Methodological quality of dental cone-beam CT diagnostic reference levels: a systematic review

Abstract

Objective

To evaluate the methodological quality of dental cone-beam CT (CBCT) diagnostic reference levels (DRLs) studies and reports, identify factors that limit DRL comparison, assess the level and methods of dose optimization, and clarify obstacles, both in dose optimization and DRL establishment.

Methods

A systematic search was applied across 5 databases and gray literature with International Atomic Energy Agency (IAEA) member countries contacted, to identify dental CBCT DRL studies and national reports. Studies/reports containing DRL establishment procedures, insufficient methodological detail, and adopted DRLs were excluded. Methodological quality was assessed using a modified Effective Public Health Practice Project tool, aligned with a checklist developed in accordance with International Commission on Radiological Protection (ICRP) 135.

Results

Of 320 reports screened, 19 were included. Diagnostic reference levels were clinical indication or field of view (FOV) based, using dose/kerma-area-product (DAP/KAP) metrics. Overall, 63% of studies had moderate methodological quality; DRL comparability was limited due to gaps or inconsistencies in reporting of exposure parameters, FOV, and heterogeneous indication grouping. Substantial dose variability was observed within the same clinical indications (2- to 10-fold) and FOV categories (3- to 10-fold), reflecting routine use of large FOVs irrespective of indication, reliance on manufacturer default protocols, limited understanding/education/training of exposure settings or optimization strategies, non-standardized system FOV settings, and unauthorized CBCT operators.

Conclusions

The majority of CBCT DRL studies are not of high quality. Improving DRL reporting quality requires interdisciplinary collaboration; digitized-dose recording with robust retrospective management of protocols, dose, indications; national dose registries for periodic dose audit; improved clinician education/engagement, strengthened regulatory oversight, and potentially international standardized frameworks.

Introduction

The rise in dental cone-beam CT (CBCT) examinations per capita¹ due to supplementing, or replacing, conventional radiography with CBCT, increases individual patient dose and collective effective dose.² Thus, establishing national diagnostic reference levels (DRLs) for dental CBCT imaging examinations is a priority.³

DRLs are a critical tool in dose optimization, introduced by the International Commission on Radiological Protection (ICRP) in Publication 73 (1996),⁴ to serve as a benchmark to identify unusually high or low doses delivered to standard-sized patients during routine clinical examinations, facilitating comparison between CBCT units, facilities, and countries. Recently, research on CBCT dosimetry and DRL establishment has increased,^5–7 highlighting a need not only to compare DRLs and their source of variance, but also to assess the quality of these studies and identify persisting challenges and obstacles.

Thus, the aim of this first systematic review (SR) of all existing dental CBCT DRLs is to evaluate the methodological quality of identified studies and the validity of comparison between these established DRLs while assessing the nature, levels, and challenges in CBCT dose optimization and in DRL establishment.

Methods

The SR was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines⁸ (Figure 1). The review protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) database (CRD42025112995).

Figure 1.

PRISMA flowchart detailing the inclusion and exclusion of articles in the systematic review. Abbreviation: DRL = diagnostic reference level.

Search strategy

The Population, Intervention, and Comparison and Outcome (PICO) framework⁹ was employed to structure the review question and define the search terms:

1. Patient, population, or problem: samples of dental patients undergoing CBCT examination

2. Intervention: exposure to radiation received by patients as part of the CBCT examination

3. Comparison: study design differences (country; size of surveys; methods of measuring dose used in the surveys)

4. Outcome: DRLs established as a result of the studies.

Search terms used included Medical Subject Heading (MeSH) for “Cone-Beam Computed Tomography”, “Diagnostic Reference Levels” and “Dentistry”. Keyword terms were manually developed from a preliminary literature search. The final list of search terms was developed collaboratively by an experienced university librarian and 2 experienced researchers in their field to ensure the validity and completeness of the terms. Following an initial pilot in 1 database to assess yield and optimize the search structure, the final strategy was applied across several databases available at the host institution, including (1) Embase, (2) Medline via OVID, (3) Scopus, (4) Cochrane Library, and (5) Web of Science (Table S1). These databases were selected as they index leading journals in dental radiology, radiography, and medical physics. The initial search was completed on November 5, 2025.

Gray literature and non-indexed publications were manually screened through Google Scholar (first 100 returns), and relevant repositories searched: the Networked Digital Library of Theses and Dissertations (NDLTD); Open Access Theses and Dissertations (OATD); DEEP (E-theses portal), and EThOS (UK’s national theses service). The past 20 years of publications in journals related to DRLs: Dentomaxillofacial Radiology (DMFR); Radiation Protection Dosimetry; and Physica Medica were reviewed for completeness. Reference lists of all included articles and any previously published reviews were examined to identify further relevant studies.

To systematically capture official national reports on dental CBCT DRLs, upon request, the International Atomic Energy Agency (IAEA) provided a regulatory authority directory¹⁰ for 157 countries which were duly contacted via the provided email addresses. For the remaining country authorities (24) not included in the directory, the respective Ministries of Health were approached using Geneva Foundation for Medical Education and Research (GFMER) website.¹¹ The search covered literature published from 1990 to the present, without language restrictions. Non-English articles were translated by radiology academics from the country of origin where feasible.

Screening process

All identified articles from each database and reports were imported into EndNote (Clarivate) and Covidence workflow management platform (Covidence, Melbourne Australia) to facilitate duplicate removal, title and abstract screening, full-text review, and data extraction. Each step of the screening and selection process was independently conducted by 2 reviewers (B.A., M.B.M.), any conflicts being resolved by a third reviewer (R.C.), ensuring that only articles/reports meeting the predefined inclusion criteria were retained for analysis.

Inclusion and exclusion criteria

During title and abstract screening, studies addressing radiation dose in CBCT examinations were included. Following full-text review, articles and national reports were included if they established DRLs for dental CBCT examinations in adult or pediatric patients at regional, national, local, or institutional levels.

Studies/reports were excluded if they lacked sufficient methodological content for quality assessment or reported adopted DRLs, described DRL establishment procedures without reporting actual DRL values, or presented radiation doses not intended as DRLs. Those focusing primarily on infrastructure or software systems for dose data management (eg, NIMIS) or on dosimetry tool evaluation without reporting DRL findings for dental CBCT were also excluded.

