “disruptor” of radiologists by influential figures from the computer science community and venture capital, perpetuating the myth that radiologists will no longer be needed in near future. More recently, an increasingly circumspect view of AI adaptation in medical imaging, including some of the same opinion leaders who originally predicted the rapid demise of radiologists, has gained acceptance., In our opinion, the persisting misconception of AI rendering radiologists obsolete should be urgently addressed and initiative should be taken by Indian government.
Another challenge in learning AI in medical students is their lack of computer knowledge as computer science is not taught as a subject in premedical school for students pursuing medicine. In the study done by Jindal and Bansal, 73% of medical students were not aware about AI language which is in accordance to our study results. As suggested by Van RG in 1993, computer languages most commonly used in AI are Python and C ++; it is required for coding and making algorithms. Python is also a general-purpose programming language which can be used across many domains and technologies. As the enthusiasm to learn AI is seen in our students, so the same can be translated for learning computer languages as well. Time has come that the medical schools strengthen their medical curriculum and include machine learning, deep learning and data management, and computer languages along with traditional classes as suggested by Ahuja in 2019. This will also create students' interest in choosing AI as future career prospect which is currently not considered as an option by even a single student in this survey.
Kolachalama and Garg in 2018 stated that AI should be incorporated into medical curriculum for under graduates. Many articles from the literature have proved it that introducing AI at an earlier point in medical career helps to shape the students better. A study by Shin et al. in 1993 demonstrated that problem-based learning increases the overall learning of medical undergraduates as compared to the traditional curriculum. The same applies to AI as it enhances the problem-based learning of the students and justifies its incorporation in medical curriculum. However, implementation of AI in medical field is full of challenges. In the current study, the majority of medical students strongly agreed for the incorporation of AI into medical syllabus and also for the scope of integration of AI in Indian medical education to enhance facilities for the patients. While in 2017, Jiang et al. stated that limited digitalization, lack of expertise, nonavailability of data, and financial constraints is another hurdle for AI to include medical curriculum. We suggest that Indian health officials should first persuade themselves and believe that AI technology is going to make a big leap in our Indian health-care industry, and that this technology is going to remain long, it can only be optimistic for a revolution.
| Conclusions|| |
AI will support the future needs of medicine by analyzing the vast amounts and various forms of data that patients and health-care institutions record in every moment. It is now accepted that AI will likely have a profound impact in the future practice of radiology and other medical fields, and it is thus inevitable that AI and other digital tools will be incorporated into clinical practice, regardless of specialty. India has a unique opportunity at this moment where AI can help in the major programs of the Government, viz., Digital India, Make in India, and Skill India, which can help to develop virtual nurse, chatbots, and virtual medical counselor. It would be a failure if we do not seize on this opportunity to better equip our future physicians with the adequate knowledge. It is our belief that the physicians of tomorrow must possess the means to employ digital tools, including AI, in a manner that is akin to rational evidence-based medication use.
Authors would like to thank ethical committee of the college and all medical students for their support throughout the study.
The institutional ethical committee clearance was taken prior and an informed consent was obtained from all participants.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Singh G, Mishra A, Sagar D. An overview of artificial intelligence. SBIT J Sci Technol 2013;2:1-4.
Srivastava SK. Artificial intelligence: Way forward for India. JISTEM USP 2018;15:1-23.
Ahuja AS. The impact of artificial intelligence in medicine on the future role of the physician. PeerJ 2019;7:e7702.
Jindal A, Bansal M. Knowledge and education about artificial intelligence among medical students from teaching institutions of India: A brief survey. Med Ed Publish 2020;9:200-1. Available from: https://doi.org/100.15694/mep. 2020.000200.1
Meskó B. The real era of the art of medicine begins with artificial intelligence. J Med Internet Res 2019;21:e16295.
Ting DS, Pasquale LR, Peng L, Campbell JP, Lee AY, Raman R, et al.
Artificial intelligence and deep learning in ophthalmology. Br J Ophthamol 2019;103:167-75.
Hogarty DT, Su JC, Phan K, Attia M, Hossny M, Nahavandi S, et al
. Artificial intelligence in dermatology where we are and the way to the future: A review. Am J Clin Dermatol 2020;21:41-7. [doi: 10.1007/s40257-019-00462-6].
Niazi MK, Parwani AV, Gurcan MN. Digital pathology and artificial intelligence. Lancet Oncol 2019;20:e253-61.
Sit C, Srinivasan R, Amlani A, Muthuswamy K, Azam A, Monzon L, et al
. Attitudes and perceptions of UK medical students towards artificial intelligence and radiology: A multicentre survey. Insights Imaging 2020;11:1-6.
Annarumma M, Withey SJ, Bakewell RJ, Pesce E, Goh V, Montana G. Automated triaging of adult chest radiographs with deep artificial neural networks. Radiology 2019;291:196-202.
Rajpurkar P, Irvin J, Ball RL, Zhu K, Yang B, Mehta H, et al.
Deep learning for chest radiograph diagnosis: A retrospective comparison of the CheXNeXt algorithm to practicing radiologists. PLoS Med 2018;15:e1002686.
Chilamkurthy S, Ghosh R, Tanamala S, Biviji M, Campeau NG, Venugopal VK, et al.
Deep learning algorithms for detection of critical findings in head CT scans: A retrospective study. Lancet 2018;392:2388-96.
Pinto Dos Santos D, Giese D, Brodehl S, Chon SH, Staab W, Kleinert R, et al
. Medical students' attitude towards artificial intelligence: A multicentre survey'. Eur Radiol 2019;29:1640-6.
Langlotz CP. Will artificial intelligence replace radiologists? Radiol Artif Intell 2019;1:e190058.
Hinton G. Deep learning – A technology with the potential to transform health care. JAMA 2018;320:1101-2.
Kolachalama VB, Garg PS. Machine learning and medical education. NPJ Digit Med 2018;1:1-3.
Shin JH, Haynes RB, Johnston ME. Effect of problem-based, self-directed undergraduate education on life-long learning. CMAJ 1993;148:969-76.
Chan KS, Zary N. Applications and challenges of implementing artificial intelligence in medical education: Integrative review. JMIR Med Educ 2019;5:e13930.
Jiang F, Jiang Y, Zhi H, Dong Y, Li H, Ma S, et al.
Artificial Intelligence in health care: Past, present and future. Stroke Vasc Neurol 2017;2:e000101.