Investigating the Role of Ultrasound in the Diagnosis of Oral Lesions: A Scoping Review.

1
Introduction
The presentation of oral and maxillofacial lesions encompasses a wide spectrum of clinical, radiological, and histological features. According to the new 2022 World Health Organization (WHO) classification of head and neck tumors, lesions in the oral cavity and tongue are divided into non‐neoplastic (or benign), epithelial tumors, and tumors of uncertain histogenesis (Vered and Wright 2022, Muller and Tilakaratne 2022, Skálová et al. 2022). This diversity necessitates a comprehensive diagnosis approach that integrates clinical examination, imaging, and histopathology examination to ensure proper diagnosis and effective treatment planning (Kumar et al. 2025).
General dentists and specialists have access to a wide range of dental imaging modalities to enhance their diagnosis. Conventional two‐dimensional (2D) radiographs are often used as the initial imaging tool to gain information about a lesion's key features and location. While important in the initial diagnostic protocol, they are limited by a lack of three‐dimensional (3D) visualization (Suomalainen et al. 2015), imposing distortions and superimposition of adjacent structures around the lesion area, no visualization of soft tissue, affecting the clinician's interpretation and decision‐making process (Antony et al. 2020; Korostoff et al. 2016). As a result, a more comprehensive form of assessment with advanced imaging modalities is often needed to provide more diagnostic information to clinicians. These include computed tomography (CT), cone‐beam computed tomography (CBCT), and magnetic resonance imaging (MRI). CT has the key advantage of providing high‐resolution imaging of both hard and soft tissues; however, it involves the highest doses of ionizing radiation among options and falls outside the scope of practice or requisition for most dentists (Bos et al. 2023; “Safety and Protection,” 2014; Wathen et al. 2013). CBCT provides high‐resolution hard tissue imaging with a lower radiation dose than CT but with poor soft tissue contrast (Bos et al. 2023; Kumar et al. 2025; “Safety and Protection,” 2014). On the other hand, MRI can provide a great soft tissue contrast and is radiation‐free but is limited by high costs and availability (Demirturk Kocasarac et al. 2018). The strengths and limitations of these imaging modalities underscore the need to explore alternatives for faster imaging assessments while minimizing ionizing radiation exposure.
Ultrasound (US) is a noninvasive imaging modality that offers benefits to clinicians, such as low cost, portability, and real‐time imaging without ionizing radiation (Demirturk Kocasarac and Angelopoulos 2018). It utilizes electrical energy to produce sound waves that are reflected differently across tissues, creating real‐time B‐mode images (Shriki 2014). Doppler US extends B‐mode imaging by enabling vascularity visualization, as seen in lesions like hemangiomas (Derindağ et al. 2021; Rossler et al. 2011) and malignant soft tissue tumors (Yamamoto et al. 2016) and intraosseous lesions (Sumer et al. 2009).
Studies have increasingly explored the diagnostic capabilities of the US to assess both soft and hard tissues in dentistry, suggesting that US could serve as a useful chairside imaging screening modality for dentists (Almeida et al. 2019; Figueredo, Catunda, et al. 2024; Figueredo, Lai, et al. 2024). Evidence supports its utility in a range of oral lesions, including the assessment of salivary gland lesions and the evaluation of cervical lymph nodes, cystic vs. solid lesions, to assist in the diagnosis of periapical lesions and periodontal status, and to provide complementary information of oral potentially malignant disorders and oral cancers for the decision of further imaging investigation and procedures (Klein Nulent et al. 2018; Ramsubeik et al. 2020). While previous reviews have provided insights into some of these applications (Musu et al. 2016; Patil et al. 2021; Tarabichi et al. 2019), they tended to focus on specific conditions or narrow diagnostic contexts/approaches. Given the rapid advancements of US imaging technology in recent years, there is a clear need for a broader synthesis that maps the existing evidence, identifies knowledge gaps, and outlines opportunities for further research. Due to the heterogeneity of clinical applications and study designs in the current literature, a scoping review methodology was selected. The purpose of this scoping review was to provide a comprehensive overview and characterize the diagnostic potential of US in oral lesions while considering MRI, CT, CBCT, or histopathology as the reference standards. Our ultimate goal was to summarize the available literature on the performance of ultrasound and its contribution to the diagnosis process of oral lesions.