Data extraction

From the included studies, data were extracted into Excel, detailing study design, sample size, country of origin, institution type, data collection period, CBCT scanner specifications, clinical indication, exposure parameters, field of view (FOV), method of dose measurement/dose descriptor, optimization strategies, quality control (QC) status, reported dose, and DRL values established according to percentile level used. Corresponding authors were contacted for additional information where necessary (eg, clinical indications, institution diversification, and age categories).

Risk of bias/quality assessment

The methodological quality of each study was independently assessed by 2 reviewers using a modified version of the Effective Public Health Practice Project (EPHPP) tool,¹² integrating a checklist (Table S2) developed in accordance with ICRP Publication 135 guidelines.³ Studies were evaluated across 6 key criteria: study design, study population, recruitment transparency, DRL reporting (adherence to ICRP guidelines), protocol transparency, and data handling. To ensure consistency and parity in evaluation, certain exemptions were applied based on study objectives and DRL types (eg, patient data-based studies were automatically rated “yes” for dosimeter calibration; typical-value studies received “yes” for institutional diversity). Scoring of the 17 questions was based on methods used in previous SRs^13,14: information clearly specified (2 points), information inferable from figures, tables, or context (1 point), and information unclear or unavailable (0 points). The 6 criteria, and subsequently the overall methodological quality, were then classified as weak (<60%), moderate (60%-85%), or high (>85%) based on the percentage of the 34 maximum possible points achievable. Any discrepancies in scoring between the 2 reviewers were resolved through discussion to reach consensus, with any unresolved issues settled by a third reviewer.

Results

Study selection

The PRISMA flowchart (Figure 1) summarizes the study selection process. The combined search strategy identified 303 articles after Covidence’s automated and manual de-duplication process. Following title and abstract screening, 21 articles were eligible for full-text review. An additional 17 reports were added to full-text review identified through manual searches, including 6 obtained from regulatory authorities (25 responses from 181 contacted). Of these 38 full-texts, 13 were national reports from the United Kingdom, United States of America, Japan, France, Latvia, Sweden, Germany, Estonia, Denmark, Greece, Malta, Italy, and Ireland. Following the full-text review, based on inclusion and exclusion criteria, 14 articles and 5 national reports were included for quality appraisal and analysis. The detailed reasons for the exclusion of 19 studies after the full-text review is documented in Table S3.

Study characteristics

Figure 2 summarizes salient characteristics of the included studies at a glance, with a more detailed overview of the extracted data presented in Table 1 of the national (n = 10), typical (n = 5), local (n = 3) and regional (n = 1) DRLs. The 10 national DRLs (NDRLs) emanated from 5 national reports, 4 journal articles, and 1 thesis. In accordance with ICRP 129,¹⁵ DRLs were established using either kerma-area product (P_ka) or dose-area product (DAP). The majority of studies (n = 13) reported DRLs for specific clinical indications, the remainder based on FOV dimensions or grouped indications. Of the 9 studies that detailed the make and model of CBCT units surveyed, Planmeca (Helsinki, Finland) devices^16–23 were most prevalent. Other commonly reported manufactures included Carestream (Atlanta, Georgia)^18,20–23; Soredex (Tuusula, Finland)^18,20,22; Sirona (Charlotte, North Carolina, USA)^18,21,22; NewTom (Imola, Italy).^18,20,24

Figure 2.

Sunburst chart displaying the proportional distribution of key characteristics across 19 included diagnostic reference level (DRL) studies/reports: study population (adult/pediatric), geographical region, DRL type (national, local, typical, and regional), and DRL categorization (clinical indications/FOV). Abbreviations: A & P = adults and pediatrics; DRL = diagnostic reference level; FOV = field of view; NS = not specified; USA = United States of America.

Table 1.

Characteristics of the national, local, typical, and regional DRL studies.

Country (year of publication)	Data collection period	Number of data	Study population	Dose measurement	Dose indices	QC	Quality score	FOV (D × H, cm, area, cm2)	DAP range	DRL based on:
National DRL
S. Korea (2011)	2011	14units	Adults	Free in air	DAP		M	16 × 18	476-3960	FOV
Finland (2016)	2015	45insts	Adults	CBCT readout	DAP		M		27-2401	CIs: 1, 2, 3, 4, 6
		18insts	Pediatrics						27-390	CI: 5
USA (2019)	2014-2015	28units	Adults	CBCT readout	DAP		M		218-2497	Grouped Cis
			Pediatrics						65-1806
Portugal (2020)	2019	69units	Adults	Free in air	Pka	Yes	S	4 × 4-14 × 14	79-3992	CIs: 1
Switzerland (2020)	2019	227insts	Adults and pediatrics	CBCT readout	Pka		M	16-140		CIs: 1, 2, 3, 4, 7
UK (2020)	2014-2017	185units	Adults		DAP		M	27-253	112-458	CIs: 1
		117units	Pediatrics					27-160	62-335	CIs: 5
Latvia (2021)	2020-2021	19insts	Adults		Pka		M		100-1400	CIs: 1, 2
France (2022)	2021-2022	150insts	Adults		DAP		M			CIs: 1, 2, 8
S. Korea (2025)	2023	301insts	Adults	Free in air	DAP		S		110-9388	CIs: 1
			Pediatrics						46-9291	CIs: 5
Japan (2025)	2024	30insts	Adults	Free in air	Pka		M	≤40 to >100		FOV
Local DRL
Tamil Nadu (2021)	2018-2020	10insts	Adults	Free in air	Pka	Yes	S	5 × 5-15 × 9	176-890	FOV
Iran (2021)		10insts		DICOM	DAP		M			CIs: 1, 2, 3, 4, 7
Missouri—Illinois (2023)	2024	25insts	Adults	CBCT readout	DAP	Yes	M	4 × 4 and 5 × 5	267-1064	CIs: 2
Typical value or institutional DRL
Malaysia (2017)	2015-2017	1insts	Adults and pediatrics	System database	DAP		M		392-1254	Grouped CIs
UAE (2021)	2021	1insts	Adults	System database	DAP	Yes	S		573-1689	Grouped CIs
Hong Kong (2022)	2016-2021	1insts	Adults	CBCT readout	DAP	Yes	S	≤40 to >100	23-4584	CIs: 1, 2, 3, 4, 9, 10
			Pediatrics							CIs: 5
Sri Lanka (2024)		2insts		System database	DAP		M	25 to ≥ 105	268-2139	CIs: 1, 3, 6, 11
Iran (2025)	2023-2024	1insts	Adults and pediatrics	System database	DAP	Yes	S	8 × 8	163-739	Cis: 1, 3, 6 Grouped CIs: males/females
Regional DRL
IAEA (2019)	2015-2017	14insts	Adults	CBCT readout	Pka	Yes	S	36-231	236-2567	CIs: 1, 12 Grouped CIs: A/P
			Pediatrics					24-100	146-796

