Aksaray Malaklı dogs are native to Central Anatolia, particularly in the regions of Aksaray, Nevşehir and Şereflikoçhisar, Turkey (Atasoy et al., 2011). Owing to their characteristic drooping lips and cheeks, these dogs are commonly referred to as “Malakli,” meaning “with large cheeks.” Distinct genetic features of the breed include black spots on the head, grayish coat coloration, a large skull and a prominent drooping mouth and lip structure (Atasoy and Kanlı, 2005).
       
In recent years, three-dimensional (3D) surface models have become highly valuable in geometric morphometric studies across anatomy (Arya et al., 2024), osteoarchaeology and forensic sciences (Waltenberger et al., 2021; Özkadif et al., 2019; Fidancı and Orhan, 2025). 3D reconstructions permit precise morphometric measurements and qualitative morphological assessments without damaging original specimens and they also facilitate digital storage and sharing (Gelati and Tanga, 2015). Surface scanners use either laser or visible light to analyze object surfaces. While optical scanners offer sufficient color detail and resolution for distance measurements and landmark placement, laser scanners provide greater accuracy and perform better in daylight conditions, albeit without capturing tissue color information (Friess, 2012). Modern scanners tend to be lightweight, portable and suitable for handheld use (Adams et al., 2015).
       
The cervical vertebrae form the bony framework of the neck, supporting its weight and allowing for head movement (König and Bragulla, 2007). All domestic mammals possess seven cervical vertebrae (Akers and Denbow, 2013). The first, second, sixth and seventh cervical vertebrae are morphologically distinct from the others and are classified as “atypical cervical vertebrae” (Waxenbaum et al., 2025).

Except for the atlas, the vertebral bodies are broad, narrowing towards the last cervical vertebra. The spinous process-a prominent dorsal projection found in most vertebrae-is relatively small and thin in the cervical region (Dyce et al., 2009). Due to their morphometric diversity and clinical importance, cervical vertebrae have been the subject of numerous studies (Zibis et al., 2016; Prameela et al., 2020), including those focused on surgical applications (da Costa et al., 2008; Falzone et al., 2022), breed-specific comparative morphometry (Fourie and Kirberger, 1999) and evolutionary anatomy (Brocal et al., 2018). The cranium-C1 (atlanto-occipital) joint provides approximately half of the total cervical flexion/extension and the C1-C2 (atlanto-axial) joint provides approximately half of the rotation. The ala atlantis of the atlas transmits loads from the skull to the cervical spine, while the dens of the axis forms a pivot within the atlas ring, enabling head rotation (Kalantar, 2013). Morphological studies of canine cervical vertebrae are essential for clarifying breed differences and providing guidance for veterinary surgical procedures.
       
Investigating the morphometric characteristics of Aksaray Malaklı dogs is particularly important, as they serve as guardian dogs, are an integral part of regional livestock farming and represent a rare breed unique to a specific area. Although several studies have addressed various features of this breed (Fidan et al., 2024; Ilgün et al., 2022; Ilgün et al., 2017; Inanç et al., 2018), to date no comprehensive morphometric analysis of the atlas (C1) and axis (C2)-the atypical cervical vertebrae-has been conducted.
       
The primary aim of the present study was to evaluate the accuracy and reliability of morphometric and angular measurements obtained from 3D surface scanner models (Fig 1) of the atlas (C1) and axis (C2) vertebrae in Aksaray Malaklı dogs (Fig 2, Fig 3). Additionally, these digital reconstructions were compared with published data from other dog breeds to identify breed-specific anatomical differences.

Collection of bones
 
This study used seven atlas and seven axis vertebrae (four males, three females) obtained from adult Aksaray Malaklı dogs, aged 5-8 years and weighing 60-75 kg. The specimens were collected from clinically healthy animals that died of natural causes between 2020 and 2025. Standard anatomical dissection procedures were followed: skin and subcutaneous tissues were first removed to expose the vertebral region. The surrounding muscles, ligaments and connective tissues were carefully dissected from the bones using scalpels and scissors. Intervertebral joints-capsules, ligaments and discs-were incised to isolate the atlas (C1) and axis (C2) from adjacent vertebrae. Any residual soft tissue was removed either mechanically or by soaking the bones in a moderate hydrogen peroxide solution to ensure clean surfaces for morphometric measurements. The specimens were catalogued accordingly. All procedures adhered to the guidelines of the Animal Ethics Committee at Aksaray University (approval no: 2025/23).
 
