Skip to main content
Download PDF

Machine learning in 3D auto-filling alveolar cleft of CT images to assess the influence of alveolar bone grafting on the development of maxilla

BMC Oral Health volume 23, Article number: 16 (2023) Cite this article

Abstract

Background

Machine learning based auto-segmentation of 3D images has been developed rapidly in recent years. However, the application of this new method in the research of patients with unilateral cleft lip and palate (UCLP) is very limited. In this study, a machine learning algorithm utilizing 3D U-net was used to automatically segment the maxilla, fill the cleft and evaluate the alveolar bone graft in UCLP patients. Cleft related factors and the surgery impact on the development of maxilla were analyzed.

Methods

Preoperative and postoperative computed tomography images of 32 patients (64 images) were obtained. The deep-learning-based protocol was used to segment the maxilla and defect, followed by manual refinement. Paired t-tests and Mann-Whitney tests were performed to reveal the changes of the maxilla after surgery. Two-factor, two-level analysis for repeated measurement was used to examine the different trends of growth on the cleft and non-cleft sides of the maxilla. Pearson and Spearman correlations were used to explore the relationship between the defect and the changes of the maxillary cleft side.

Results

One-year after the alveolar bone grafting surgery, different growth amount was found on the cleft and non-cleft sides of maxilla. The maxillary length (from 34.64 ± 2.48 to 35.67 ± 2.45 mm) and the alveolar length (from 36.58 ± 3.21 to 37.63 ± 2.94 mm) increased significantly only on the cleft side while the maxillary anterior width (from 11.61 ± 1.61 to 12.01 ± 1.41 mm) and posterior width (from 29.63 ± 2.25 to 30.74 ± 2.63 mm) increased significantly only on the non-cleft side after surgery. Morphology of the cleft was found to be related to the pre-surgical maxillary dimension on the cleft side, while its correlation with the change of the maxilla after surgery was low or not statistically significant.

Conclusion

The auto-segmentation of the maxilla and the cleft could be performed very efficiently and accurately with the machine learning method. Asymmetric growth was found on the cleft and non-cleft sides of the maxilla after alveolar bone graft in UCLP patients. The morphology of the cleft mainly contributed to the pre-operation variance of the maxilla but had little impact on the maxilla growth after surgery.

Peer Review reports

Introduction

As a congenital maxillofacial hypoplasia, the prevalence of unilateral cleft lip and palate (UCLP) is 6.64 per 10,000 births worldwide and even higher in some developed area, with the deformity of midfacial areas, such as the incomplete formation of lip, alveolar, palate, accompanied by problems of appearance, pronunciation and feeding [1,2,3,4].

Alveolar bone grafting (ABG) was of great importance in the cleft lip and palate team approach. It could not only fill the cleft area but also support the alar base, close oroantral fistulae, and help the eruption of canines within the cleft area [5]. ABG included primary alveolar bone grafting (PABG) and secondary alveolar bone grafting (SABG). PABG was rarely used because of its adverse effects on the development of the maxilla [6]. For SABG, the latest cephalometric studies showed that surgeries performed after 2 years of age (and after palatoplasty) had limited negative effect on subsequent maxillofacial development regardless of its timing [7, 8]. However, head films were 2D images, with disadvantages like no volumetric information and structural overlap, and might bring limitation to the research results. More importantly, lateral cephalometric images are not possible to assess the effect of surgery on the cleft side and the non-cleft side respectively. Three-dimensional (3D) computed tomography (CT) has obvious advantages in evaluating asymmetric structures [9]. It had been demonstrated that maxillary asymmetry concentrated in the dentoalveolar area near the cleft and nasal chamber [10,11,12]. However, few studies explored the effect of SABG on the maxilla development with CT [13].

Besides, it is important to access the volume of cleft accurately by CT before SABG. Firstly, it can avoid taking excessive or inadequate bone and reduce trauma to the donor site [14]. Secondly, the volume of clefts can guide where to take bone. When the requirement of bone is small, the mandibular angle can be the donor site, so there is no need for an extra-oral incision [15]. However, to obtain the accurate cleft volume, most researchers segmented the maxilla and filled the clefts through threshold division and manual modification, time-consumingly and laboriously [5, 16]. Machine learning is a branch of artificial intelligence that uncovers patterns in data automatically and then applies the detected patterns for future data prediction [17]. Machine learning based auto-segmentation of 3D images has been developed rapidly in recent years. However, the application of this new method in the research of patients with unilateral cleft lip and palate (UCLP) is very limited.