Clinical indication codes: 1, implant planning; 2, endodontics; 3, dentoalveolar pathologies/maxillofacial evaluation; 4, third molar; 5, supernumerary tooth; 6, sinus and nasal pathologies; 7, tooth and position anomalies; 8, impacted tooth (exodontia); 9, periodontal bone defects; 10, orthognathic evaluation; 11, TMJ; 12, developmental disturbances. Abbreviations: A = adult; CIs = clinical indications; DAP = dose area product; DRL = diagnostic reference level; FOV = field of view; IAEA = international atomic energy agency; M = moderate; P = pediatric; P_ka = kerma area product; QC = quality control; S = strong; S. Korea = South Korea; USA = United States of America; W = weak.

The most active period for CBCT DRL publications was 2020-2021 and 2025, during which 9 studies (47%) were published. However, the majority of data collection occurred in 2015 and 2019. Before this surge, the foundation for CBCT optimization initiative was laid by the United Kingdom,²⁵ and of the reviewed articles/reports, the earliest national dental CBCT DRL emanated from South Korea (2011).²⁰

Methodological quality assessment

A detailed breakdown of the quality assessment for each study, along with the rationale for rating, is presented in Table S4. Seven studies/reports (37%) were rated as strong, while the remainder (63%) received moderate rating. All strongly rated studies reported key elements as per ICRP guidelines; QC assessment; institution diversification; DRL per clinical indications with 1 exception²⁶; and FOV details with 2 exceptions.^19,27 Study design, recruitment transparency, and data handling (restrictions on routine practice²¹) were rated as strong in 95% of studies/reports. Conversely, compliance with DRL reporting as per ICRP guidelines received weak ratings in 26% and protocol transparency classified as strong in 42% of included studies/reports.

Of the NDRL studies, 20% were rated as strong, compared with 60% of the typical value studies. Some studies applied terms inconsistent with ICRP recommendations (Table 2). Two local DRL studies^26,28 simply reported a DRL without stipulation of type and a third,²¹ incorrectly reported a local DRL as a regional DRL. Typical values were all correctly categorized^16,19,24,29 with 1 exception.¹⁷ The aforementioned study¹⁷ did not categorize the DRL and reported the third quartile value instead of the appropriate measure (median), additionally reporting identical values for the maximum, mean, and third quartile, which is inconsistent with the box plot that was presented in the study.

Table 2.

International Commission on Radiological Protection (ICRP)⁴ definition of regional, national, local, and typical/institutional diagnostic reference levels (DRLs).

DRL	Data collection region	Statistics
Regional	Group of countries with comparable practices, typically within the same continent	Median value of each country’s DRL or 75th percentile of dose data distribution
National	Healthcare facilities from an entire country; typically, 20-30 facilities, or 30%-50% of all facilities in smaller countries	75th percentile of the median values from each individual healthcare facilities
Local	A limited number of healthcare facilities within a country (approximately 10-20)	75th percentile of the median values from each facility
Typical or institutional DRL	A single facility equipped with either multiple CBCT units or a single unit	Median value of the dose distribution, based on at least 20 datasets per clinical indication

Abbreviation: CBCT = cone-beam CT.

Reporting gaps were also identified; exposure parameters were absent in 6 studies^{16,22,23,30–32}; FOV details were missing in 8 studies,^{17,19,22,23,27,28,30,33} and dose measurement methodologies were unclear in 3 national reports.^23,30,31 Data collection period was not explicitly reported in 5 studies,^{20,21,28,29,32} the target population was unspecified (adult or pediatric) in 2 studies,^28,29 and methodological information was not provided in 3 national reports.^23,30,33 In 1 study,²¹ practitioners were instructed to use FOV size of either 4 × 4 cm/5 × 5 cm, while all other exposure parameters followed existing clinical protocols. In another study,²⁸ the reported DRL values differed between the abstract and the tables.

Dose measurement and quality control

Figure 3 illustrates the dose measurement methodologies used across all DRLs types. National and local DRL studies commonly employed free-in-air measurements using external dosimeters or recorded CBCT readouts under standard or optimized exposure settings. Two studies^27,34 reported that approximately one-quarter of the CBCT devices did not display dose values, possibly attributable to use of older devices. Patient dose data recorded from system databases were predominantly used for typical value studies, and innovatively, 1 study²⁸ recorded patient dose via Digital Imaging and Communication in Medicine Structured Reporting (DICOMSR) with a cloud-fog software architecture to enable automatic local DRL establishment over 6 phases. However, changes in exposure parameters were not reported on a per-phase, per-indication basis.

Figure 3.

Dose measurement methodologies across DRL types: distribution of dose measurement methods used in establishment of national (n = 10), local (n = 3), regional (n = 1), and typical (n = 5) diagnostic reference levels (DRLs). Reported methods include free in air measurements using dosimeters, CBCT unit readouts, DICOM data extraction, and system database records. No details; Dosimeters; CBCT readout; DICOM; System database.

Of the 7 studies that used patient dose data, 3^16,18,28 recorded prospectively, the remainder^17,19,24,29 using retrospective dose data. Additionally, Japan also reported air kerma at the rotation center (Kiso). In the 2025 report,³⁵ KISO was constant (17 mGy) across all FOV categories, whereas in the 2020 report,³⁶ KISO varied with FOV size, this being attributable to the impact of differing measurement methodology.

Studies that additionally reported effective dose confirmed that the salivary glands received relatively high effective doses^19,24 compared with other head and neck organs, especially during implant, sinus and lesion assessment and endodontics where the need for dose optimization is pertinent.^37,38 Some studies (37%) confirmed dosimeter calibration and QC tests were performed routinely; others omitted such information (Table 1). Notably, only 1 study²² quantified the display dose percentage (DDP) error from the QC report and applied a correction factor to the reported values to approximate true DAP values. Two additional studies^26,34 verified that the DDP was within the manufacturer stated tolerance level, whereas the remaining studies did not report any assessment or validation of DDP accuracy.