Surface scanning procedure
 
High-resolution, color surface models of each bone were obtained using a Revopoint POP 3 Plus Standard Edition scanner (Shenzhen, China) in color mode with default settings. Scans were conducted in a dark room to minimize ambient light and optimize image quality. The scanner was calibrated prior to each session according to the manufacturer’s instructions. Images were processed with Revo Scan v5.5.2 software and exported in OBJ format. Each atlas and axis model was subsequently analyzed using Meshmixer (v3.5). The digital models were cross-checked against the original bones and standard anatomical references (Evans and De Lahunta, 2012; Barone, 1999) and manual corrections were made whenever necessary to ensure validation.
 
Morphometric analysis
 
All OBJ files were imported into 3D Slicer (v5.6.2) (Fedorov et al., 2012) , where 25 linear and one angular measurement were performed on each bone. Detailed osteometric landmarks are provided in Table 1 and 2 (Von Den Driesch, 1976). The dens angle (DA^o) of the axis was measured as described by Takahashi et al., (2017). For this, 3D models were sagittally bisected in Meshmixer and the angle between the floor of the vertebral canal and the dens was measured in ImageJ (v1.54). All measurements were completed by a single researcher. To assess intra-observer repeatability and reduce measurement bias, each measurement was repeated three times at different intervals in a blinded fashion and the average value was used for analysis. Thus, reliability and objectivity were ensured. All anatomical terms conform to the Veterinaria (2017).





Statistical analysis
 
Statistical analyses were conducted using SPSS version 29.0. Descriptive statistics (mean, standard deviation, minimum and maximum) were calculated for all parameters. Relationships between continuous variables were assessed using the Pearson correlation coefficient; when parametric assumptions were not met, Spearman’s correlation was applied. Statistical significance was set at p<0.05 and p<0.01. Intra-observer reliability was assessed using the Intraclass Correlation Coefficient (ICC) with 95% confidence intervals (McGraw and Wong, 1996; Longo et al., 2018). Sex based differences (sexual dimorphism) were not evaluated due to the limited availability of specimens from this rarely bred, endemic dog breed, which precluded balanced male–female sampling.

The morphometric features of the atlas and axis in Aksaray Malakli dogs were evaluated by performing three repeated measurements on each 3D model. Intraclass correlation coefficients (ICCs) were all statistically significant and exceeded 0.75, indicating high measurement reliability (Table 3). For the atlas, ICC values ranged from 0.921 to 0.999, while for the axis, they ranged from 0.769 to 0.998. All parameters demonstrated ICC values above 0.75, a threshold considered satisfactory. The lowest ICC was observed for the dens-to-axis length ratio (DALR) of the axis (0.769) (Table 3).


       
The mean height of the axis (HAx) was 53.21±5.54 mm, which was higher than the mean height of the atlas (HAt) at 36.67±3.50 mm. Several atlas parameters-including the breadth of the caudal articular surface (BFcdAt), height of the cranial vertebral foramen (FVHAtcr), height of the caudal vertebral foramen (FVHAtcd), width of the cranial vertebral foramen (FVWAtcr) and width of the caudal vertebral foramen (FVWAtcd)-were greater than their axis counterparts (BFcdAx, FVHAxcr, FVHAxcd, FVWAxcr, FVWAxcd). Additionally, the mean surface area and total volume were higher for the axis (surface area: 13,101.33± 2,633.01 mm²; volume: 29,331.98±8,153.71 mm³) compared to the atlas (surface area: 12,047.94±1,842.67 mm²; volume: 25,037.09±6,636.64 mm³) (Table 3).
       