2
Materials and Methods
The reporting of this scoping review followed the Preferred Reporting Items for a Systematic Review and Meta‐analysis Extension for Scoping Reviews (PRISMA‐ScR) and the JBI Manual for Evidence Synthesis (Tricco et al. 2018; Aromataris et al. 2024). A protocol was prepared and guided all phases of this scoping review.
2.1
Research Question
The following structured question guided this review: What is the available evidence regarding the diagnostic potential of ultrasound in diagnosing oral lesions in the oral cavity compared to other imaging modalities or histopathology? It is based on the PCC framework (Koufogiannakis and Brettle 2016) in which the following principles were applied: P (Population): Patients with oral lesions; C (Concept): diagnostic potential of ultrasound to assess oral lesions; C (Context): oral lesions in the oral cavity.

2.2
Eligibility Criteria
Diagnostic studies assessing the applications of US in the diagnosis of non‐endodontic lesions in the oral cavity were included. For this study, the oral cavity was defined as the maxillo‐mandibular complex, buccal and alveolar mucosa, lips (including up to the vermillion border), the floor of the mouth (including the sublingual and submandibular glands), and the hard and soft palates. Studies including patients of all ages, regardless of medical comorbidities, were considered. Only articles written in English were considered, with no restriction on the year of publication.
The following exclusion criteria were applied: (1) Cleft/lip palate patients and prenatal US; (2) Any studies that include US assessment of endodontic lesions, including but not limited to periapical cysts, periapical granulomas, periapical abscesses as they were previously investigated (Natanasabapathy et al. 2021); (3) US utilized exclusively for tumor staging, evaluating reconstructive surgery, and guiding treatment, as these are considered interventional or exploratory applications; (4) US in salivary glands and areas other than the oral cavity as they were recently explored (Ramsubeik et al. 2020; Rogalska et al. 2022); (5) US of venous‐vascular malformation or phlebolith as as these are anatomical variations rather than pathological lesions; (6) Studies lacking proper reference standard, or reliability analyses; (7) Scoping and systematic reviews, literature reviews, letters, personal opinions, technical notes, book chapters, conference abstracts, unpublished theses, anecdotes, any in vitro and in vivo animal studies, case reports, case series were not considered.

2.3
Information Sources and Search Strategy
With the assistance of an expert librarian, search strategies for the following electronic databases were performed: OVID Medline, OVID Embase, Web of Science (All Databases), Pubmed, and Scopus. Google Scholar was used for a partial gray literature search for the first 100 results (filtered by “relevance”). For all databases, the end search date was June 20, 2025. Full search queries for each database can be found in Appendix S1.

2.4
Study Selection and Data Extraction
After searches were performed, all identified citations were exported to Covidence software, available at https://www.covidence.org/home, in which duplicates were automatically removed and cross‐checked by two reviewers (SV, SY). This software was also used to streamline the selection process, in which each reviewer independently made a selection decision using this tool. The selection process was completed by two reviewers independently (SV, SY) after calibration with review experts (FTA and CPP). In the first phase, titles and abstracts of articles were reviewed with the inclusion criteria as a basis for which articles could be considered for a full readthrough. In the second phase, the full text of those studies screened successfully in the first step was analyzed independently by the two reviewers (SV, SY), allowing us to further eliminate studies that did not meet our inclusion criteria. Any disagreements were resolved by consensus, with consultation by FTA and CPP as needed. A structured data extraction table was employed to collect the study and sample characteristics, intervention characteristics, and main outcomes.

2.5
Data Items and Measures
Diagnostic performance was assessed using sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), overall accuracy, and Interclass Correlation Coefficient (ICC) for agreement and consistency. These measures provided a comprehensive evaluation of the reliability and clinical utility of the US in detecting oral lesions.

2.6
Risk of Bias (RoB) Assessment
A critical appraisal of the potential risk of bias among the individual studies was performed to improve the understanding of their methodological limitations. The RoB assessment was conducted independently by four reviewers (SV, SY) after calibration with a systematic review expert (FTA). Disagreements were solved by a third reviewer (FTA). The Joanna Briggs Institute Checklist for Systematic Reviews and Research Syntheses tool for cross‐sectional and cohort studies (JBI's Tools Assess Trust, Relevance, and Results of Published Papers: Enhancing Evidence Synthesis, n.d.) was used for this assessment. Each study was analyzed for several qualities of study design, yielding answers of “yes”, “no”, “unclear”, or “not applicable”.