Exposure parameters

The mAs varied considerably across clinical indications and FOV categorizes. Exposure/scan durations ranged from 2.3 to 36 seconds, with no marked difference between medium and large FOV examinations, the longest scan durations (36 seconds) being observed for medium FOVs.²⁹ To clarify, all studies reported exposure time (ie, total time of exposure to X-ray beam) except 2^21,38 which reported a scan time, not clarifying if this scan time equaled exposure time or if dose reducing, pulsed exposure was being used. Small FOV examinations, for example, endodontic examinations (FOV: 4 × 4 cm and 5 × 5 cm²¹) reported varied scan durations ranging from 9 to 27 seconds. In contrast, the identified range of tube voltage varied less, ranging from 50 to 120 kVp, with 4 studies using a fixed 90 kVp.^16,17,19,29 One study³⁴ did identify that only 12 distinct kVp setting were available among 29 different CBCT models from 19 different manufacturers.

Field of view

Field-of-view data (Figure 4) was reported in over half of the studies/reports (n = 11), with values ranging between 16 and 288 cm². Of these, 5^{16,18,29,32,35} categorized FOVs based on surface area (height × diameter), 2^26,34 by height, and 1²⁰ categorized without specifying the classification criteria. The remaining 3^21,24,31 did not categorize FOVs.

Figure 4.

Diagnostic reference levels based on field of view dimensions: distribution of DRL values across FOV dimensions. Abbreviations: DRL = diagnostic reference level; FOV = field of view. ≤40 cm²; >40 to ≤100 cm²; >100 cm².

Small FOV

DRLs for small FOVs (≤40 cm²) ranged from 333 to 893 mGycm², with all exceeding 500 mGycm² except one.¹⁶ Endodontics^16,21,29 was the most common indication within this category. Voxel sizes were detailed in only 2 studies: one¹⁶ indicated that over 50% of examinations used voxel sizes >150 to ≤300 µm, 40% used ≤150 µm, and less than 5% used >300 µm, while the other²¹ reported voxel sizes between 75 and 125 µm for small FOV examinations.

Medium FOV

Medium FOV (>40 to ≤100 cm²) DRLs^{16,18,24,29,31,34,35} ranged between 265 and 1584 mGycm² and were most commonly associated with pre-implant planning, maxillofacial evaluation, periodontal bone defects, and sinus pathology. It was¹⁶ observed that 70% of the medium FOV examinations were performed with voxel sizes between >150 and ≤ 300 µm, 18% with ≤150µm and 12% with >300µm.

Large FOV

Large FOV (>100 cm²) DRLs^{16,20,29,34,35} ranged from 229 to 3203 mGycm². Two studies^16,34 specified the clinical indications, including placement of upper first molar implants, third molar assessment, maxillofacial evaluation, orthognathic surgery, and developing dentition in pediatric patients. The lowest DRL identified for a large FOV (229 mGycm²)¹⁶ was significantly lower than that for small and medium FOV DRLs, emanating from a survey where 93% of the scans were performed with low-dose mode/voxel size >300 µm, in over 95% of cases, primarily for orthodontic and maxillofacial evaluation.

Correlation analysis

Analyzing the included studies, FOV was identified as having a moderate to strong positive correlation of 0.8518, 0.7520, 0.4834 with dose values. In contrast, tube current-exposure time product (mAs) was identified as having a low positive (r = 0.31)³⁴ and weak negative correlation (r = −0.042)²⁰ with dose, while tube voltage (kVp) showed a weak positive (r = 0.103),²⁰ (r = 0.03)¹⁸ correlation. A weak positive correlation (r = 0.175)¹⁷ was reported between patient age and dose level.

Clinical indications

It is apparent from Figure 5 that implant planning, endodontic assessment, dentoalveolar pathology/maxillofacial evaluation, third molar assessment, and pediatric examinations were the most frequently reported indications. The DRLs corresponding to these indications are examined in greater depth subsequently. Periodontal,¹⁶ orthognathic,¹⁶ developing dentition,¹⁶ obscure pain condition,²² TMJ,²⁹ and single impacted tooth³⁰ were less frequently reported indications, each with only 1 reported DRL among all included studies. A complete tabulation of all DRLs from the reviewed studies/reports (in addition to national reports which were excluded due to insufficient methodology) is presented in Tables S5 and S6.

Figure 5.

Diagnostic reference levels across clinical indications: multigraph showing DRL values across clinical indications from the 19 included studies/reports. Other NDRL indications: obscure pain condition and exodontia, other typical/regional DRL indications: orthognathic evaluation, periodontal bone defects, developmental disturbances, and TMJ. Numbering under each bar represents the countries from 13 studies that provided DRLs for specific clinical indications: 1. United Kingdom, 2. Finland, 3. Switzerland, 4. Latvia, 5. France, 6. Portugal, 7. South Korea (2025), 8. Iran (2021), 9. Iran (2025), 10. Hong Kong, 11. Sri Lanka, 12. IAEA, 13. Missouri—Illinois. Abbreviations: DRL = diagnostic reference level; NDRL = national diagnostic reference level; TMJ = temporomandibular joint. National DRL; Local DRL; Typical DRL; Regional DRL.

Clinical indication based DRLs were reported for site-specific individual indications^{21,22,27,31,34} (eg, implant planning-maxillary first molar, third molar assessment). Diagnostic reference levels were also reported for grouped indications, combining subcategories of the same indications (eg, all implant planning: single/multiple/peri-implantitis/mandible/maxilla, all dentoalveolar pathologies)^{16,23,24,28,29,32} allowing comparison with DRLs reported for similarly grouped subcategories. In contrast, some DRLs reflected cross-indication groupings, making their DRLs not directly comparable^18,19 (Tables S5 and S6). Inconsistencies existed in the terminology used for identical indications, the reporting of exposure parameters and FOV details.

Implant planning

Specific DRLs for implant planning (Table 3) were addressed in 12 studies producing 14 DRLs. Reported collective DRLs¹⁹ that included implant assessment but grouped it with indications that would necessitate diverse imaging protocols were excluded from Table 3.

Table 3.

Diagnostic reference levels and exposure factors/parameters for implant planning.