In this study we used deep-learning-based protocol, to segment the maxilla and the cleft automatically in CT of SABG patients. This is continuation of method of our previous article about “Machine learning in 3D auto-filling alveolar cleft of CT images” [18,19,20]. The aims of this study were to answer the following questions (1) to evaluate the impact of SABG on maxillary development three-dimensionally; (2) to analyze the association between the size of the cleft and the preservation of the grafted bone; (3) to analyze the association between the preserved grafted bone and 3D development of maxilla after the operation.

Methods

Subjects

This retrospective study was approved by the Institutional Ethical Committee of Peking University School and Hospital of Stomatology (PKUSSIRB-202280135) and conducted in accordance with the tenets of the Declaration of Helsinki. Informed consent was obtained from all patients before the CT was taken.

Firstly, we performed the retrospective screening in the archived database for all patients with UCLP who had pre-surgery (T1) and one-year post-surgery (T2) CT from 2012 to 2020. The CTs were taken for surgery planning and treatment effect evaluation. The inclusion criteria were as follows:

  1. (1)

    All of the subjects had been diagnosed with non-syndromic UCLP and received primary lip and palate repair.

  2. (2)

    The subjects had never undergone previous alveolar bone grafts, orthodontic treatment, maxillofacial neoplasia, trauma, or orthognathic surgery.

  3. (3)

    The subjects were 8–11 years old.

  4. (4)

    CT before and 1 year after SABG were accessible.

In total, we included 32 patients (64 Spiral CTs), consisting of 17 males and 15 females, with a mean age of 9.59 ± 0.97 years (age range: 8–11 years). The CT images were taken at two different times (T1-T2). The mean difference between T1 and T2 was (11.86 ± 0.81) months.

The CT machine (Optima CT520 series, GE MEDICAL SYSTEMS, American) was used under a standard scanning protocol: 120 kV, 14 mAs, 64 × 16.5 cm field of view, and 1.25 mm slice thickness.

Maxilla and defect segmentation

The 64 CT images were saved as Digital Imaging and Communication in Medicine (DICOM) files. Our team had adopted 3D U-net as the neural network architecture and used the cross-entropy loss as the loss function to train a maxi segmentation model of normal people and UCLP patients. The Dice similarity coefficient (DSC) of our segmentation model of maxi and cleft segment had been reported to reach 88% and 83% [18,19,20].

Description of measurement

For 3D morphometric quantification, we determined landmarks by referring to previous literatures [10] and made some adjustments according to the characteristics of Spiral CT, as defined in Table 1 and illustrated in Figure 1A, B. Landmarks were identified on the surface of the 3D segmented model and verified in the multiple planar reformat mode. Three reference planes (the Frankfort horizontal plane, midsagittal plane, and coronal plane) were established as a coordinate system. The maxilla was separated by the midsagittal plane to calculate the volume of the cleft and non-cleft sides respectively. To avoid the impact of teeth on volume, we erased the teeth before calculating maxilla volume. All parameters are defined in Table 1.

Table 1 3D cephalometric landmarks, reference planes and measurements of the maxilla and defect
Full size table
Fig. 1
figure 1

The main landmarks on the 3D segmented model. A Frontal view (cleft side and non-cleft side; landmarks on the non-cleft side are denoted by superscript); B Lateral view (cleft side);

Full size image

The length, width and height were measured by calculating the distances between the position (voxel coordinates) of landmarks, and the volumes of the maxilla and defect were measured based on the segmentation voxel counting. These measurements were carried out using ITK-SNAP [21] (version 3.8.0; www.itksnap.org).

Data collection and analyses

All measurements were conducted by two trained examiners. The intraclass correlation (ICC), is greater than 0.96, confirming the consistency of the measurements.

The data were presented as mean values and standard deviations.