2.7
Synthesis of Results
The included studies were grouped into benign and malignant tumors. For malignancies, the US ability to assess depth of invasion, tumor thickness, and specific malignancy features was synthesized. The data were analyzed and reported using text and tables, presenting a descriptive summary of study characteristics and main results.

3
Results
3.1
Study Selection
The identification, preliminary screening, and inclusion and exclusion of studies are provided in Figure 1. There were 3512 studies screened in phase 1 and 166 studies assessed for eligibility in phase 2. A total of 34 studies were included in this scoping review.

3.2
Characteristics of Included Studies
The included studies were published from 1989 to 2024 as per a biometric analysis in Figure 2. Benign soft tissue lesions were assessed by one study (Izzetti et al. 2020), benign intraosseous lesions were assessed by one study (Lu et al. 2009), and malignancies were assessed in 32 studies. Thirty‐three studies were cross‐sectional design (Angelelli et al. 2017; Brouwer De Koning et al. 2019; Bulbul et al. 2021; Caprioli et al. 2022; Chammas et al. 2011; Choi et al. 2015; Cocker et al. 2021; Faraji et al. 2018; Iida et al. 2018; Izzetti et al. 2020, 2021; Kaltoft et al. 2024; Kodama et al. 2010; Konishi et al. 2021; Kumar et al. 2024; Kurokawa et al. 2005; Lodder et al. 2011; Lu et al. 2009; Markovic‐Vasiljkovic et al. 2023; Millesi et al. 1990; Natori et al. 2008; Nilsson et al. 2022; Noorlag et al. 2020; Ogura et al. 2013; Rocchetti et al. 2021; Shintani et al. 2001; Smiley et al. 2019; Songra et al. 2006; Takamura et al. 2022; Takashima et al. 1989; Yesuratnam et al. 2014; Yoon et al. 2020, 2021; Yuen et al. 2008). Most studies used gray‐scale or B‐mode US with frequencies ranging between 5 and 17 MHz. Two studies utilized ultra‐high frequency ultrasound (UHFUS) with 70 MHz (Izzetti et al. 2020, 2021) and two studies also explored Power Doppler US for vascularity analysis of the lesions (Lu et al. 2009; Ogura et al. 2013).
As per location and type of lesions, most assessments occurred in tongue lesions, followed by malignancies in the floor of the mouth, buccal mucosa, palate, lip, alveolar mucosa, retromolar trigone, and mandible. These assessments were conducted across various US modalities and for multiple diagnostic indications. The information of included studies is summarized in Table 1.

3.3
Results of Individual Studies
To facilitate the presentation of the results in this section, the studies were grouped as follows: (1) studies assessing benign or premalignant soft tissue lesions; (2) studies assessing benign intraosseous lesions; (3) studies assessing malignancies.
3.3.1
Benign or Premalignant Soft Tissue Lesions
The study of Izzetti et al. (2020) described the diagnostic capabilities of the US in non‐cancerous oral soft tissue lesions. It included a large sample of benign or premalignant lesions (n = 142) and a 70 MHz UHFUS to assess lesion growth pattern, echogenicity, mucosal thickness, and vascularization. The lesions were categorized as autoimmune disease (lichen planus), mucosal growths (traumatic fibromas, fibrous epulis, mucoceles, and HPV papillomas), and oral potentially malignant disorders (OPMD) (such as leukoplakia and erythroplakia). When comparing diagnoses provided with histopathology as the gold standard, UHFUS displayed sensitivity (autoimmune diseases: 91%, mucosal growths: 100%, potentially malignant lesions: 93%) and specificity (autoimmune disease: 98%, mucosal growths: 98%, OPMD: 97%) values above 90% for all soft lesions assessed. The positive predictive value (PPV) was above 95% for both the autoimmune disease (97%) and the mucosal growths (99%) but much lower for OPMD (83%). This same trend was observed in the negative predictive value (NPV), where autoimmune diseases (98%) and mucosal growths (100%) had higher values compared to OPMD (88%).