Correlation analyses among the 3D-measured parameters for the atlas and axis are presented in Table 3 and 4. The greatest breadth of the atlas wings (GB), its surface area (SAAt) and its volume (VAt) all showed strong positive correlations (e.g., GB and SAAt: r = 0.976, p<0.01; GB and VAt: r = 0.921, p<0.01). GB was also highly correlated with atlas height (HAt, r = 0.953, p<0.01) and the greatest breadth of the cranial articular surface (BFcr, r = 0.900, p<0.01). Significantly, the strong positive association between GB and SAAt indicates that as the atlas wings (ala atlantis) widen, the atlas surface area-and concomitantly its volume-increases proportionally, suggesting that lateral expansion of the wings is a primary contributor to overall atlas size.


       
The greatest length of the atlas (GL) and the distance from the cranial to caudal articular surfaces (GLF) both showed strong positive associations with surface area (SAAt) and volume (VAt), particularly with surface area (GLF and SAAt: r = 0.956, p<0.01). The BFcr value was also highly correlated with both surface area (SAAt: r = 0.927, p<0.01) and volume (VAt: r = 0.877, p<0.01) (Table 4).
       
Pearson correlation was used to assess the relationships and the analysis was verified for consistency with our predefined criteria. Breadth of the caudal articular surface of the atlas (BFcdAt) was strongly correlated with the height of both the cranial (FVHAtcr, r = 0.907, p<0.01) and caudal (FVHAtcd, r = 0.911, p<0.01) vertebral foramina (Table 4). The length of the dorsal arch (LAd) also showed positive correlations with surface area and volume, although correlations with other parameters were generally weaker. Overall, surface area (SAAt) and volume (VAt) demonstrated strong associations with multiple morphometric features of the atlas (Table 4).
       
For the axis, Pearson correlation analysis indicated significant positive relationships among several parameters (Table 5). The length of the dorsal canal axis (LDCA) was strongly correlated with both the surface area (SAAx, r = 0.904, p<0.01) and volume (VAx, r = 0.864, p<0.05) of the axis, as well as with its height (HAx, r = 0.900, p<0.01).



The greatest breadth across the transverse processes (BPtr) also showed strong correlations with surface area (r = 0.973, p<0.01) and volume (r = 0.970, p<0.01). Similarly, axis height (HAx) was positively correlated with both SAAx and VAx. The length of the dens (LD) was significantly correlated with both surface area and volume. The width of the cranial vertebral foramen of the axis (FVWAxcr) was significantly associated with BPtr (r = 0.762, p<0.05) and SAAx (r = 0.728). As in the atlas, both surface area (SAAx) and volume (VAx) of the axis demonstrated strong correlations with a variety of morphometric features (Table 5).
       
Notably, DALR showed weak-to-moderate and generally non-significant correlations with the other morphometric variables, indicating that it does not systematically increase or decrease in tandem with morphometric lengths and functions largely as a size-independent ratio.
       
Previous research on the Aksaray Malakli breed has mainly investigated general anatomical features or linear measurements of long bones (Özüdoğru et al., 2023; Fidan, 2023). However, there is a notable lack of breed-specific morphometric data on the atlas and axis vertebrae, which are crucial for understanding cervical biomechanics in large dog breeds. In the present study, we utilized 3D modeling to achieve highly accurate, reproducible morphometric measurements of these vertebrae, thus establishing a methodological foundation for both clinical and research applications. Comparative data for the atlas and axis in various species are summarized in Tables 6 and 7.
       
Our results demonstrate that the atlas and axis of Aksaray Malaklı dogs are morphologically distinct from those of smaller domestic breeds and other mammals. 3D modeling significantly increased the reliability and reproducibility of our measurements, consistent with the growing evidence supporting 3D imaging in veterinary anatomy (Smith and Jones, 2016; Hildebrand, 2019). Such precision is especially important for generating breed-specific reference data that can inform future anatomical, surgical and diagnostic work.
       
Comparative analysis shows the greatest breadth (GB) of the atlas in Malaklı dogs (108.92±9.33 mm) is, as expected, less than that found in animals such as cattle and horses, but much greater than in smaller species such as Persian cats (Derakhshi et al., 2024) and baboons (Tominaga et al., 1995). Including a range of other species in the comparison highlights how vertebral dimensions adapt to varying biomechanical demands across mammals (Table 6).


