Concurrent validity of cephalometric analysis using smartphone application (oneCeph), Dolphin Imaging software, and conventional method

INTRODUCTION

Cephalometric analysis remains one of the most indispensable diagnostic tools in orthodontics, providing quantitative assessment of craniofacial morphology, skeletal relationships, dentoalveolar characteristics, and soft tissue profile.^[1] It plays a fundamental role in diagnosis, treatment planning, growth assessment, treatment evaluation, and long-term follow-up of orthodontic patients. Since the introduction of cephalometric radiography by Broadbent, manual tracing has traditionally been regarded as the reference standard for cephalometric analysis because of its established reliability and widespread clinical acceptance.^[2]

Despite its diagnostic value, conventional manual tracing is labor-intensive, time-consuming, and susceptible to both random and systematic errors. Variations in landmark identification, operator experience, radiographic quality, and manual measurement techniques may influence the accuracy and reproducibility of cephalometric measurements.^[3] With advances in digital technology, computer-assisted cephalometric analysis has progressively replaced manual tracing by offering automated calculations, image enhancement, digital storage, and improved workflow efficiency while minimizing calculation errors.^[4]

Among commercially available digital software, Dolphin Imaging has become one of the most extensively validated and widely used cephalometric analysis systems in orthodontic practice. Numerous investigations have demonstrated that Dolphin Imaging provides measurements comparable to conventional manual tracing, making it a reliable tool for clinical diagnosis and research.^[1] However, commercial software requires dedicated hardware, licensing costs, and workstation availability, which may limit accessibility in many clinical settings.

The widespread use of smartphones has created new opportunities for mobile-based orthodontic applications. Smartphone-assisted cephalometric analysis offers several practical advantages, including portability, affordability, ease of use, and immediate access without requiring sophisticated computer systems. OneCeph, introduced in 2016, is among the most commonly used Android applications for cephalometric analysis and has shown promising accuracy in preliminary validation studies. Nevertheless, concerns remain regarding the influence of smaller screen size, touch-based landmark identification, and operator-dependent variability on measurement reliability.^[5,6]

Although previous investigations have evaluated either smartphone applications or desktop software individually against manual tracing, only limited studies have directly compared conventional manual tracing, Dolphin Imaging software, and OneCeph application using identical cephalometric parameters within the same patient population. Furthermore, evidence evaluating the concurrent validity and agreement of these three modalities remains relatively limited, particularly in the Indian population.

Therefore, the present study was undertaken to evaluate the concurrent validity, reliability, and agreement of cephalometric measurements obtained using OneCeph smartphone application and Dolphin Imaging software by comparing them with conventional manual tracing, which served as the reference standard.

MATERIAL AND METHODS

The calculated sample size was 36. Lateral cephalograms of patients between the ages of 18 and 25 years with Class I skeletal base were retrieved [Figure 1]. All the lateral cephalograms were thoroughly evaluated to fulfill the standardized conditions necessary for the study. Standardization was done by obtaining the cephalograms taken with the same machine (PLANMECA PROLINE XC). 15 selected cephalometric parameters were used to make tracings. The study was divided into 3 groups, traced, and verified by a secondary investigator.

Close

Figure 1: Lateral cephalogram of patient with class I skeletal base.

Export to PPT

The cephalometric radiographs were manually traced using 0.3 mm pencil on a sheet of acetate tracing paper [Figures 2 and 3]. Tracing of 15 parameters was performed and cephalometric measurements were calculated using Dolphin software [Figures 4 and 5] and then mobile application (OneCeph) [Figures 6-8] where the software performed all the measurements based on predefined analysis. All landmarks were marked by a single examiner and verified by a secondary investigator to avoid bias.

Close

Figure 2: Acetate paper, 0.3 mm Hb pencil, and scale.

Export to PPT

Close

Figure 3: Cephalometric analysis performed using the manual method.

Export to PPT

Close

Figure 4: Calibration of the lateral cephalogram in Dolphin Imaging software.

Export to PPT

Close

Figure 5: Identification of landmarks and calculated measurements.

Export to PPT

Close

Figure 6: Uploading of lateral cephalogram and various cephalometric analyses in OneCeph.

Export to PPT

Close

Figure 7: Calibration of the lateral cephalogram in OneCeph application.

Export to PPT

Close

Figure 8: Identification of landmarks and cephalometric analysis in OneCeph application.

Export to PPT

Ethical approval for this study was obtained from the Institutional Review Board before the commencement of the study.

RESULTS

Thirty-six lateral cephalograms were analyzed using three cephalometric tracing methods: Conventional manual tracing, OneCeph smartphone application, and Dolphin Imaging software.