When the data showed a normal distribution, t-tests were used to compare different sides at the same time, or the same side before and after operation; otherwise, Mann-Whitney tests were used. Two-factor, two-level analysis for repeated measurement (r-ANOVA) was used to examine whether the change of the cleft side was the same as non-cleft sides after the operation. Pearson and Spearman correlations were used to explore the relationship between the defect and the maxillary cleft side. All statistical analyses were performed using SPSS (Version 25.0; IBM Co.). The level of significance was set at P < 0.05.

Results

Measurement and analysis of UCLP maxilla and defect

The parameters of the defect structure were shown in Table 2. The analysis and measurement of the cleft and non-cleft sides of maxilla before and one-year after SABG (T1 and T2) were shown in Table 3. Significant differences of maxilla dimension were noticed between the cleft and the non-cleft sides before the surgery, including length and width of maxilla and alveolar bone, anterior alveolar height and the whole maxilla volume. After the surgery, among the above parameters, only the posterior width and alveolar anterior width of maxilla became insignificantly different between the cleft and non-cleft sides. The anterior maxillary height, posterior maxillary height, maxillary volume, and posterior alveolar height increased significantly on both the cleft and non-cleft sides. The maxillary length and alveolar length increased significantly only on the cleft side. The anterior maxillary width and posterior maxillary width increased significantly only on the non-cleft side.

Table 2 Defect structure parameters and measurements
Full size table
Table 3 Data measurement and analysis of the cleft and/or non-cleft sides of maxilla before operation (T1) and/or after operation (T2)
Full size table

The results of two-factor, two-level analysis for repeated measurement (r-ANVOA) of the maxilla were also shown in Table 3. Since new bone had been grafted in the maxillary cleft, the maxillary volume was not included in the two-factor, two-level analysis. The P values of maxillary length, anterior width, posterior width, and alveolar length were less than 0.05, indicating that there was an interaction between different sides and time effects. There was different growth trend on the cleft and non-cleft sides in these items after surgery (Fig. 2).

Fig. 2
figure 2

A The trend of maxillary length overtime on the cleft and non-cleft sides. The length of the cleft sides increased significantly while that on non-cleft sides changed insignificantly. B The trend of maxillary anterior width overtime on the cleft and non-cleft sides. The anterior width of the non-cleft sides increased significantly while that on cleft sides changed insignificantly. C The trend of maxillary posterior width overtime on the cleft and non-cleft sides. The posterior width of the non-cleft sides increased significantly while that on cleft sides changed insignificantly. D The trend of alveolar length overtime on the cleft and non-cleft sides. The length of the cleft sides increased significantly while that on cleft non-sides changed insignificantly

Full size image

Correlation between maxillary and defect

When the data showed a normal distribution, Pearson correlation was used to explore the relation between parameters; otherwise, Spearman was used. The results were shown in Table 4.

Table 4 Pearson and Spearman correlation analysis of defect and maxilla on the cleft side
Full size table

Discussion

This study applied machine learning method to segment the maxilla and the cleft automatically in CT of UCLP patients before and 1-year after SABG surgery to quantify the defect related maxillary morphological characteristics and post-surgery growth change. We optimized the 3D U-net with a non-rigid registration technique and combined it with manual refinement, making it satisfied with the clinical study [18,19,20]. Based on this technique, auto-segmentation of one sample only took several minutes.