Country/region	Terminology	DRL		Exposure parameters
		Numbers of data	DRL (mGycm2)	kVp	mA/mAs	Exposure time (sec)	FOV (D × H, cm, area, cm2)
Implant planning
UK	Maxillary first molar	185insts	265	–	–	–	27-253
Portugal	Maxillary first molar	69units	820	60-100	3-12Δ	2.3-24	16-196
S. Korea 2025	Maxillary first molar	301insts	1856	75-120	2.8-13Δ	5-24	–
Iran 2021	Single implant	10insts	310	–	–	–	–
Iran 2025	Implant	110patients	341	110	8.62∇	3.6	8 × 8
Finland	Single missing tooth	32insts	360	–	–	–	–
	Imaging of maxilla or mandible	28insts	610	–	–	–	–
Hong Kong	Implant (implant planning in 1 sextant; at least 2 sextants; peri-implantitis)	837patients	458	–	–	–	40-100
Switzerland	Single implant on maxilla and mandible	170insts	683	–	–	–	20-100
France	Single implant without guide and without sinus lift	150insts	700	–	–	–	–
Latvia	Implantation in upper or lower jaw	19insts	685	–	–	–	–
	Implantation in both jaws	19insts	1000	–	–	–	–
Sri Lanka	Implant placement: Assessment of mandible, maxilla, pre-maxilla, dilaceration, implant assessment	390patients	1307	–	–	–	–
IAEA	Implant	135patients	1584	85	71∇	–	75

$\Delta$ , mA; $\nabla$, mAs; - data not available. Abbreviations: DRL = diagnostic reference level; FOV = field of view; IAEA = international atomic energy agency; kVp = tube voltage; mA = tube current; mAs = tube current-time product; S. Korea = South Korea; UK = United Kingdom.

DRLs ranged from 265 to 1856 mGycm², the lowest being the UK NDRL³¹ (2020) and South Korea NDRL²⁷ (2025) being the highest, both referring specifically to pre-implant assessment for the maxillary first molar, as did one other NDRL.³⁴ Two studies^22,23 reported 2 separate DRLs; single and multiple sites implant assessment. No DRLs were reported for implant planning that necessitated surgical intervention involving the maxillary sinus, for grafting purposes or for zygomatic implants.

Although exposure parameters and FOV reporting was inconsistent among studies, available data enabled some comparison. Comparing 2 NDRL studies^27,34 for adult implant planning, where mean kVp was similar, the study²⁷ reporting higher mean mA (↑75%) and scan durations²⁷ (↑200%) similarly reported markedly higher implant DRLs (↑126%). Small (<40 cm²) and medium (>40 to ≤100 cm²) FOVs were most frequently used for implant assessment, with a minority of studies^16,31,34 reporting large FOVs (>100 cm²) as well. Voxel size was reported in 1 study,¹⁶ ranging from <150 to >300 µm, with 150-300 µm being the most common (65%). An optimized protocol was used in about one-third of the scans performed in 2 studies.^16,22

Endodontics

Endodontics DRLs (n = 7) were generally reported under the broad category of endodontics; and even where subcategories were identified,^16,22 only a single DRL was reported (Table 4). Diagnostic reference level values ranged from the lowest¹⁶ of 373 mGycm²(20% of scan performed using low-dose modes) to highest²¹ of 838 mGycm², the latter reporting a DAP range of 267-1064 mGycm². Considering only NDRLs, the Finnish DRL²² of 550 mGycm² (25% of examinations performed with optimized protocols) was the lowest, 14%-21% lower than those reported by Switzerland, Latvia, and France. A smaller FOV (≤40 cm²)^16,21,32 and smaller voxel size¹⁶ (≤150 μm) were most commonly employed, whereas exposure parameters were inconsistently reported, with only 1 study²¹ providing this detail.

Table 4.

Diagnostic reference levels and exposure factors/parameters for endodontics, dentoalveolar pathologies, third molar, and pediatric examinations.

Country/region	Terminology	DRL		Exposure parameters
		Number of data	DRL (mGycm2)	kVp	mA/mAs	Exposure time (sec)	FOV (D × H, cm, area, cm2)
Endodontics
Hong Kong	Periapical lesion; root canal morphology; root fracture	553patients	373	–	–	–	≤40
Iran 2021	Endodontics	10insts	480	–	–	–	–
Finland	Periapical lesion and root canal morphology	30insts	550	–	–	–	–
Latvia	Endodontics	19insts	610	–	–	–	–
Switzerland		53insts	639	–	–	–	25-40
France		150insts	700	–	–	–	–
Missouri—Illinois		25insts	838	85-120	3-14Δ	9-27a	16 and 25
Dentoalveolar pathologies
Hong Kong	Maxillofacial evaluation	1203patients	229	–	–	–	≤40 to >100
Iran 2025	Lesions	23patients	387	110	${7.5}^{\nabla }$	3.6	8 × 8
Finland	Jaw cysts and tumours	28insts	500	–	–	–	–
Switzerland	Dentoalveolar pathologies (eg, cyst, periodontal, and periapical lesions)	90insts	569	–	–	–	16-100
Iran 2021		10insts	510	–	–	–	–
Sri Lanka	Pathological conditions including cysts, tumours, and lesions	44patients	1013	90	–	–	–
	Surgery assessment	22patients	1266	90	–	–	–
	Facial trauma	177patients	1585	90	–	–	–
Third molar
Hong Kong	Third molar	398patients	229	–	–	–	≤40 to >100
Finland	Assessment of the relationship between a wisdom tooth and the mandibular canal	38insts	380	–	–	–	–
Iran 2021	Third molar	10insts	410	–	–	–	–
Switzerland	Wisdom tooth	176insts	662	–	–	–	16-100
Pediatric examinations
Hong Kong	Developing dentition (tooth eruption, supernumerary teeth, and occlusal problems)	50patients	130	–	–	–	≤40 to >100
UK	Impacted maxillary canine tooth	117insts	170	–	–	–	27-160
Finland	Unerupted and supernumerary tooth (excluding wisdom tooth)	18insts	290	–	–	–	–
S. Korea 2025	Mesiodens or impacted canine	301insts	1350	65-100	2-13Δ	5-24	–

^aScan time cited, not exposure time. $\Delta$, mA; $\nabla$, mAs; - data not available. Abbreviations: DRL = diagnostic reference level; FOV = field of view; kVp = tube voltage; mA = tube current; mAs = tube current-time product; S. Korea = South Korea; UK = United Kingdom.

