Attitude of Brazilian Dentists and Dental Students Regarding The Future Role of Artificial Intelligence in Oral Radiology: A Multicenter Survey
Attitude of Brazilian Dentists and Dental Students Regarding The Future Role of Artificial Intelligence in Oral Radiology: A Multicenter Survey
Attitude of Brazilian Dentists and Dental Students Regarding The Future Role of Artificial Intelligence in Oral Radiology: A Multicenter Survey
birpublications.org/dmfr
RESEARCH ARTICLE
Attitude of Brazilian dentists and dental students regarding
the future role of artificial intelligence in oral radiology: a
multicenter survey
1
Ruben Pauwels and 2Yumi Chokyu Del Rey
Aarhus Institute of Advanced Studies, Aarhus University, Aarhus, Denmark; 2Department of Dental Materials and Prosthodontics,
1
Ribeirão Preto Dental School, University of São Paulo, São Paulo, Brazil
Objectives: The aim of this study was to assess the attitude of dentists and dental students in
Brazil regarding the impact of artificial intelligence (AI) in oral radiology, and to evaluate the
effect of an introductory AI lecture on their attitude.
Methods: A questionnaire was prepared, comprising statements regarding the future role
of AI in oral radiology and dentistry. A lecture of approx. 1 h was prepared, comprising the
basic principles of AI and a non-exhaustive overview of AI research in medicine and dentistry.
Participants filled in the questionnaire prior to the lecture. After the lecture, the questionnaire
was repeated.
Results: Throughout 7 sessions at 6 locations, 293 questionnaires were collected. The
majority of participants were undergraduate dental students (57%). Prior to the lecture, there
was a strong agreement regarding the various future roles and expected impact of AI in oral
radiology. Approximately, one-third of participants was concerned about AI. After the lecture,
agreement regarding the different roles of AI in oral radiology increased, overall excitement
regarding AI increased, and concerns regarding the potential replacement of oral radiologists
decreased.
Conclusions: A generally positive attitude towards AI was found; an introductory lecture was
beneficial towards this attitude and alleviated concerns regarding the effect of AI on the oral
radiology profession. Given the unprecedented, ongoing revolution of AI-augmented radi-
ology, it is pivotal to incorporate AI topics in dental training curricula.
Dentomaxillofacial Radiology (2021) 50, 20200461. doi: 10.1259/dmfr.20200461
Cite this article as: Pauwels R, Del Rey YC. Attitude of Brazilian dentists and dental students
regarding the future role of artificial intelligence in oral radiology: a multicenter survey. Dento-
maxillofac Radiol 2021; 50: 20200461.
Introduction
In the last few years, artificial intelligence (AI) has investigated applications involve the detection of
become the hottest topic in radiology.1 New AI methods pathosis and the segmentation of anatomical and
using deep learning (DL), often involving convolu- pathological structures.3
tional neural networks (CNNs),2 along with improved In dentistry, AI applications have been investigated
computational capacity has resulted in the explora- as early as 1992.3,4 Whereas most of the research in
tion of various AI applications. The most commonly subsequent years involved cephalometric landmark
detection, CNN-based DL has recently been investi-
Correspondence to: Prof. Ruben Pauwels, E-mail: pauwelsruben@hotmail.com gated for tasks such as osteoporosis detection, tooth
Received 24 September 2020; revised 14 December 2020; accepted 14 December detection/numbering, and the detection of various
2020 types of oral pathosis.2,3,5,6 Seeing that most of the AI
Attitude of dentists regarding AI
Pauwels, Del Rey 2 of 7
research in radiology involve imaging techniques with and professors from local institutions. Participation in
large patient populations, such as chest radiography and the survey was voluntary, anonymous and was not part
mammography, dental radiology is highly suited for AI of any student evaluation. Hence, no ethical approval
because of the high frequency of imaging performed was required.
worldwide. Based on data of the United Nations Scien- As the aforementioned study by Pinto dos Santos et
tific Committee on the Effects of Atomic Radiation al13 was aimed at undergraduate medical students, the
(UNSCEAR),7 the European Commission’s Radiation following adjustments were made to make it more suit-
Protection report 180 [2] and the 2014–2015 NEXT able for the field of oral radiology:
survey in the United States [3],7–9 a conservative esti-
mate has been made that at least 1.5 billion dental radio- • Answer options regarding professional background
graphic examinations are performed annually at the were adapted to represent different types of students
time of writing.10 and professionals in dentistry.