3.3.2
Benign Intraosseous Lesions
One study assessing benign intraosseous lesions investigated US capabilities in screening mandibular ameloblastomas (Lu et al. 2009). Color Doppler Flow Imaging (CDFI) was used to predict tumor proliferation by analyzing blood flow signals. The authors also predicted these proliferations from the destruction profile of the surrounding bony cortex using gray‐scale B‐mode US. The US showed a sensitivity of 100% for both indications. However, it was more specific in detecting active tumor proliferations (94%) than in predicting the lesion from cortical destruction (88%).

3.3.3
Malignancies
3.3.3.1
Tumor Thickness (TT)
Eighteen studies assessed the relationship between TT, determined by different modalities, and histopathological TT (Angelelli et al. 2017; Brouwer De Koning et al. 2019; Chammas et al. 2011; Choi et al. 2015; Izzetti et al. 2021; Kodama et al. 2010; Konishi et al. 2021; Kurokawa et al. 2005; Lodder et al. 2011; Natori et al. 2008; Noorlag et al. 2020; Shintani et al. 2001; Smiley et al. 2019; Songra et al. 2006; Yesuratnam et al. 2014; Yoon et al. 2020, 2021; Yuen et al. 2008). Of these, 10 studies assessed the ability of the US alone to estimate histopathological TT values, with seven studies finding strong correlation values for US derived TT (Kodama et al. 2010; Konishi et al. 2021; Chammas et al. 2011; Yuen et al. 2008; Songra et al. 2006; Izzetti et al. 2021; Yoon et al. 2020). Natori et al. (2008) reported a significant correlation between US and histopathologic TT (p < 0.001). Angelelli et al. (2017) evaluated the level of infiltration using TT, concluding that US demonstrated a 91% sensitivity (SN), 100% specificity (SP), PPV of 100%, and NPV of 83%. Choi et al. (2015) compared US‐derived preoperative measurements with postoperative histopathologic measurements and supported the predictability of US to measure tongue tumor size (width = 0.91, thickness = 0.97, anterior–posterior dimension = 0.90). Five studies compared the correlation between MRI and US‐derived measurements with histopathological TT (Brouwer De Koning et al. 2019; Lodder et al. 2011; Noorlag et al. 2020; Smiley et al. 2019; Yesuratnam et al. 2014). Three of these studies were in agreement that US correlated better than MRI with histopathological measurements (Brouwer De Koning et al. 2019; Lodder et al. 2011; Yesuratnam et al. 2014), however, a large degree of heterogeneity was observed with respect to the range of values of correlation coefficients (US: 0.87 (Lodder et al. 2011)—0.67 (Brouwer De Koning et al. 2019); MRI: 0.69 (Yesuratnam et al. 2014)—0.38 (Brouwer De Koning et al. 2019)). Smiley et al. (2019) reported no difference in accuracy between US (0.48) and MRI (0.40), and Noorlag et al. (2020) reported that both modalities performed equally (US: 0.78; MRI: 0.72). One study comparing the correlation of US and CT derived TT to histopathology supported the superiority of US to CT (US: 0.93; CT: 0.10) (Yoon et al. 2021). Two studies were identified comparing US, CT, and MRI with histopathological TT, with both studies supporting strong correlation values of US compared to CT and MRI (US: 0.98, CT: 0.97, MRI: 0.91 (Shintani et al. 2001); US: 0.97, CT: 0.82, MRI: 0.77 (Kurokawa et al. 2005)).
US‐derived TT was found to be strongly correlated with histopathological examination TT (Chammas et al. 2011; Izzetti et al. 2021; Kodama et al. 2010; Konishi et al. 2021; Songra et al. 2006; Yoon et al. 2020; Yuen et al. 2008), however, the varying degrees of heterogeneity in correlation values should be noted. Several studies suggested that US performs better than CT (Kurokawa et al. 2005; Shintani et al. 2001; Yoon et al. 2021) and MRI (Brouwer De Koning et al. 2019; Kurokawa et al. 2005; Lodder et al. 2011; Noorlag et al. 2020; Shintani et al. 2001; Smiley et al. 2019; Yesuratnam et al. 2014) in the evaluation of TT, while two studies reported that US performed equally to MRI (Noorlag et al. 2020; Smiley et al. 2019). One study further investigated the use of US to measure lesions in various dimensions preoperatively and compared these measurements to postoperative histopathological results. This study demonstrated that US can accurately measure the size of tumors in its width, thickness, and in the anterior–posterior dimension (Natori et al. 2008).