Dentoalveolar pathologies/maxillofacial evaluation

This was the third most commonly reported DRL (n = 6) with the clinical indication having the most inconsistent/varied terminology (Table 4). Some studies reported^28,32 under the broad category of dentoalveolar pathologies/maxillofacial evaluation, 1 study¹⁶ further identified 5 subcategories but grouped them under a single DRL value (229 mGycm²). Alternatively, DRLs^22,29 were largely reported under the subcategory tumours/cysts/radiolucencies where the values were 500 and 1013 mGycm². Of the 3 studies for this indication that reported FOV, 1³² predominantly used a small FOV (25 cm²), another²⁴ employed a fixed medium FOV (8 × 8 cm²), and the third¹⁶ primarily used large FOVs (>100 cm²) with larger voxel sizes (>300 µm), with 84% of examinations performed using low-dose protocols.

Third molar assessment

Four identified DRLs were reported under third molar assessment ranged from 229 to 662 mGycm². Two studies employed optimized protocols, for 29%²² and 92%¹⁶ of the scans performed. In the latter, about 40% of the scans were acquired using large FOVs (>100 cm²), and 60% with large voxel sizes (>300 µm). Exposure parameters were not reported, and FOVs^16,32 ranged from 16 to 100 cm².

Pediatric DRL

Nine studies incorporated pediatric cohorts in their dose assessment. Of these, 6 studies^{16,18,22,27,31,33} explicitly established DRLs for pediatric patients (Figure 5), the remaining 3 studies^17,24,32 reported DRLs collectively, combining adult and pediatric dose data. Of the 6 pediatric specific DRLs, 2^18,33 were excluded from the present analysis since their datasets were smaller than the ICRP recommended minimum of 20 examination (IAEA: 1 data; NEXT: 6 data). However, the Finnish pediatric NDRL²² was included, as its dataset (18 data) was close to the ICRP recommendation. Variations were observed in age categorization: 2 studies^16,18 categorized pediatrics as <16 years, 2 studies^19,24 as <18 years, and 2 studies^31,33 specifically referred to 12-year-olds. The lowest pediatric DRL was reported in the United Kingdom³¹ (170 mGycm²) and the highest in South Korea (1350 mGycm²), this having increased from 1208 mGycm² in a comparable South Korea-based study in 2019.²⁷ Two studies reported optimized protocols, performed in 44%²² and 96%¹⁶ of cases.

Prevalence of dose optimization

Only 21% of reviewed studies explicitly reported the use of optimization strategies, such as low-dose, 180° acquisition protocols, and employing a large voxel size. In contrast, the Japanese national report³⁵ indicated that practitioners were only using standard scan mode (107/115 datasets) and high-resolution mode (8/115) in all 30 centers surveyed. One study¹⁶ did report that low dose protocols were consistently used across all indications until image quality (IQ) feedback identified that this was inappropriate for certain indications except pediatric examinations. Studies^30,32 also highlighted that the majority of CBCT operators relied on manufacturer setting and had limited knowledge on manipulation of CBCT settings, using large FOVs irrespective of clinical indications²⁹ or whether patients were pediatric²⁷ (cited indications were: mesiodens, impacted canine).

Wide interquartile ranges (IQRs) and SDs reflected the wide variation in exposure protocols and parameters for the same clinical indications, SD (886 mGycm²) exceeding the reported DRL (820 mGycm²)³⁴ for a single implant planning DRL. Additionally, in some cases, IQRs were approximately two-thirds of the reported DRL values^16,27,32 compared to an IQR of approximately one-third of the corresponding implant planning DRL value in the UK NDRL³¹ report.

The diversity in protocols obviously stems not only from operator preferences but also from differences in equipment configuration and manufacturer specifications/capabilities, which also impacts on DRL establishment. Obviously, FOV and voxel size options were not homogenous among manufacturers, some models offering fewer options than others.^20,21 Cone-beam CT filtration variations were identified among makes and models¹⁸ (2.5-4.5 mm aluminum [Al], 0.5 mm copper [Cu] + 2.5 mm Al). The impact of filtration was effectively demonstrated in 1 survey²¹; lower displayed and measured DAPs were reported at higher mAs setting than at lower mAs setting, with all other exposure parameters and the FOV remaining constant, due to the presence of additional 0.2 mm Cu filter in one of the units, preferentially absorbing low energy photons and altering the beam spectrum. This underscores the importance of comprehensive and transparent reporting for a meaningful dose comparison across CBCT units.

CBCT operators

Cone-beam CT operators (those making the exposure) varied according to regulatory requirements of the reporting country. Dentists/physicians^{16,21,22,26,30} and radiographers^19,22,26 were the most common operators of CBCT units. The thesis on Finnish NDRLs stated that dental assistants and hygienists collectively performed CBCT scans more commonly than dentists as is permissible under Finnish regulations (supervising dentist: present during scan, completed supplementary CBCT training) but identified that some supervising dentists had not completed this required training. According to French regulations, dental assistants and hygienists are not authorized to perform CBCT scans; however, it was reported that exams were performed routinely by these groups in addition to dental practice secretaries. The French national report also identified practitioners’ limited knowledge of device settings and protocols, and noted error, gaps and inaccuracies exist in manufacturers’ technical specification and user guides.

Meta analysis

A statistician confirmed that due to the heterogeneity of the included studies a meta-analysis was not performed.

Discussion

The main aims of this SR were to identify key methodological elements that may undermine the quality of DRL surveys while identifying factors that limit DRL comparison; assess the level and methods of dose optimization; and clarify the obstacles both in dose optimization and DRL establishment.

Methodological quality assessment

Quality ratings on adherence to ICRP guidelines and protocol transparency were often moderate or weak. This was largely associated with apparent unfamiliarity or misinterpretation with certain elements of ICRP guidelines on DRL establishment and the absence of proper retrospective dose data management systems, resulting in inconsistent reporting of exposure parameters, FOV, dose levels, and clinical indications. In addition, lack of awareness on DRL categorization, awareness of dose optimization strategies, and the use of inappropriate statistical measures further contributed to incomplete or inaccurate reporting. Considering the vast differences in the diversity of practices and technologies included in a local compared to national or regional DRL survey, misclassification of DRL type could result in a failure to prompt optimization where needed or prompt excessive dose reduction, thereby compromising IQ.

Unsurprisingly, NDRL studies were shown to be most susceptible to such methodological errors, typically receiving a moderate quality rating in this SR; data were collected from multiple institutions within a nation, often via mailed questionnaire taking 1-3 years for collection, analysis, and publication. The accuracy and completeness of these questionnaires depended heavily on participants cooperation, knowledgeability, and awareness on dose optimization. In contrast, typical value studies generally involve a single institution, allowing researchers more control over fulfillment of data collection and QC verifications, consequently, most receiving a strong rating in this SR.