Table 1 presents the descriptive statistics of the fifteen cephalometric parameters evaluated using the three methods. The mean values obtained by manual tracing, OneCeph, and Dolphin Imaging software were highly comparable for all skeletal, dental, and soft tissue measurements. The mean sella-nasion-point A (SNA) values were 80.57 ± 4.06°, 80.38 ± 4.19°, and 80.84 ± 4.13° for manual tracing, OneCeph, and Dolphin Imaging, respectively. Similarly, sella-nasion-point B (SNB), A-nasion-point B (ANB), upper incisor-nasion point A plane [UI-NA (°)], lower incisor-nasion point B plane [LI-NB (°)], upper incisor and nasion point A plane [UI-NA (mm)], lower incisor and nasion point B plane [LI- NB (mm)], N⊥ Point A, N⊥ Pogonion, nasolabial angle, Jarabak ratio, frankfort mandibular plane angle (FMA), frankfort mandibular incisor plane angle (FMIA), incisor mandibular plane angle (IMPA), and mandibular plane angle demonstrated minimal variation among the three methods, indicating excellent agreement in cephalometric measurements.

Pearson correlation analysis demonstrated strong positive correlations between manual tracing and the two digital methods for all evaluated parameters [Table 2]. Correlation coefficients ranged from 0.730 to 0.994, indicating excellent agreement between the methods. Dolphin Imaging demonstrated slightly higher correlations than OneCeph for SNA (r = 0.966), SNB (r = 0.951), N ⊥ Point A (r = 0.944), N ⊥ Pogonion (r = 0.985), nasolabial angle (r = 0.992), FMA (r = 0.963), FMIA (r = 0.973), IMPA (r = 0.969), and mandibular plane angle (r = 0.952). Conversely, OneCeph demonstrated marginally higher correlations for ANB (r = 0.842) and Jarabak ratio (r = 0.994). Comparable correlation coefficients were observed for the remaining dental parameters, indicating excellent consistency between both digital methods and conventional manual tracing.

The intraclass correlation coefficient (ICC) analysis further confirmed the reliability of the digital methods when compared with manual tracing [Table 3]. Both OneCeph and Dolphin Imaging demonstrated excellent agreement for the majority of cephalometric measurements, with ICC values exceeding 0.90 for most parameters. Dolphin Imaging exhibited slightly higher ICC values than OneCeph for SNA (0.96 vs. 0.94), SNB (0.95 vs. 0.94), and UI-NA (°) (0.98 vs. 0.97), whereas OneCeph demonstrated slightly higher agreement for ANB (0.84 vs. 0.73). Overall, both digital methods exhibited excellent reproducibility and reliability, with Dolphin Imaging demonstrating marginally superior agreement with conventional manual tracing for most cephalometric parameters.

DISCUSSION

The transition from conventional manual tracing to digital cephalometric analysis has significantly influenced contemporary orthodontic diagnosis by improving efficiency, reproducibility, and data management. While manual tracing remains the reference standard for cephalometric evaluation, computer-assisted and smartphone-based applications have gained widespread acceptance owing to their ability to reduce analysis time and eliminate manual calculation errors. The present study evaluated the concurrent validity of the OneCeph smartphone application and Dolphin Imaging software by comparing their cephalometric measurements with conventional manual tracing.^[5]

The results demonstrated no statistically significant differences among manual tracing, Dolphin Imaging software, and OneCeph for any of the fifteen cephalometric parameters evaluated. These findings indicate that both digital platforms produce measurements comparable to those obtained using conventional manual tracing and can therefore be considered clinically reliable alternatives.

The absence of significant differences observed in the present study is consistent with the findings of Shrestha and Kandel,^[1] Mohan et al.,^[5] and Zamrik and İşeri,^[6] who reported excellent agreement between OneCeph and manual tracing for most angular and linear cephalometric measurements. These authors concluded that smartphone-assisted cephalometric analysis provides measurements that are clinically comparable to conventional methods while substantially reducing the time required for analysis. The present findings further strengthen this evidence by simultaneously comparing OneCeph with both manual tracing and well-established commercial Dolphin software. The present findings are further supported by a recent study by Chugh et al.,^[7] who compared the OneCeph smartphone application with Dolphin Imaging software and reported good to excellent interobserver and intraobserver reliability for both systems. The authors observed no statistically significant differences between the two methods for hard- and soft-tissue cephalometric measurements, confirming that smartphone-assisted cephalometric analysis provides measurements comparable to established computer-based software. Although Dolphin Imaging required less time for analysis, the diagnostic reliability of OneCeph remained clinically acceptable, supporting its use in routine orthodontic practice.

Similarly, previous investigations evaluating Dolphin Imaging software have consistently demonstrated excellent agreement with conventional cephalometric analysis. Gregston et al.^[8] reported that computerized cephalometric software generated measurements comparable with manual tracing, particularly for skeletal parameters, while Celik et al.^[9] concluded that the minor differences observed between digital and conventional methods were statistically and clinically insignificant. The findings of the present study corroborate these observations, as all evaluated skeletal, dental, and soft tissue parameters showed comparable mean values irrespective of the tracing method used.