In the context of this ongoing AI revolution, it • A question was added to ask licensed dentists regard-
is essential to ensure that current and future clinical ing their specialty, if applicable.
practitioners are up-to-date regarding the current state • Questions pertaining ‘radiology’ were specified to
and potential of this technology. Whereas the number ‘oral radiology’, and ‘medicine/physician’ was re-
of published research studies on AI in radiology is placed with ‘dentistry/dentist’.
increasing considerably, evidence-based guidelines and • Answer options regarding the applications of AI in
position statements are pending, especially in dentistry. oral radiology were expanded by adding ‘treatment
A lack of information, or an excess of misinformation, planning’ and ‘image processing’. Specific examples
regarding the role of AI in the future of clinical practice were added for most of the options to ensure that the
could lead to negative attitudes and poorly informed difference between them was clear.
career choices.11 On the other hand, the limited sample • Specific cut-off dates were provided in questions
sizes in the majority of currently published AI studies regarding the potential replacement of oral radiol-
in dentistry indicate that, for many applications, clin- ogists (≤15 years from now) and dentists in general
ical implementation is several steps away.2,3 The overall (≤30 years from now) by AI-based systems.
consensus, however, seems to be that it is a matter of time
before many radiological tasks are “augmented”12 using The original questionnaire was prepared in English; it
AI, with the extent of the AI’s role being highly task- was translated into Portuguese by a native speaker.
specific. Thus, it is important to provide students and A lecture of approx. 1 h was prepared, comprising the
clinicians with accurate, objective and up-to-date infor- following topics:
mation during this transitional phase. In this context,
it is valuable to assess the current attitude of clinical • Introduction: examples of AI in fiction and current
dentists as well as undergraduate students in dentistry society (5’).
regarding the future role of AI. Therefore, the aim of • Overview of the basic principles of AI: examples of
this study was to assess the attitude of (future) dentists animal recognition, showing improved AI perfor-
regarding the use of AI in oral radiology, as well as the mance over time and indicating the limitations of
effect of an introductory AI lecture on their attitude. conventional image processing approaches for com-
plex recognition tasks (5’).
• Basic concepts: distinction between AI/machine
Methods and materials learning/DL, classification vs regression tasks (with
radiological example), different types of AI-based
A questionnaire was prepared based on the study from learning (supervised/unsupervised/reinforcement)
Pinto dos Santos et al13. The survey included demo- (10’).
graphic information, professional background, famil- • Supervised learning: example of a diagnostic appli-
iarity with AI, applicability of AI in oral radiology/ cation, data labelling, concepts related to training
dentistry, and attitude regarding the impact of AI (fitting a model with weights, loss, gradient descent,
in oral radiology/dentistry. The questionnaire was moment, data augmentation), brief illustration of
provided to participants of seminars or congresses at variety of machine learning algorithms (10’).
six different locations in Brazil: (1) University of São • Convolutional neural networks: basic concepts, bene-
Paulo, School of Dentistry of Ribeirão Preto (Ribeirão fits for image-based tasks (10’).
Preto, SP); (2) University of São Paulo, School of • Non-exhaustive overview of published research stud-
Dentistry of Bauru (Bauru, SP); (3) Federal Univer- ies on AI applications in medicine and dentistry.
sity of Espírito Santo (Vitória, ES); (4) XXIth annual The selected studies comprised various types of al-
meeting of SNNPqO (Campina Grande, PB); (5) State gorithms, and types of tasks, e.g. diagnosis, segmen-
University of Northern Paraná (Jacarezinho, PR); tation, image enhancement, prediction of patient
(6) Faculty of Medicine and Dentistry São Leopoldo outcome (15’).
Mandic (Campinas, SP). The participants comprised a • Future prospects: steps towards AI implementation
heterogeneous group of students, clinicians, researchers in radiology, how to get involved in AI research (5’).