Actual patient dose data, although ideally recommended,³ were not favored by NDRL studies due to logistical and practical difficulty of obtaining consistent information on patient dose data, exposure parameters, and clinical indications across multiple facilities. Instead, CBCT readouts and manufacturer radiation dose tables provide readily accessible dose data that can be collected quickly subsequent to confirming the settings used for the particular examination protocol in that facility, without the need for a dedicated dose management system and with minimal practitioner involvement and limited ethical approval for studies. Nevertheless, accurate dose values remain essential for reliable DRLs and only a few studies documented their attempts to ascertain this, applying dose display error correction²² and ensuring the values were within tolerance level.^26,34

Incorporating data from all facility types, encompassing an array of equipment technology; clinical practices (general and specialized); imaging protocols; and medical physics support, ensures that NDRLs are both realistic and broadly applicable, ultimately supporting optimization efforts across an entire countries healthcare system. Japanese DRLs³⁵ were computed solely from data collected at university hospitals (n = 29/30). These DRLs were substantially higher than other studies, mainly due to the use of manufacture’s default protocols (n = 108/115) and the absence of optimized protocols. This suggests that there is still lack of knowledge and initiatives on dose optimization from healthcare personnel involved in imaging and perhaps limited input or control from a regulatory authority.

Updating DRLs is advised every 3-5 years,³ in order to account for continuous technological advancements and clinical practices, enhancing the relevancy and reliability of DRL values. To date, based on reviewed NDRLs studies and reports, South Korea and Japan appear to be the only nations that have published revised NDRLs. Most other NDRLs have not yet exceeded the ICRP recommended revision period, with 2 exceptions, established 6³³ and 9²² years ago which may no longer accurately reflect current clinical practice or technological advancements.

Comparing DRLs

Clinical indication vs FOV based DRL

Figure 4 highlights that similar FOV categories report markedly varied DRLs, due to the same FOV categories being used for diverse clinical indications, each requiring different IQ for accurate diagnosis and hence acquisition protocols, this also being influenced by operator preferences and knowledge. Differing practices with regard to selection of FOVs for particular diagnostic tasks were evident, with some countries reporting a wide range of FOVs from small to large, even for single implant planning,^16,18,31 and in some cases, large FOV were used irrespective of indications; this was evident for pediatric cases also.²⁷ Furthermore, the lower IQ required for some diagnostic tasks permitted larger voxel sizes and low-dose mode, resulting in significantly reduced dose levels even when using large FOVs.¹⁶ Therefore, FOV-based DRLs are a very crude assessment. Diagnostic reference levels should be predominantly influenced by the clinical task, rather than FOV size, otherwise comparisons between institutions or studies becomes meaningless. This limits the ability of DRLs to fulfill their purpose of identifying unusually high or low dose levels for a given clinical examination.

Normalizing DRL values to a commonly used 5 × 5 cm FOV facilitates standardized inter-CBCT comparison and is useful for approximate benchmarking. However, the normalization formula applied in this study³² implicitly assumes proportional area-dose scaling, whereby a given percentage reduction in FOV results in an equivalent percentage reduction in dose. This assumption does not fully capture CBCT dose behavior across different FOVs. Empirical measurements demonstrated that dose does not scale linearly with FOV even when exposure parameters are held constant,^21,40 likely reflecting the effects of factors such as beam non-uniformity, collimation characteristics, rotation geometry, and scatter behavior. Therefore, FOV-normalized DRLs should be interpreted cautiously, as they introduce uncertainty and may not accurately represent the true dose output of individual CBCT systems.

The reviewed studies categorized the FOV in 2 ways; one was based on height^26,34 while others were based on surface area.^{16,18,29,32,35} Diagnostic reference level establishment is generally based on FOV area, the dose metric (dose/kerma-area-product [DAP/KAP]) being calculated by multiplying air kerma and beam area. Height-based categorization is considered more appropriate for computing effective dose,⁴¹ organ exposure increasing with height.

As recommended,³ indication-specific DRLs largely facilitated a more direct comparison and insight into the sources of DRL variation. It was evident that facilities that performed scans using small voxel sizes (70-80 µm) and standard manufacture protocols reported higher DRLs,^21,27,35 whereas centers that used large voxel sizes (≤150 and >300 µm) and in combination with optimized protocols reported lower DRL values.¹⁶ Facility type also contributed to this variation: university hospitals, with access to medical physicist and often superior technologies generally reported lower doses^16,18,24 than private dental clinics^21,34 where dose optimization knowledge and support was much reduced. Reporting DRLs according to specific clinical indications, ensuring that all relevant exposure parameters and QA assessment details are documented, permits meaningful comparisons. This supports identification of the underlying causes of dose variability and targeted corrective actions, such as protocol modification, provision of additional training on optimization practices, or implementation of system advances or software updates.

Heterogeneity

Comparison and validity of published DRLs can be limited by the lack of a uniform approach and/or terminology across studies, as evidenced for implant planning and dentoalveolar pathologies (Tables 3 and 4). Heterogeneous groupings of clinical indications into a single category to establish DRLs^16,19,29 were due to insufficient data (ICRP guidelines: <20 datasets) or to inconsistent terminology, contributed to by questionnaires not being completed by a dentist (eg, medical physicists not fully familiar with the nuanced naming conventions used for clinical indications).

Dose optimization

Operators and regulatory control

Cone-beam CT is an evolving technology, decreasing in expense, more accessible, and with a dynamic range of makes/models with accompanying diversity in IQ and optimization capabilities.³⁹ It is apparent from studies^{19,22,23,28,30,31,35} that many operators/practitioners do not understand technological/physics terminology or dosimetric data nor do they understand how hardware design and beam conditioning influence dose output, which is essential for informed CBCT unit procurement, protocol selection, and effective dose optimization in clinical practice. Additionally, some operators appear unwilling or unable to adapt protocols according to clinical/IQ requirements.²⁶ Regular support from an appropriately trained MPE, familiar with dental imaging and its requirements, is essential to optimize imaging practices and close the clinical-medical physics knowledge gap. This review identified that some countries still lack the legislative basis for radiation safety and appropriate infrastructure for effective patient dose management.^18,29 This review also observed instances of weak regulatory authority control^22,29,30,34 associated with poor communications of practitioners’ regulatory obligations, resulting in practitioner flagrant noncompliance with national CBCT regulations, CBCT training, and general radiation protection requirements.