3.3.3.2
Depth of Invasion (DOI)
Several studies measured the US ability to assess DOI compared to histopathological standards with ICC values ranging between 0.83 and 0.96 (Table 1). Kaltoft et al. (2024) concluded that there was a strong correlation between US and histopathological DOI (0.86) and a moderate correlation between MRI and histopathological DOI (0.57), further supporting that the strong correlation between US and histopathological DOI led to more accurate T‐staging of tongue squamous cell carcinoma (SCC) (US: 86.7%, MRI: 56.7%). Markovic‐Vasiljkovic et al. (2023) showed that both US and CT correlated strongly with histopathology (0.94 and 0.86); however, a weak correlation was found for US and CT when measuring the diameter of the lesions (GD values 0.27 and −0.61). Kumar et al. (2024) reported an ICC of 0.78 for US‐DOI compared to histopathology, whereas contrast‐enhanced MRI (CEMRI) achieved a coefficient of 0.68. This study further noted that US provides better predictive values for DOI than CEMRI for lesions that are less than or equal to 5 mm. US‐measured DOI values overestimated histopathological values in three studies (Izzetti et al. 2021; Noorlag et al. 2020; Takamura et al. 2022), with values between 0.14 mm (Izzetti et al. 2021) and 4.7 mm (Noorlag et al. 2020). Conversely, one study showed that US‐measured DOI differed from histopathological‐DOI by a mean difference of −0.5 mm. In contrast, the mean difference for MRI‐DOI was 3.9 mm, suggesting that MRI‐measured DOI tends to be overestimated, particularly in T1–T2 tumors (Nilsson et al. 2022). Noorlag et al. (2020) observed that US overestimated DOI less than MRI for tumors with invasion less than 10 mm, though the trend reversed for tumors with invasion greater than 10 mm. Takamura et al. (2022) compared the estimation potential of US to CT and MRI and found the mean overestimation was lowest with US (0.2 mm) compared to MRI (1.9–2.5 mm) and CT (2.6 mm). Interestingly, as with TT (Brouwer De Koning et al. 2019), the overall spread of DOI measurements around the mean was closer for US (5.5 mm) compared to MRI (5.8–8.2 mm) and CT (9.5–9.9 mm) (Takamura et al. 2022). Three studies assessed sensitivity and specificity of US in evaluating DOI (Caprioli et al. 2022; Iida et al. 2018; Rocchetti et al. 2021), all reporting sensitivity greater than 90%, while specificity ranged from 70.6% (Iida et al. 2018) to 100% (Caprioli et al. 2022; Rocchetti et al. 2021). Cocker et al. (2021) also assessed the accuracy of DOI measurements made by the US and found that a higher proportion of measurements were made within a smaller tolerance thickness than CT and MRI.
US‐DOI showed a correlation of 90% with histopathological DOI in multiple studies (Izzetti et al. 2021; Kaltoft et al. 2024; Takamura et al. 2022). US tends to overestimate DOI (Izzetti et al. 2021; Noorlag et al. 2020; Takamura et al. 2022); however, in comparison to MRI and CT, the variance of DOI measurements and mean overestimation were smaller for US (Brouwer De Koning et al. 2019; Nilsson et al. 2022; Takamura et al. 2022). The US also showed 90% sensitivity in measuring DOI and had varying degrees of specificity (Caprioli et al. 2022; Iida et al. 2018; Rocchetti et al. 2021).

3.3.3.3
Malignancy Features
US demonstrated moderate to high inter‐clinician agreement in assessing lesion boundaries, echogenicity, internal architecture, and vascularity (Ogura et al. 2013). SN and SP of US in detecting tumor spread beyond the lamina propria in the tongue were greater than 90% in one study (Rocchetti et al. 2021). In another study, the US was as sensitive and specific as MRI in detecting the expansion of tongue tumors across the midline (100% specificity and sensitivity for US and MRI), but was much less sensitive and more specific than MRI in detecting the spread of the tumor outside the tongue region (SN: 33.3% (US) vs. 100% (MRI); SP: 100% (US) vs. 75% (MRI)) (Takashima et al. 1989).