Table 1 Professional background of participants Table 2 Distribution of participants with dental license
Background Participants % Participants %
Undergraduate student (dentistry) 167 57.0 General practitioner 21 19.3
Undergraduate student (other) 5a 1.7 Specialist (oral radiology) 30 27.5
Postgraduate/PhD student (oral radiology) 30 10.2 Specialist (other) 57 52.3
Postgraduate/PhD student (other dental 29 9.9 Prosthetic technician 1 0.9
disciplines) Total 109
Postdoctoral researcher 5 1.7
Professor 43 14.7
Dentist 12 4.1 were specialists in fields other than oral radiology
(clinical practitioner & none of the above) (52.3%) (Table 2). The age of the participants was 22.6
Other 2b 0.7 ± 4.1 years for undergraduate students and 34.3 ± 11.9
Total 293 years for others; 69% of the participants were female
a
Medicine (n = 2), biology (n = 1), biotechnology (n = 1), computer (Table 3). A symmetric distribution was found regarding
science (n = 1). the participants’ self-assessment of their technological
b
Systems analyst (n = 1), unspecified researcher (n = 1). proficiency, with a small number finding themselves
decidedly proficient (9.4%) or inept (7.7%).
The participants’ overall awareness of AI is shown
Care was taken to provide all information in an objec- in Table 4. Whereas 63.0% of the participants was
tive manner, as to not to steer the audience towards any unfamiliar with the application of AI in radiology, a
particular opinion regarding the role of AI. The lecture substantial amount (24.7%) assessed that they already
was given in English by a single presenter throughout all have a basic understanding of this technology. Most of
sessions. A PDF file of the lecture can be provided by the prior awareness of AI came from media (93.3%),
the corresponding author at request. After the presen- including social media (90.0%). Almost half of the
tation, questions and answers were mostly in English, participants had already attended some type of presen-
although a translator was available. Immediately after tation regarding AI (not counting the lecture they
the lecture, the participants were asked to fill in the last attended on the day of the survey).
part of the questionnaire, which repeated the questions
pertaining their view on the applications and role of Attitude regarding AI and effect of introductory lecture
AI in dentistry. Results were analyzed using descriptive Opinions regarding the potential applications of AI in
statistics. Answers prior to and subsequent to the lecture oral radiology are found in Table 5. Percentages in the
were compared using the Wilcoxon matched- pairs following subsection always refer to absolute, not rela-
signed rank test. Furthermore, the answers regarding tive, changes in response frequency (e.g. a change in
the attitude towards AI were split up using subpop- response frequency from 40 to 60% would be indicated
ulations based on gender, professional level, dental as ’+20%’, not ‘+50%’). Prior to the lecture, most of
specialty, and self-reported technological proficiency. To the participants partially or fully agreed that AI could
assess differences in response due to professional level, play a role in automatic detection of pathosis on images,
undergraduate students were compared with all other treatment planning, image processing and selection of
participants. To assess the effect of the dental specialty, appropriate imaging techniques. However, only a slight
a three-way intercomparison was performed between majority (62.6%) agreed that it could provide the final
general practitioners, specialists in oral radiology, and diagnosis. After the lecture, a significantly increased
other dental specialists. In terms of technological profi- agreement was found for each potential AI application
ciency, participants who answered “agree” or “rather except treatment planning.
agree” on this question were compared with others. Responses regarding the participants’ personal
The Mann–Whitney test was used for all comparisons perception regarding the use of AI in oral radiology and
between subpopulations. A significance level of 0.05 was
used for all tests.
Table 3 Demographics and self-assessment of technological profi-
ciency
Results Median First/third quartile Min/Max
Age (years) 24 21/32 18/63
Throughout 7 sessions at the aforementioned 6 loca-
tions, 293 questionnaires were collected. General data Male Female N/A
regarding the participants are summarized in Table 1. Gender (%) 89 (30.4%) 199 (67.9%) 5 (1.7%)
The majority of the participants were undergraduate Agree Rather agree Rather Disagree
students in dentistry (57.0%), followed by postgraduate/ entirely disagree entirely
PhD students (20.2%) and professors (14.7%). Out of Technologically 9.4 43.2 39.7 7.7
proficient (%)
the participants with a degree in dentistry, the majority
Dentomaxillofac
Radiol, 50, 20200461 birpublications.org/dmfr
Attitude of dentists regarding AI
Pauwels, Del Rey 4 of 7
dentistry are found in Table 6. Before the lecture, for • There was a generally increased excitement regarding
most of the questions, a high consensus (>85% agree- the use of AI in oral radiology (+5.6%; p < 0.001)
ment or disagreement) was seen for most statements; and dentistry (+5.5%; p < 0.001). On the other hand,
mixed responses were found regarding being worried by there was no significant change in responses regard-
AI development (33.5% agreed) and the substitution of ing whether AI will improve oral radiology or den-
oral radiologists by AI programs in the next 15 years tistry, although a higher degree of agreement was
(22.9% agreed). The most notable effects of the lecture found for both questions.
were: • Although there was a high overall agreement that AI
should be part of dental training curricula prior to
• Over 11% of the responders switched from being
the lecture (94.4%), the percentage of participants
somewhat agreed to completely agreed that AI rep-
with strong agreement significantly increased after
resents a revolution in oral radiology, and 5.5% of
the lecture (+7.2%; p < 0.05).
binary responses switched to agreeing that it will rev-
olutionize dentistry (p < 0.01).