The strong correlations observed between manual tracing and both digital methods indicate that the accuracy of cephalometric analysis depends predominantly on precise landmark identification rather than on the software platform itself. Once anatomical landmarks are correctly identified, both OneCeph and Dolphin automatically calculate linear and angular measurements, thereby minimizing human error associated with manual measurement using rulers and protractors. This explains the high Pearson correlation coefficients obtained across almost all variables.

Among the evaluated digital methods, Dolphin Imaging software demonstrated marginally higher correlation coefficients and intraclass correlation values for most cephalometric parameters compared with OneCeph. Although these differences were small and clinically insignificant, they may be attributed to several technical advantages of desktop-based software. Dolphin Imaging provides a larger display, superior image resolution, enhanced contrast adjustment, multiple magnification options, and more precise cursor control using a computer mouse, all of which facilitate accurate landmark localization. In contrast, landmark placement on a smartphone depends on touch-screen interaction and a comparatively smaller display, factors that may introduce minimal operator-dependent variability despite image magnification capabilities.

Interestingly, OneCeph demonstrated comparable or slightly better agreement for selected parameters such as ANB angle and Jarabak’s ratio. Since these measurements are derived from multiple skeletal landmarks and represent calculated angular relationships rather than isolated linear measurements, minor variations in landmark placement may exert less influence on the final measurement. Consequently, the observed differences between the two digital methods are unlikely to have any meaningful impact on orthodontic diagnosis or treatment planning.

Another important observation in the present study was the consistently high intraclass correlation coefficients for both digital methods when compared with manual tracing. Intraclass correlation coefficients exceeding 0.90 for most parameters indicate excellent agreement and confirm the reproducibility of both software platforms. These findings are in accordance with those reported by Roden-Johnson et al.,^[10] who demonstrated superior reproducibility and reliable landmark identification using computerized cephalometric systems. Likewise, Erkan et al.^[11] concluded that differences among various computerized cephalometric software programs were primarily related to landmark identification rather than computational inaccuracies.

Landmark identification continues to represent the greatest source of variability in cephalometric analysis irrespective of the tracing modality employed. Sayinsu et al.^[12] demonstrated that anatomical landmarks such as Point A, Point B, Gonion, and Porion are inherently more difficult to identify because of anatomical superimposition and image quality, leading to greater measurement variability. Similar challenges exist for both manual and digital tracing. Therefore, adequate operator training and experience remain essential for obtaining reproducible cephalometric measurements regardless of the software platform.

From a clinical perspective, the findings of the present study have important implications. The increasing incorporation of digital workflows into orthodontic practice has created a need for reliable, portable, and economical cephalometric analysis systems. Although Dolphin Imaging software demonstrated marginally superior agreement with manual tracing, the differences were not clinically significant. Consequently, OneCeph may be considered a practical alternative in situations where access to commercial software is limited. Its portability, ease of use, offline functionality, and low cost make it particularly useful in outreach programs, educational institutions, private practices, and teleorthodontic consultations.

The rapid evolution of digital orthodontics has expanded beyond conventional computer-assisted tracing toward artificial intelligence (AI)-assisted landmark detection. Recent investigations have demonstrated that AI-assisted cephalometric systems substantially reduce tracing time while maintaining high diagnostic accuracy when clinician verification of landmark placement is performed. However, fully automated landmark detection still requires clinician supervision, particularly for soft-tissue landmarks and anatomically complex regions. These findings indicate that contemporary orthodontic practice is progressively moving toward a hybrid workflow in which artificial intelligence assists clinicians rather than replacing expert judgment.^[13]

CONCLUSION

Within the limitations of the present study, it can be concluded that the manual and digital methods of identifying cephalometric landmarks are comparable and provide similar results. The zoom-in feature available on smartphones enables magnification of specific areas, thereby facilitating detailed examination and accurate landmark identification by the examiner. The computerized cephalometric analysis systems employed in this investigation demonstrated accurate results, indicating that computer-aided cephalometric analysis can be effectively applied in clinical practice. Mobile application-assisted cephalometry showed good agreement with both PC-aided cephalometry and manual tracing and may be preferred when user-friendliness and portability are prioritized. OneCeph was found to be an easy, reliable, and accurate alternative to conventional manual tracing. It can be accessed on a smartphone without an internet connection, thereby saving clinical time and resources. Smartphone-based cephalometric analysis applications demonstrated satisfactory validity and reliability. Furthermore, OneCeph showed close agreement with the manual method, which was considered the gold standard, whereas Dolphin software demonstrated the highest dependability among the evaluated digital cephalometric methods.