3.4
Risk of Bias Assessment
The most frequent problems observed across studies were in the definition of inclusion criteria (n = 8), description of study subjects and setting (n = 7), the criteria used for measurement of the condition (n = 7), and outcome measurement (n = 7). The details are presented in Appendices S3 and S4.

4
Discussion
This scoping review overviews the current knowledge regarding the potential of US to be used as an adjunct chairside tool in the diagnosis of oral lesions. US screening of benign and/or premalignant soft tissue lesions and intraosseous lesions showed high SN and SP. Additionally, there was a strong correlation of US‐derived measurements of TT and DOI of malignancies with histopathological measurements. Previous systematic reviews showed agreement of the US ability to measure TT (Klein Nulent et al. 2018; Marchi et al. 2020; Tarabichi et al. 2019). A meta‐analysis focused on tongue SCC demonstrated that US TT is highly correlated to HP TT, suggesting that US is valuable for preoperative estimation of tongue SCC TT (Tarabichi et al. 2019). A similar meta‐analysis that evaluated US on oral SCC showed that it has high potential alongside MRI in preoperative assessment of TT as US TT showed good correlation with HP TT (Marchi et al. 2020). The accuracy of US to determine TT was also demonstrated in a meta‐analysis of oral SCC in which a high correlation was found between US TT and HP TT in T1–T2 oral cancers (Klein Nulent et al. 2018). While TT measurement with US has been previously investigated, this scoping review presents additional parameters in which US can potentially improve diagnostic accuracy.
One study assessed the use of UHFUS in the diagnosis of benign soft tissue lesions, demonstrating strong accuracy values for all described categories of lesions (Izzetti et al. 2020). While the diagnostic efficacy of UHFUS has been demonstrated, this mode of US was also shown to be capable of recognizing characteristic biomarkers of each lesion type (growth type, mucosal thickness, echogenicity, and vascularization). The high sensitivity (100%) detected for mucosal growths suggests that all cases of lesions belonging to this category were identified and correctly classified by UHFUS. This study also highlights the potential of UHFUS in preoperative examination and as an adjunct tool in the diagnostic workup for benign soft tissue lesions (Izzetti et al. 2020). Further studies exploring UHFUS are encouraged.
With respect to bony lesions, US was capable of identifying ameloblastoma proliferation based on cortical bone destruction with strong sensitivity and specificity; additionally, the use of CDFI was utilized to characterize tumor proliferation based on blood flow signaling (Lu et al. 2009). This study showcases that the assessment of mandibular odontogenic tumors may be facilitated with the use of US and may even provide a direction on the treatment approach, as it provides key information on tumor proliferation activity (Lu et al. 2009). In agreement, the high SN of US to detect bone lesions has been previously reported in a systematic review where jaw lesions were detected with 100% SN and 94% SP (Musu et al. 2016). This study used a 10–13 MHz probe and an extraoral scan protocol.
Our scoping review suggests that the malignant characteristics of oral lesions could be detected by the US. One study showed a high detection rate of tongue tumor spread by US (Rocchetti et al. 2021), and another study demonstrated that the US is comparable to MRI for detecting expansion of tongue tumors within the tongue itself (Takashima et al. 1989). Our findings allow us to conclude that US tumor detection is dependent on the size and location of the lesion. US has been shown to perform accurately in the detection of tongue tumors (Natori et al. 2008). One study further supports the performance of the US in detecting (base of tongue) BOT lesions (Faraji et al. 2018). This is in agreement with a previous systematic review on tongue SCC in which the US performed similarly with MRI in the estimation of TT of oral SCCs, with the majority of included studies being tongue tumors (Marchi et al. 2020). However, a different study demonstrated that US had significantly less sensitivity in detecting tumors compared to MRI (Takashima et al. 1989). These discrepancies in results indicate that more studies are needed to assess the performance of US on tongue lesions, and a stricter protocol for investigation is warranted. In terms of location, US was more sensitive in detecting whether a tumor had invaded the adjacent mandibular bone (93%) compared to conventional radiography (90%) and CT (86%) (Millesi et al. 1990).
The heterogeneity of US diagnostic metrics values between studies suggests that the type of US transducer, its frequency, and the acquisition protocol can impact the performance of the technology. Interestingly, one study reported differences in results depending on the transducer type (Lodder et al. 2011). In addition, US is heavily operator dependent and it requires a level of training. It is essential that studies focus on developing standardized imaging acquisition and assessment protocols and guidelines for the optimal use of US in dental practice. This will facilitate clinical implementation and validation of US as a chair‐side tool in dentistry.
None of the included studies tested new handheld pocket US systems to assess the lesions. Although these devices have been previously tested for different purposes in medicine, with great diagnostic performance (Popat et al. 2024), there is limited investigation of the device in dental studies. Handheld US has several advantages over traditional cart‐based ultrasound systems, including reduced costs and real‐time imaging at the point of care (Andersen et al. 2019). Future research should explore the diagnostic potential of these systems, as their cost‐effectiveness and compact size make them ideal candidates for use as chairside tools, expanding their potential as a screening tool.
The findings from this scoping review indicate that the dental field could be heading in a similar direction as our medical colleagues, where US can be used as an initial method for assessing lesions and subsequently guide the clinician to make a more informed decision in the diagnostic workup and management of the patient (Rix et al. 2018). By providing valuable information about TT, DOI, malignant characteristics, and detection of tumor invasion, it holds significant potential as a chairside tool in dentistry. It can serve as a valuable complement in clinicians' decision‐making processes on selecting further imaging modalities.
4.1
Limitations, Knowledge Gaps, and Future Directions
This review identified knowledge gaps related to the diagnostic value of specific US systems, such as ultra‐high‐frequency US (UHFUS) and handheld pocket US. Additionally, there is a limited body of research evaluating the diagnostic potential of US for assessing benign lesions, including both intraosseous bony and soft tissue lesions. Further research is encouraged in the following topics: (1) exploring US applications in diverse regions of oral lesions, ranging from bony lesions to non‐cancerous soft tissue lesions; (2) exploring different types of new US systems, including UHFUS and handheld pocket US.