       
Within canines, our data indicate that Malaklı dogs, like other large breeds, exhibit clear differences from small and toy breeds-particularly in atlas width and other clinically relevant morphometric ratios (e.g., DALR). For example, the mean GB and HAt values for Malaklı dogs are not only higher than those in Persian and Van cats, but also consistent with the expectation that larger breeds have more robust cervical vertebrae. These differences may influence vulnerability to cervical disorders, tolerance to injury and surgical considerations.
       
Regarding the axis, the sagittal diameter (e.g., FVHAxcr, FVHAxcd) in Aksaray Malaklı dogs was greater than the transverse diameter-a trend seen in sheep and foxes, but not in all species (such as Van cats, baboons, or humans). In large breeds, a relatively larger sagittal diameter may reinforce the vertebra against greater forces associated with head movement, which is especially relevant for working on guardian breeds like the Malaklı (Table 7).


       
Dens hypoplasia-a key factor in atlantoaxial instability (AAI), especially in small breeds-can be evaluated using the dens-to-axis length ratio (DALR). Prior studies report DALR values of 0.41±0.12 in dogs overall, 0.36±0.009 in AAI-affected dogs and 0.40±0.007 in healthy controls (Takahashi et al., 2017); other studies have reported similar or lower values in affected dogs (Cummings et al., 2018; Kim et al., 2022; Planchamp et al., 2022). The DALR observed in Malakli dogs was 0.29±0.02-lower than typical reference ranges and similar to values found in AAI-affected dogs-suggesting a potential predisposition to AAI, though reference values specific to large breeds remain to be fully established.
       
The dens angle (DA^o) is another key measurement for AAI diagnosis (Takahashi et al., 2017). DA^o for toy breeds with AAI averages 41.0^o±1.0^o, significantly higher than in healthy toy breeds (37.6^o±0.7^o) and healthy Beagles (34.8^o±0.5^o). In Malakli dogs, the DA was 31.89^o±5.76^o, lower than reported for both healthy Beagles and toy breeds, suggesting a lower risk of AAI in this breed. However, more comparative morphometric data are needed to clarify the clinical significance of these values for large breeds.
       
Recent work by Lievens et al., (2025) compared MSCT, CBCT and optical white light desktop scanning (OWLDS) for cranial modeling, concluding that OWLDS produces highly accurate, repeatable 3D models without the need for segmentation (mean deviation 0.04/ mm versus MSCT). These findings support the value of optical scanning for reliable morphometric analysis, due to its operator-independence, lack of radiation and efficiency.
 
LIMITATIONS
 
This study has several limitations. The relatively small sample size and unequal sex distribution (four males and three females) limit the capacity for comprehensive statistical comparisons and may reduce the generalizability of the findings. Because this breed is endemic and rarely bred, obtaining specimens was challenging, which further constrained sample size. Additionally, while correlation analysis is useful, many variables tend to move in the same direction, potentially masking underlying structures or key trends. Therefore, the use of multivariate approaches such as Principal Component Analysis (PCA) or regression models in future studies with larger, more balanced cohorts may be valuable for revealing hidden clusters or fundamental patterns and for confirming and expanding upon these results.

3D surface scanning is a reliable and repeatable technique for morphometric analysis of canine vertebrae when standardized protocols are followed. This approach is well suited for anatomical studies. The morphometric baseline data provided here for Aksaray Malaklý dogs enrich the current literature and lay the groundwork for future comparative research involving large dog breeds.

The present study was not supported by any funding sources.
 
Disclaimers
 
The views and conclusions expressed in this article are solely those of the authors and do not necessarily represent the views of their affiliated institutions. The authors are responsible for the accuracy and completeness of the information provided, but do not accept any liability for any direct or indirect losses resulting from the use of this content.
 
Informed consent
 
All procedures were approved by the Aksaray University Ethics Committee.

The authors declare that there are no conflicts of interest regarding the publication of this article. No funding or sponsorship influenced the design of the study, data collection, analysis, decision to publish, or preparation of the manuscript.