• Interestingly, there was a non-significant tendency to Effect of gender, professional level and technological
switch from slightly disagree (−16.2%) to completely proficiency on attitude regarding AI
disagree (+14.2%) for the question of whether oral ra- The following significant differences were found:
diologists will be replaced by AI. On the other hand,
agreement regarding long-term (<30 years) replace- • No significant difference in response between gen-
ment of dentists by AI-based alternatives increased ders, except for the statement that AI will revolution-
by 8.2% (p < 0.001), and agreements that dentists will ize dentistry in general, for which females showed
never be replaced decreased by 7.4% (p < 0.01). higher agreement (p = 0.02).
• Undergraduate students showed a significantly dif- of oral radiologists (higher agreement, p = 0.01) and
ferent response regarding the replacement of oral excitement regarding AI’s impact on the dental pro-
radiologists (higher agreement; p = 0.03), excitement fession (higher agreement, p = 0.02).
regarding AI’s impact on the dental profession (low-
er agreement; p = 0.04) and the statement that AI
should be part of dental training (lower agreement; Discussion
p < 0.01).
• General practitioners showed an increased excite- In the midst of the ongoing AI revolution that is
ment regarding the impact of AI on the dental pro- expected to affect profoundly the future of radiology,
fession than specialists in oral radiology (p = 0.03) as this study assessed the current perception of AI in oral
well as other dental specialists (p < 0.01). Note that radiology and dentistry. Whereas there was a generally
the sample size was somewhat limited for these sub- optimistic attitude towards AI, it was found that an
groups, which limits the statistical power. introductory lecture on this topic had an immediate,
• Participants with technological proficiency answered positive effect on the perception of dental students and
differently for statements regarding the replacement professionals.
Dentomaxillofac
Radiol, 50, 20200461 birpublications.org/dmfr
Attitude of dentists regarding AI
Pauwels, Del Rey 6 of 7
A degree of concern and uncertainty by (present and the rapid changes that are needed in curricula for
future) dental professionals can be expected, considering radiology residents as well as continuing education.21
the growing evidence regarding the performance of AI It is expected (and recommended) that AI tools in
tools for dedicated tasks.2,3,5 Furthermore, people have dentistry as well as ethical standards, regulations and
been made aware of the benefits of AI outside of medi- guidelines, will be developed in parallel to those in
cine for relatively complex tasks (e.g. autonomous vehi- medicine.5
cles). Although it is possible to predict the extent of the The adaptation of teaching curricula and guide-
impact of AI in dentistry, as well as the timeline of this lines in dentistry, as well as the actual implementa-
impact, it is conceivable that the role of responsibilities tion of AI tools in clinical practice, may require a
of radiologists will evolve alongside the development of different approach in countries with or without an oral
diagnostic AI tools. radiology specialty. The current study was performed
Although this is the first survey of this kind in in a country that has a recognized oral radiology
dentistry, several recent studies have assessed attitudes specialty. At the time of writing, it is estimated that 53
regarding the impact of AI in medical (i.e. non-dental) countries worldwide recognize this specialty; notably,
radiology. The current survey was adapted from the this is only the case in 3 out of 27 EU member states
study by Pinto dos Santos et al13 involving German and 7 of 53 countries in the WHO European Region
medical students. They found that students are aware, (private communication, Axel Ruprecht, University
but not concerned about the role of AI in radiology or of Florida, 2019). In countries with an oral radiology
general medicine. A majority of students judged that specialty, the ‘augmented radiology’12 concept will
AI topics should be included in the medical training need to be discussed in considerable detail, to ensure
curriculum. Other studies involving Canadian14 and that current and future specialists in this field are
Swiss15 medical students showed a higher degree prepared to make use of AI to its full extent. On the
of concern, indicated by their reduced intention of other hand, one could stratify the approach towards
choosing radiology as a specialty due to their uncer- teaching, regulations and ethical guidelines among
tainty regarding the future, and possible obsolescence, all countries, regardless of the existence of an oral
of this profession within a few decades. Similarly, two radiology specialty. While it can be envisaged that