Based on the segmentation results, morphological dimensions of maxilla and cleft were measured. Different growth was noticed between the cleft and non-cleft sides. The maxillary length, maxillary volume, alveolar length, and alveolar anterior height on non-cleft sides were larger than those of cleft sides. The anterior and posterior width of the maxilla on cleft sides were larger than those on non-cleft sides. The above results of our study agreed with the previous studies [10, 12]. For those asymmetries, Agarwal et al. [22] explained it based on the dysplasia of UCLP: the bone clefts made the lateral incisors and/or the canines absent in the defect sides, along with the alveolar missing and the premaxillary portion of the defect side was skewed along with the incisors, which shortened the alveolar vertical height and maxillary and alveolar length; the deformed pyriform margin derived from the attachment of the accessory cartilages and the lateral crus served to pull down the nose further caudally, leading to a tilting of the nasal tripod toward the cleft sides and then the width of maxillary increased. However, the anterior and posterior alveolar widths on cleft sides were larger than those of non-cleft sides, which was different from previous studies. Li [12] reported that the means of anterior and posterior alveolar widths on defect sides were larger than those of non-defect sides, but the difference was not significant. Wang [10] reported those indicators of cleft sides were smaller. The reason why our result on alveolar width was different from previous studies might due to individual differences in the self-healing of alveolars after repairing cleft lips. Berkowitz stated in his treatise [23] that intraoral and extraoral muscles worked together to ensure the normal development of the maxilla during the growth of the embryo. Extraoral muscles lost their continuity in UCLP, resulting in the lateral pull of the cleft sides by the abnormal labio-buccal muscles, and at the same time, the tongue facilitated the abnormal lateral movement. After repairing the cleft lip, the continuity of the perioral muscles was restored, and the cleft sides moved inward to reduce the width of the maxilla. But the alveolar inward movement might vary from person to person [24] (Fig. 3), resulting in disagreements among researchers.

Fig. 3
figure 3

Complete unilateral cleft lip and palate (CUCLP) before (A) and after (B) lip surgery. With the establishment of muscle continuity, the lesser segment moves medially, while the premaxillary portion of the larger segment moves medio-inferiorly, both acting to reduce the cleft width. Any of the following segmental relationships can result. B No contact between segments. The inferior turbinate on the cleft side makes premature contact with the bowed nasal septum. C The premaxillary portion of the larger segment overlaps the smaller segment. D The segments form a butt joint showing good approximation. (Both figures and notifications come from ref [23])

Full size image

Many researchers used lateral cephalograms to explore the development of the maxillary of UCLP, and the effect of SABG on the maxilla. Berkowitz concluded [23] that the length of the maxilla increased rarely from 5 to 18 years old and negated the SABG had a deleterious effect on the growth of midface. For the limitation of lateral cephalograms, we cannot measure the cleft and non-cleft sides separately. Therefore, in this study, CT was used to explore the impact of SABG on maxillary growth and development for both sides respectively.

According to the results of this study, the change of the maxillary length, the anterior and posterior width of maxillary, and the alveolar length on cleft sides were different from those on non-cleft sides one year after surgery. The maxillary and alveolar lengths on the non-cleft sides were significantly longer than those on cleft sides at both pre-operative and post-operative time points. However, compared to pre-operation, these lengths of the non-cleft sides changed insignificantly after surgery which was consistent with the above Berkowitz’s conclusion. But these lengths on cleft sides increased significantly, indicating that the cleft side tend to have a catch-up growth in length after the surgery. SABG had a positive effect on the length growth of both the maxilla and alveolar on the cleft sides.

Anterior and posterior widths of the maxilla on non-cleft sides increased significantly within 1 year after surgery, while these widths of the cleft sides had no significant change after treatment. These differences indicated the overall transversal maxillary growth of the cleft sides were lower than those of the non-cleft sides. The maxillary anterior and posterior widths were larger on the cleft sides. This may related to lateral transposition of the cleft segment. However, the difference of the maxillary posterior width on the cleft and non-cleft sides became insignificant one-year after surgery. This may come from the reason that the degree of deformity of the maxillary posterior width on cleft sides was less than that of maxillary anterior width so that when the posterior maxilla of non-cleft side became wider, the width on both sides tended to be the same. The alveolar anterior and posterior widths on two sides changed insignificantly after surgery. However, an increase trend of mean was found on the non-cleft side and a derease trend of mean on the cleft side. Since the anterior alveolar width became larger along with the eruption of the canine, the insignificant change indicated that SABG might have limited benefit to the transverse growth of the alveolar bone, especially to the anterior alveolar bone.

As for the height of maxilla and alveolar bone, about 1 mm increase was noticed one-year after the surgery, except for the anterior alveolar height. On the cleft side, this height was significantly lower than that of the non-cleft side at both T1 and T2. This result indicated that SABG had limited effect on the vertical growth of the alveolar bone on the cleft side. One year after the surgery, the volume of maxilla on both sides increased due to growth and grafting, whereas the volume of the cleft side was still smaller than that of the non-cleft side, indicating that the SABG could not eliminate the defect completely.