Manufacturers/engineers

Evidently, some CBCT units had fixed kVp/limited mA adjustment ranges, varied FOV size capabilities, and manufacturers’ default settings often provided unnecessarily high IQ for various diagnostic tasks. Inaccuracies in manufacturers’ technical specification and user guides³⁰ and wide variations in FOV/exposure setting configurations programmed by the X-ray engineers for the same CBCT make and model and examination at different facilities lead to concomitant variations in dose.²⁶

Manufacturer standardization of FOV options or freely adjustable FOVs, to prevent exposing anatomical regions beyond those required for diagnosis and mandatory dose display¹⁵ in addition to accurate documentation of specifications, provision of model-specific practical training for engineers and operators initially and periodically (introduction of new technology/software updates), would all aid the implementation of dose optimization strategies.

Optimized protocols

Optimized protocols based on diagnostic tasks can lead to substantial dose reduction.¹⁶ Diagnostic tasks requiring lower IQ permitted larger voxel sizes and low-dose modes, resulting in significantly reduced dose levels,¹⁶ reinforcing that DRLs should be reported according to clinical task rather than FOV size.

Achieving effective dose optimization therefore requires interdisciplinary collaboration between medical physicists, practitioners, radiologists, radiographers, manufacturers, and indeed information technology experts, particularly in the development of a national registry capable of automatically collecting dosimetric data. This would enable more frequent DRL updates and help identify practices that require additional education or training programs based on the feedback from dose audits. This approach could also increase participation rates in DRL surveys and make DRLs and optimizations, an intrinsic part of clinical practice.

Pediatric

This SR identified limited data for pediatric patients, only a few studies having established pediatric DRLs (Table 3). Some studies combined it with adult datasets due to negligible sample size.^17,24,32 Frequency of pediatric CBCT imaging is difficult to assess as often it emanates from sub-populations involving tertiary referral centers which skews the findings.⁷ One regional DRL study¹⁸ reported that pediatric patients constituted 11% of all CBCT exposures (37/335). Given the heightened radiosensitivity of children and the apparent dearth of dose optimization practices with study²⁷ showing that the same acquisition protocols are often being used for both adults and pediatrics with resultant significantly higher pediatric DRLs, attention to pediatric DRLs is essential.

Identified obstacles to dose optimization and establishing DRLs are summarized in Table 5, including responses from regulatory authorities who had ongoing surveys/inspections, but were not yet able to establish DRLs.

Table 5.

Obstacles and challenges in DRL establishment and dose optimization identified from reviewed studies/reports and cited by regulatory authorities that had ongoing surveys/inspections, but not yet established DRLs.

Obstacles and challenges in DRL establishment and dose optimization
Clinical-related factors Limited awareness/understanding of the terms justification and optimization Unfamiliarity of practitioners with optimization strategies Low response rate to radiation dose data collection survey Questionnaires often completed by dental assistants Operation of CBCT equipment by unauthorized personnel, for example, dental assistant, contrary to national regulatory requirements
Education and training Limited education and training in radiation protection and CBCT exposure parameter selection Limited understanding of CBCT exposure settings, system hardware, and their impact on dose output and image quality. Use of unnecessarily large FOVs and identical exposure protocols irrespective of clinical indication
Infrastructure and data availability Lack of adequate infrastructure to support retrospective data collection in clinics Incomplete retrospective reports, including missing clinical indication descriptors, FOV details, and exposure parameters Weak regulatory control/poor communication of regulatory obligations
Manufacturers Non standardized FOV setting and exposure parameters across different manufacturers Absence of DAP/KAP displays in certain CBCT systems
General methodological challenges Time consuming nature of DRL establishment Chances of uncertainty or potential errors in collected dose levels Non-standardized nomenclature for clinical indications Insufficient data for certain indications, particularly for sub-indications. Inconsistencies and confusion among researchers regarding different DRL types and their purposes

Abbreviations: CBCT = cone-beam CT; DAP/KAP = dose/kerma-ar-product; DRL = diagnostic reference level; FOV = field of view.

Strengths and limitations

This is the first SR on dental CBCT DRLS, focusing on methodological quality of these studies/reports and on strategies to elevate standards in DRL reporting. In addition to reviewing published studies/reports, all IAEA listed regulatory authorities/Ministries of Health were contacted in attempt to source all data; even with this, however, some relevant data may not have been identified, and this may limit the completeness of the global stakeholder representation. National CBCT DRL reports from several countries were excluded due to insufficient methodological detail or the use of adopted DRL values although these data have been included in Tables S5 and S6. This review also did not analyze the reconstruction types used in CBCT examinations as most studies provided insufficient information; however, this would be of relevance given that iterative reconstruction techniques have the potential to improve IQ and reduce dose.

Conclusion

Direct comparison of DRLs from 19 studies/reports was not always possible due to methodological inconsistency, heterogenous data collection, and reporting practices. Variability in DRL values stems from a myriad of factors, including unfamiliarity with optimized CBCT acquisition protocols, poor radiation protection education, variety of CBCT unit technologies, and muted input from regulatory authorities who appear to be reacting slowly to the surge of CBCT use. The majority of NDRLs studies/reports, rated only as being of moderate methodological quality, reflects the spectrum of systematic challenges, including inadequate retrospective dose management systems, limited familiarity with DRL categorizations, and guidelines among researchers and imaging personnel, as well as logistical barriers involved in DRL establishment process.

Key recommendations

• Effective implementation of dose optimization requires strong interdisciplinary collaboration, routine regulatory audits focusing on operator training/education and dosimetry, digitation of dose records, and development of a national dose registry to support periodic DRL updates.

• Improving the quality of DRL surveys should necessitate the adoption of comprehensive and standardized data-collection frameworks. Key elements include: clinical indication, FOV, exposure parameters, QC verification, operator training and education, optimization strategies, CBCT make and model, institution type, sample size, survey response rate, study population, statistical measures, calibration information, dose measurement method, data collection period, and IQ analysis. Consistent reporting of these parameters will improve comparability across studies and ensure that DRLs fulfil their intended role as effective tools for guiding dose optimization in clinical practice.

• Collaboration between international radiation protection bodies such as ICRP or International Radiation Protection Association (IRPA) and dental organizations such as International Association of Dentomaxillofacial Radiologists (IADMFR) could establish and promote a standardized approach for dental CBCT DRL development to elevate standards.

Supplementary material

Supplementary material is available at Dentomaxillofacial Radiology online.

Conflicts of interest

None declared.

Funding

None declared.