separate surveys involving US- and UK-based medical certain radiological tasks will be gradually performed
students showed a conviction in the future role of AI,
by AI for the purpose of reduced reporting time and/
coupled with a decreased interest in radiology by
or higher diagnostic efficacy, this does not imply that
almost half of the students.16,17 In the US study, it was
such tasks should be removed entirely from radiolog-
found that most of the current awareness of AI comes
ical courses, regardless of the specialty. The propor-
from online articles, which can be expected to result
tion of the curriculum that should be dedicated to AI
in a more negative outlook due to overhyping.16 In the
will undoubtedly be debated in extensive detail in the
UK study,17 participants that received prior teaching
in AI were less adverse towards choosing radiology as next years. This is likely to depend on the specialty in
a profession, indicating the positive impact that can be question; as a hypothetical example, orthodontists and
achieved by providing objective and realistic informa- endodontists in training may require practical training
tion regarding AI.11 in AI if these tools become part of routine practice in
Another US survey at a radiology residency their specialty. Furthermore, undergraduate teaching
program revealed a lack of acquaintance to the curricula can contain AI topics regardless of post-
state of the art in medical AI, leading to concerns graduate career options, although there may be a need
by trainees regarding the impact of AI.18 A French for a higher emphasis on the possibly changed, not
survey comprising radiologist residents and senior reduced, role of oral radiologists in the near future
radiologists showed similar sentiments regarding a to ensure that newly graduated dentists can make an
lack of information in AI, and a willingness to famil- educated choice of their specialty.
iarize themselves with this topic at a theoretical and Further effort should be undertaken to get a cross-
practical level.19 The French participants were gener- section of the current perception regarding AI in
ally optimistic regarding the future impact of AI, dentistry in different countries and regions. Whereas
expecting a reduced error rate and a change in work- the sample size in this study was limited due to the
flow that allows for more patient interaction.19 requirement of being physically present for a lecture,
Recently, a consortium of European and North this was purposely done to assess the immediate
American societies released a statement regarding the effect of such a lecture on the participants’ percep-
ethical use of AI in radiology; among others, the need tion. The use of electronic surveys could be consid-
to develop codes of ethics and practice was stressed.20 ered as a more wide-scale approach to evaluate the
Several other national associations or communi- attitude of students and clinicians. Furthermore, the
ties have released statements that also emphasized effect of the lecture found in the current study could
the need for teaching21,22; the French radiology be reproduced using webinars, if physical lectures are
community in particular went into detail regarding not feasible.
REFERENCES
1. Hosny A, Parmar C, Quackenbush J, Schwartz LH, Aerts HJWL. 13. Pinto Dos Santos D, Giese D, Brodehl S, Chon SH, Staab W,
Artificial intelligence in radiology. Nat Rev Cancer 2018; 18: 500– Kleinert R, et al. Medical students' attitude towards artificial
10. doi: https://doi.org/10.1038/s41568-018-0016-5 intelligence: a multicentre survey. Eur Radiol 2019; 29: 1640–6.
2. Hwang J-J, Jung Y-H, Cho B-H, Heo M-S. An overview of deep doi: https://doi.org/10.1007/s00330-018-5601-1
learning in the field of dentistry. Imaging Sci Dent 2019; 49: 1–7. 14. Gong B, Nugent JP, Guest W, Parker W, Chang PJ, Khosa F, et al.
doi: https://doi.org/10.5624/isd.2019.49.1.1 Influence of artificial intelligence on Canadian medical students'
3. Hung K, Montalvao C, Tanaka R, Kawai T, Bornstein MM. preference for radiology specialty: ANational survey study. Acad
The use and performance of artificial intelligence applications in Radiol 2019; 26: 566–77. doi: https://doi.org/10.1016/j.acra.2018.
dental and maxillofacial radiology: a systematic review. Dento- 10.007
maxillofac Radiol 2020; 49: 20190107. doi: https://doi.org/10. 15. van Hoek J, Huber A, Leichtle A, Härmä K, Hilt D,
1259/dmfr.20190107 von Tengg-Kobligk H, et al. A survey on the future of radiology
4. Mol A, van der Stelt PF. Application of computer-aided image among radiologists, medical students and surgeons: students and
interpretation to the diagnosis of periapical bone lesions. Dento- surgeons tend to be more skeptical about artificial intelligence
maxillofac Radiol 1992; 21: 190–4. doi: https://doi.org/10.1259/ and radiologists may fear that other disciplines take over. Eur J
dmfr.21.4.1299632 Radiol 2019; 121: 108742. doi: https://doi.org/10.1016/j.ejrad.
