Truace
Subscribe
70/100

Moderate evidence

Cite
More
Export CSVExport JSONEmailLinkedInX

Review of 7 studies finds AI helped orthodontic students learn, but few programs teach it

Study title: Integrating Artificial Intelligence into Orthodontic Education: A Systematic Review and Meta-Analysis of Clinical Teaching Application

Evidence type
Systematic review or meta-analysis
Sample
n = 1,101
Sources
1 sourcePeer-reviewed
Published
2 min read
Truvace record
TRV-2026-2834cite or check it

Study title: Integrating Artificial Intelligence into Orthodontic Education: A Systematic Review and Meta-Analysis of Clinical Teaching Application

A systematic review and meta-analysis of seven studies covering 1,101 participants examined how AI is used to teach orthodontics. The participants included 449 orthodontic students, 408 orthodontists, 149 faculty members, 41 program directors and 30 general dentists, plus 24 trainees in one quasi-experimental study. Tools ranged from cephalometric landmarking platforms to ChatGPT-based learning modules.

Read the study abstract

Background/Objectives: Artificial intelligence (AI) is rapidly emerging as a transformative force in healthcare education, including orthodontics. This systematic review and meta-analysis aimed to evaluate the integration of AI into orthodontic training programs, focusing on its effectiveness in improving diagnostic accuracy, learner engagement, and the perceived quality of AI-generated educational content. Materials and Methods: A comprehensive literature search was conducted across PubMed, Scopus, Web of Scien…

Integrating Artificial Intelligence into Orthodontic Education: A Systematic Review and Meta-Analysis of Clinical Teaching Application
"A Navy dentist exames a patient at sea" by Official U.S. Navy Page, public domain.

64Gain side

How the two sides scoreClose call

64Problem side

The gain

A systematic review of seven studies with 1,101 participants found AI-assisted cephalometry training improved orthodontic students' landmarking accuracy, and pooled ChatGPT content quality scored 3.69 on the Global Quality Scale (two studies).

The problem

A systematic review of seven orthodontic education studies found 87.8% of North American programs lacked formal AI seminars and faculty cited high costs (89.3%) and lack of technical knowledge (94.7%).

A fixed formula sets both scores from the evidence and its sources. Reader votes don’t move them.

What the scores mean

G is the gain side, out of 100. It goes up when the gain is specific, when it measurably helped people and when the sources behind it are strong.

P is the problem side, out of 100. It goes up when the problem is specific, when it measurably hurt people and when the sources behind it are strong.

In brief

Only two studies had comparable data for pooling. They gave a Global Quality Scale score of 3.69 (95% CI 3.58–3.80) for ChatGPT-generated orthodontic content, described as moderate perceived quality, with heterogeneity of 64.5%. All other outcomes were summarised narratively because of inconsistent measures and missing statistics.

The findings point to learner benefit alongside slow institutional uptake. Students and junior faculty were more enthusiastic, while experienced orthodontists were more skeptical of AI content. Most programs lack the training, expertise and curricular space to teach it, and the evidence base is small and mostly non-randomized.

Main points

  1. Seven studies were included from 896 screened records and 107 full-text reviews; sample sizes ranged from 24 to 298 participants.
  2. Two studies were pooled: ChatGPT-generated orthodontic content scored 3.69 on the Global Quality Scale (95% CI 3.58–3.80, I2 = 64.5%).
  3. Lin et al. found AI-assisted digital cephalometry improved students' landmarking accuracy, with three rounds of training raising accuracy over time.
  4. In a North American survey, 56.1% of residency programs had adopted or planned AI instruction, mostly for research (60.9%) rather than diagnosis (39.1%); 87.8% lacked formal AI seminars.
  5. Orthodontists were more skeptical of chatbot answers (mean GQS 2.87–3.23) than students and general dentists (3.93–4.10).
  6. Among faculty in one survey, 72% supported AI in postgraduate curricula, but 89.3% cited cost and 94.7% cited lack of technical knowledge as obstacles.
  7. Risk of bias was low in three studies, moderate in three and serious in one; GRADE rated four studies moderate quality and three low.

What the review found when it combined the studies

These are the review’s own pooled results, copied from its abstract. Truvace did not recalculate them.

  • Perceived quality of ChatGPT-generated orthodontic educational content on the Global Quality Scale

    Pooled mean GQS score 3.69 (95% CI 3.58 to 3.80)

    search to May 2025

    The review’s own words
    The pooled mean GQS score was 3.69 (95% CI: 3.58 to 3.80), indicating a moderate-to-high perceived quality of information.

Code checks every number above against the review’s text before it is shown. The short labels are an AI model’s reading of that text, so the review’s own sentence sits under each one. A pooled figure is only as strong as the studies behind it. How Truvace uses pooled results

The full story

The review followed PRISMA 2020 and was registered in PROSPERO. Searches of PubMed, Scopus, Web of Science and Embase ran through May 2025, and two reviewers screened and extracted data independently. Risk of bias was assessed with ROBINS-I and certainty with GRADE.

On skills, Lin et al. reported that AI-supported cephalometric training improved student landmarking accuracy through real-time visual feedback. Metin et al. found chatbot-assisted clinical reasoning scored comparably to human-guided sessions in accuracy and satisfaction, though ChatGPT-4 was less consistent than human responses.

Perceptions split by experience. In Kurt et al., patients rated ChatGPT highest (mean GQS 4.27 for clear aligners, 4.40 for lingual orthodontics) while orthodontists gave 3.43–3.67. In Gupta et al., 74% of faculty agreed AI improved cephalometric accuracy, yet 52% in another survey worried about AI errors in treatment planning.

Adoption lagged belief. Mengi et al. found 62.5% of clinicians did not use AI for cephalometric analysis because of cost and technical barriers, and more than 60% were not using AI in routine practice, even though 84% of respondents agreed it could improve diagnostic quality.

What this doesn’t fix

Only two studies could be pooled and I2 was 64.5%; the other outcomes relied on narrative synthesis of mostly small, non-randomized studies and self-reported survey data, which limits generalizability.

Sources

1 on file
  1. Journal of Clinical Medicinedoi:10.3390/jcm14155487
    Peer-reviewed2025-08-04

Talk it over

Four questions to start onThis record found a gain and a problem in the same place. Which matters more to you, and who might answer differently?

Start with a question

Review of 7 studies finds AI helped orthodontic students learn, but few programs teach ithttps://truvace.com/item/integrating-artificial-intelligence-into-orthodontic-education-a-systematic-revi

  1. This record found a gain and a problem in the same place. Which matters more to you, and who might answer differently?

    The gain A systematic review of seven studies with 1,101 participants found AI-assisted cephalometry training improved orthodontic students' landmarking accuracy, and pooled ChatGPT content quality scored 3.69 on the Global Quality Scale (two studies).

    The problem A systematic review of seven orthodontic education studies found 87.8% of North American programs lacked formal AI seminars and faculty cited high costs (89.3%) and lack of technical knowledge (94.7%).

    Answer in the thread

  2. How do we know? What can this kind of evidence show, and what can’t it?

    The evidence This rests on a systematic review or meta-analysis with a sample of 1,101, from Journal of Clinical Medicine.

    Answer in the thread

  3. What does it leave open? What else would you want to know before acting on it?

    The record’s own caveat Only two studies could be pooled and I2 was 64.5%; the other outcomes relied on narrative synthesis of mostly small, non-randomized studies and self-reported survey data, which limits generalizability.

    Answer in the thread

  4. Have you seen AI used this way where you study, work or live? Did it go the same way?

    Answer in the thread

The questions are the same on every record. The details under them are taken from this record as written.

What readers say

More from Health