4.2
Clinical Translation of the Study Findings
Unlike other imaging modalities, US provides real‐time visualization without exposing the patient to ionizing radiation. Its portability and accessibility make it particularly advantageous for point of care use (Shriki 2014). In contrast, MRI and CT imaging are typically limited to hospital‐based settings and are often associated with significant wait times. For example, in countries like Norway and Canada, more than 60% of patients experience delays of over a month before obtaining a specialist appointment (OECD 2020).
These advantages position US as a promising tool for initial assessment of lesions in dental practice, as an additional chair‐side tool. However, successful integration into routine care will require further investigation of the learning curve, operator training requirements, and standardization of protocols to ensure consistent diagnosis across dental practitioners.

5
Conclusion
The potential of the US in assessing oral lesions was demonstrated by its high sensitivity and specificity in detecting lesion presence, location, TT, and DOI. As US continues to emerge as a valuable imaging modality, prospective studies are also needed to evaluate its clinical applicability, particularly in how its use in the initial assessment of lesions influences diagnostic accuracy and treatment decisions.

Author Contributions

Camila Pacheco‐Pereira: conceptualization, writing – original draft, methodology, writing – review and editing, supervision, formal analysis. Sheryn Villarey: methodology, writing – review and editing, writing – original draft, formal analysis. Swarna Yerebairapura Math: methodology, writing – review and editing. Konrad Lehmann: writing – original draft, methodology, formal analysis. Carlos Alberto Figueredo: methodology, writing – review and editing. Fabiana T. Almeida: writing – original draft, conceptualization, investigation, methodology, writing – review and editing, supervision, formal analysis.

Conflicts of Interest
The authors declare no conflicts of interest.

Supporting information

Appendix S1: odi70107‐sup‐0001‐AppendixS1.docx.

Appendix S2: odi70107‐sup‐0002‐AppendixS2.docx.

Appendix S3: odi70107‐sup‐0003‐AppendixS3.docx.

Appendix S4: odi70107‐sup‐0004‐AppendixS4.docx.