5. Pauwels R. A brief introduction to concepts and applications of 2019.108742
artificial intelligence in dental imaging. Oral Radiol 2020;16 Aug 16. Park CJ, Yi PH, Siegel EL. Medical student perspectives on the
2020. doi: https://doi.org/10.1007/s11282-020-00468-5 impact of artificial intelligence on the practice of medicine. Curr
6. Leite AF, Vasconcelos KdeF, Willems H, Jacobs R. Radiomics Probl Diagn Radiol 2020;27 Jun 2020. doi: https://doi.org/10.1067/
and machine learning in oral healthcare. Proteomics Clin j.cpradiol.2020.06.011
Appl 2020; 14: e1900040. doi: https://doi.org/10.1002/prca. 17. Sit C, Srinivasan R, Amlani A, Muthuswamy K, Azam A,
201900040 Monzon L, et al. Attitudes and perceptions of UK medical students
7. United Nations Scientific Committee on the Effects of Atomic towards artificial intelligence and radiology: a multicentre survey.
Radiation (UNSCEAR).Sources, effects and risks of Ionizing Insights Imaging 2020; 11: 14. doi: https://doi.org/10.1186/s13244-
radiation, UNSCEAR 2008 report to the general assembly with 019-0830-7
scientific annexes. Vol. 1. New York, NY, USA: United Nations; 18. Collado-Mesa F, Alvarez E, Arheart K. The role of artificial intel-
2010. ligence in diagnostic radiology: a survey at a single radiology resi-
8. European Commission.. Medical radiation exposure of the Euro- dency training program. J Am Coll Radiol 2018; 15: 1753–7. doi:
pean population. Radiation protection no 180. Luxembourg: Publi- https://doi.org/10.1016/j.jacr.2017.12.021
cations Office of the European Union; 2015. 19. Waymel Q, Badr S, Demondion X, Cotten A, Jacques T. Impact
9. Farris K, Spelic D. Nationwide evaluation of X-Ray trends: high- of the rise of artificial intelligence in radiology: what do radiolo-
lights of the 2014-15 next dental survey. Proceedings of the 47th gists think? Diagn Interv Imaging 2019; 100: 327–36. doi: https://
national conference on radiation control. Frankfort, KY, USA: doi.org/10.1016/j.diii.2019.03.015
Conference of Radiation Control Program Directors; 2015. pp. 20. Geis JR, Brady A, Wu CC, Spencer J, Ranschaert E, Jaremko JL,
E-15-4. et al. Ethics of artificial intelligence in radiology: summary of
10. International Atomic Energy Agency (IAEA). Radiation protec- the joint European and North American multisociety statement.
tion in dental radiology. In: Safety Reports Series. Vienna, Austria: Insights Imaging 2019; 10: 101. doi: https://doi.org/10.1186/
International Atomic Energy Agency; 2021. pp. 1–81. s13244-019-0785-8
11. Park SH, Do K-H, Kim S, Park JH, Lim Y-S. What should 21. SFR-IA Group, CERF, French Radiology Community. Artificial
medical students know about artificial intelligence in medicine? J intelligence and medical imaging 2018: French radiology commu-
Educ Eval Health Prof 2019; 16: 18. doi: https://doi.org/10.3352/ nity white paper. Diagn Interv Imaging 2018; 99: 727–42. doi:
jeehp.2019.16.18 https://doi.org/10.1016/j.diii.2018.10.003
12. Tajmir SH, Alkasab TK. Toward augmented radiologists: changes 22. Tang A, Tam R, Cadrin-Chênevert A, Guest W, Chong J, Barfett J,
in radiology education in the era of machine learning and artifi- et al. Canadian association of radiologists white paper on artifi-
cial intelligence. Acad Radiol 2018; 25: 747–50. doi: https://doi. cial intelligence in radiology. Can Assoc Radiol J 2018; 69: 120–35.
org/10.1016/j.acra.2018.03.007 doi: https://doi.org/10.1016/j.carj.2018.02.002
Dentomaxillofac
Radiol, 50, 20200461 birpublications.org/dmfr