Brudnicki showed in the latest papers that SABG indeed increased the volume of alveolars but had an adverse effect on the maxilla growth, although might be limited and regardless of timing [7, 8, 13]. Combined with the results of this study, we further speculated that on one hand, SABG increased the volume and length of the maxilla, especially on the cleft sides; on the other hand, the transversal growth of the maxilla on the cleft sides was more negatively affected by SABG than the non-cleft sides. Since the traversal deficiency could be resolved by orthodontic maxilla expansion, maybe it is better to perform SABG at earlier age than generally assumed to be optimal for subsequent maxillary growth.

Although it is of challenge to determine factors that contribute to maxillary variability, the cleft on the maxilla has been considered as one factor of maxillary dysplasia. In our study, the length of the cleft was found to have the strongest relationship with the volume of the cleft. However, the correlation between increased bone in the cleft region or percentage of bone fill and the morphology of the cleft (length, width, height, and volume) was not high or even had no statistical significance, which was in accordance with previous papers [9, 25,26,27]. The growth of height of the cleft contributed to the increase of volume of maxilla, anterior height, and posterior height of alveolar ridge. And the anterior width of both maxilla and alveolar ridge was positively associated with the width of the cleft. All of these agreed with the previous study [10]. Besides, we also found that the anterior and posterior heights of maxilla were associated with the morphology of the cleft at a medium level. In terms of the change of maxillary and alveolar parameters after the SABG surgery, almost no change was related to the morphology of the cleft, the bone increased, and the percentage of bone fill.

The limitations of the present study included its retrospective character and a relatively small number of participants. More importantly, there was no control group. Although we compared the cleft sides to the non-cleft sides to explore the changes of cleft sides after surgery, the growth of non-cleft sides could not represent that of UCLPs at the same age without receiving SABG. And the growth of UCLP may differ from the normal children at the same age, so it would be better to have a control group of normal children at the same age, which was hard to achieve due to ethical reasons. In addition, the one-year follow up for the assessment was short and CBCT could be better than CT for bone evaluation.

Conclusions

Cleft and non-cleft sides of the maxilla have different growth trends after alveolar bone graft in UCLP patients. The growth amount of the maxillary width on the cleft side is smaller than that on the non-cleft side. Bone graft surgery is beneficial to the length growth of the maxilla and alveolar ridge but has limited benefit to the width and height of the maxilla and the alveolar width on the cleft side. The morphology of the cleft contributes to the variability of the maxilla, while has almost no effect on the percentage of the preserved grafted bone and the growth change of the maxilla after the surgery.

Availability of data and materials

The datasets generated and analysed during the current study are not publicly available due to compromised individual privacy but are available from the corresponding author on reasonable request.

References

  1. Group IW. Prevalence at birth of cleft lip with or without cleft palate: data from the International perinatal database of typical oral clefts (IPDTOC). Cleft Palate Craniofac J. 2011;48(1):66–81.

    Article  Google Scholar 

  2. Zhu Y, Miao H, Zeng Q, Li B, Wang D, Yu X, Wu H, Chen Y, Guo P, Liu F. Prevalence of cleft lip and/or cleft palate in Guangdong province, China, 2015–2018: a spatio-temporal descriptive analysis. BMJ Open. 2021;11(8):e046430.

    Article  Google Scholar 

  3. Li L, Yu HT, Wang XD, Zhou F, Wang F, Wang CF. Analysis of birth defect rate trend of cleft lip and palate in Shanghai from 2007 to 2016. Zhonghua Kou Qiang Yi Xue Za Zhi. 2018;53(5):301–6.

    Google Scholar 

  4. McCrary H, Skirko JR. Bone grafting of alveolar clefts. Oral Maxillofac Surg Clin North Am. 2021;33(2):231–8.

    Article  Google Scholar 

  5. Shaheen E, Danneels M, Doucet K, Dormaar T, Verdonck A, Cadenas de Llano-Perula M, Willems G, Politis C, Jacobs R. Validation of a 3D methodology for the evaluation and follow-up of secondary alveolar bone grafting in unilateral cleft lip and palate patients. Orthod Craniofac Res. 2022;25(3):377–83.

    Article  Google Scholar 

  6. Santiago PE, Schuster LA, Levy-Bercowski D. Management of the alveolar cleft. Clin Plast Surg. 2014;41(2):219–32.

    Article  Google Scholar 

  7. Brudnicki A, Sawicka E, Brudnicka R, Fudalej P. Effects of different timing of alveolar bone graft on craniofacial morphology in unilateral cleft lip and palate. Cleft Palate-craniofac J Off Publ Am Cleft Palate-Craniofac Assoc. 2020;57(1):105–13.

    Article  Google Scholar 

  8. Brudnicki A, Sawicka E, Fudalej P. Maxillofacial morphology in post-pubertal patients with unilateral cleft lip and palate following early vs. late secondary alveolar bone grafting. J Cranio-maxillo-fac Surg Off Publ Eur Assoc Cranio-Maxillo-Fac Surg. 2021;49(9):809–14.

    Article  Google Scholar 

  9. Oberoi S, Chigurupati R, Gill P, Hoffman WY, Vargervik K. Volumetric assessment of secondary alveolar bone grafting using cone beam computed tomography. Cleft Palate Craniofac J. 2009;46(5):503–11.

    Article  Google Scholar 

  10. Wang X, Pastewait M, Wu TH, Lian C, Tejera B, Lee YT, Lin FC, Wang L, Shen D, Li S, et al. 3D morphometric quantification of maxillae and defects for patients with unilateral cleft palate via deep learning-based CBCT image auto-segmentation. Orthod Craniofac Res. 2021;24(Suppl 2):108–16.

    Article  Google Scholar 

  11. Suri S, Utreja A, Khandelwal N, Mago SK. Craniofacial computerized tomography analysis of the midface of patients with repaired complete unilateral cleft lip and palate. Am J Orthod Dentofac Orthop. 2008;134(3):418–29.

    Article  Google Scholar 

  12. Li H, Yang Y, Chen Y, Wu Y, Zhang Y, Wu D, Liang Y. Three-dimensional reconstruction of maxillae using spiral computed tomography and its application in postoperative adult patients with unilateral complete cleft lip and palate. J Oral Maxillofac Surg. 2011;69(12):e549-557.

    Article  Google Scholar 

  13. Brudnicki A, Regulski P, Sawicka E, Fudalej P. Alveolar volume following different timings of secondary bone grafting in patients with unilateral cleft lip and palate. A pilot study. J Clin Med. 2021;10(16):3524.

    Article  Google Scholar 

  14. Kasaven C, Ivekovic S, McIntyre G, Gillgrass T, Thomson D, Menhinick A, Mossey P. Validation of the volumetric measurement of a simulated maxillary alveolar bone defect using cone-beam computed tomography. Cleft Palate-Craniofac J Off Publ Am Cleft Palate-Craniofac Assoc. 2013;50(6):e115-120.

    Article  Google Scholar 

  15. Kang NH. Current methods for the treatment of alveolar cleft. Arch Plast Surg. 2017;44(3):188–93.

    Article  Google Scholar 

  16. Feng B, Jiang M, Xu X, Li J. A new method of volumetric assessment of alveolar bone grafting for cleft patients using cone beam computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol. 2017;124(2):e171–82.

    Article  Google Scholar 

  17. Lee J, Jun S, Cho Y, Lee H, Kim G, Seo J, Kim N. Deep learning in medical imaging: general overview. Korean J Radiol. 2017;18(4):570–84.

    Article  Google Scholar 

  18. Chen S, Wang L, Li G, Wu TH, Diachina S, Tejera B, Kwon JJ, Lin FC, Lee YT, Xu T, et al. Machine learning in orthodontics: Introducing a 3D auto-segmentation and auto-landmark finder of CBCT images to assess maxillary constriction in unilateral impacted canine patients. Angle Orthod. 2020;90(1):77–84.

    Article  Google Scholar 

  19. Zhang Y, Pei Y, Chen S, Guo Y, Ma G, Xu T, Zha H, IEEE. Volumetric registration-based cleft volume estimation of alveolar cleft grafting procedures. In: IEEE 17th international symposium on biomedical imaging (ISBI). Iowa, IA; 2020. p. 99–103.

  20. Zhang Y, Pei Y, Guo Y, Chen S, Xu T, Zha H. Cleft volume estimation and maxilla completion using cascaded deep neural networks. In: Machine learning in medical imaging. 2020. p. 332–341.

  21. Yushkevich PA, Piven J, Hazlett HC, Smith RG, Ho S, Gee JC, Gerig G. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage. 2006;31(3):1116–28.

    Article  Google Scholar 

  22. Agarwal R, Parihar A, Mandhani PA, Chandra R. Three-dimensional computed tomographic analysis of the maxilla in unilateral cleft lip and palate: implications for rhinoplasty. J Craniofac Surg. 2012;23(5):1338–42.

    Article  Google Scholar 

  23. Berkowitz S, Berkowitz S, Berkowitz S. The effect of clefting of the lip and palate and the palatal arch form. In: Berkowitz S, editor. Cleft lip and palate. Berlin Heidelberg: Springer; 2013. p. 61–85.

    Chapter  Google Scholar 

  24. Aduss H, Pruzansky S. The nasal cavity in complete unilateral cleft lip and palate. Arch Otolaryngol. 1967;85(1):53–61.

    Article  Google Scholar 

  25. Ozawa T, Omura S, Fukuyama E, Matsui Y, Torikai K, Fujita K. Factors influencing secondary alveolar bone grafting in cleft lip and palate patients: prospective analysis using CT image analyzer. Cleft Palate Craniofac J. 2007;44(3):286–91.

    Article  Google Scholar 

  26. Linderup BW, Cattaneo PM, Jensen J, Kuseler A. Mandibular symphyseal bone graft for reconstruction of alveolar cleft defects: volumetric assessment with cone beam computed tomography 1-year postsurgery. Cleft Palate Craniofac J. 2016;53(1):64–72.

    Article  Google Scholar 

  27. Honma K, Kobayashi T, Nakajima T, Hayasi T. Computed tomographic evaluation of bone formation after secondary bone grafting of alveolar clefts. J Oral Maxillofac Surg. 1999;57(10):1209–13.

    Article  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

This research was supported in part by the Ningxia Hui Autonomous Region key Research and Development program (2022BEG02031), National Natural Science Foundation of China (51972005, U21A2055, 82101074) and National Key R&D Program of China (2016YFC1000804).

Author information

Authors and Affiliations

  1. Department of Orthodontics, Peking University School and Hospital of Stomatology & National Center of Stomatology & National Clinical Research Center for Oral Diseases & National Engineering Research Center of Oral Biomaterials and Digital Medical Devices, Beijing, People’s Republic of China

    Xin Zhang, Niu Qin & Si Chen

  2. Department of Oral and Maxillofacial Surgery, Peking University School and Hospital of Stomatology & National Center of Stomatology & National Clinical Research Center for Oral Diseases & National Engineering Research Center of Oral Biomaterials and Digital Medical Devices, Beijing, People’s Republic of China

    Zhibo Zhou

Authors
  1. Xin Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Niu Qin
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Zhibo Zhou
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Si Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

XZ designed the work with ZZ and SC. XZ and NQ finished the acquisition, analysis, and interpretation of data. Then XZ drafted the work and revised it under the guidance of ZZ and SC. Each author has approved the submitted version (and any substantially modified version that involves the author's contribution to the study) and has agreed both to be personally accountable for the author's own contributions and to ensure that questions related to the accuracy or integrity of any part of the work, even ones in which the author was not personally involved, are appropriately investigated, resolved, and the resolution documented in the literature. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Zhibo Zhou or Si Chen.

Ethics declarations

Ethics approval and consent to participate

This retrospective study was approved by the Institutional Ethical Committee of Peking University School and Hospital of Stomatology (approval number: PKUSSIRB-202280135) and conducted in accordance with the tenets of the Declaration of Helsinki. Informed consent was obtained from all patients before the CT was taken.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, X., Qin, N., Zhou, Z. et al. Machine learning in 3D auto-filling alveolar cleft of CT images to assess the influence of alveolar bone grafting on the development of maxilla. BMC Oral Health 23, 16 (2023). https://doi.org/10.1186/s12903-023-02706-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12903-023-02706-8

Keywords

BMC Oral Health

ISSN: 1472-6831

Contact us