Progressive Deep Learning for Automated Spheno-Occipital Synchondrosis Maturation Assessment

Ahmet Enis Cetin; Amanda Nikho; Emadeldeen Hamdan; Lauren Mills; Marouane Tliba; Mohammed H. Elnagar; Omid Halimi Milani

arxiv: 2604.10945 · v1 · submitted 2026-04-13 · 💻 cs.CV · cs.LG

Progressive Deep Learning for Automated Spheno-Occipital Synchondrosis Maturation Assessment

Omid Halimi Milani , Amanda Nikho , Marouane Tliba , Lauren Mills , Emadeldeen Hamdan , Ahmet Enis Cetin , Mohammed H. Elnagar This is my paper

Pith reviewed 2026-05-10 16:28 UTC · model grok-4.3

classification 💻 cs.CV cs.LG

keywords spheno-occipital synchondrosismaturation assessmentprogressive learningCBCT imagingdeep neural networkscurriculum trainingfine-grained recognition

0 comments

The pith

Sequentially activating deeper layers in a neural network improves accuracy for assessing spheno-occipital synchondrosis maturation stages from CBCT scans.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper aims to show that a training approach which gradually activates deeper parts of a neural network, starting with basic structures and moving to finer details, can better handle the continuous changes in bone fusion at the spheno-occipital synchondrosis. This matters because current manual assessments vary a lot between experts, especially when the fusion is halfway done, affecting decisions on when to do orthodontic work or surgery. By making the model learn in stages like a clinician does, the method gets more reliable results on tricky middle stages. It does this without altering the network design or the way errors are calculated, just by changing how training happens over time. Experiments on different types of networks confirm better stability and performance.

Core claim

The authors claim that training a network by progressively activating its deeper blocks over time creates a depth-based curriculum that aligns feature learning with the biological progression of synchondrosis fusion. This leads to more stable optimization and higher classification accuracy than training the full network at once, with the largest gains on ambiguous intermediate maturation stages. These improvements hold across convolutional and transformer models and require no changes to the architecture or loss function.

What carries the argument

The progressive representation-learning framework, which sequentially grows the model by activating deeper blocks over training time to encode coarse cranial base morphology first before specializing in subtle fusion patterns.

If this is right

Higher accuracy particularly on transitional fusion stages
More stable training optimization across different network types
Effective without modifications to architecture or loss functions
Applicable to modeling other continuous biological processes in imaging

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

This training strategy might generalize to other medical imaging tasks involving gradual changes, such as bone age assessment or tumor progression.
It highlights that the order of feature learning can substitute for complex model designs in fine-grained anatomical classification.
Future work could test if this method reduces the amount of training data needed for similar problems.

Load-bearing premise

That growing the model depth sequentially will naturally align with the biological stages of fusion and improve separation of close stages by mirroring how experts reason.

What would settle it

Running the same experiments but finding no accuracy gain or even a drop for the progressive method on a standard test set of CBCT images would disprove the main claim.

Figures

Figures reproduced from arXiv: 2604.10945 by Ahmet Enis Cetin, Amanda Nikho, Emadeldeen Hamdan, Lauren Mills, Marouane Tliba, Mohammed H. Elnagar, Omid Halimi Milani.

**Figure 3.** Figure 3: Progressive-growing training. The backbone is expressed as an ordered sequence of blocks. At each stage, a single classification head is attached to the currently active depth and trained. Earlier heads are removed as deeper blocks are activated and training continues. At inference time, the final fully grown network is used with a single forward pass. To enable progressive network training, we partition … view at source ↗

**Figure 4.** Figure 4: Illustration of progressive growing for Transformer encoders: at stage k, only the first k blocks are enabled and trained, while deeper blocks remain inactive until later stages. During stage k, the model consists of the embedding stem, the positional encoding, the class token, all encoder layers in T1 ∪ · · · ∪ Tk, and the final classification head. Thus, deeper encoder blocks are only introduced after ea… view at source ↗

read the original abstract

Accurate assessment of spheno-occipital synchondrosis (SOS) maturation is a key indicator of craniofacial growth and a critical determinant for orthodontic and surgical timing. However, SOS staging from cone-beam CT (CBCT) relies on subtle, continuously evolving morphological cues, leading to high inter-observer variability and poor reproducibility, especially at transitional fusion stages. We frame SOS assessment as a fine-grained visual recognition problem and propose a progressive representation-learning framework that explicitly mirrors how expert clinicians reason about synchondral fusion: from coarse anatomical structure to increasingly subtle patterns of closure. Rather than training a full-capacity network end-to-end, we sequentially grow the model by activating deeper blocks over time, allowing early layers to first encode stable cranial base morphology before higher-level layers specialize in discriminating adjacent maturation stages. This yields a curriculum over network depth that aligns deep feature learning with the biological continuum of SOS fusion. Extensive experiments across convolutional and transformer-based architectures show that this expert-inspired training strategy produces more stable optimization and consistently higher accuracy than standard training, particularly for ambiguous intermediate stages. Importantly, these gains are achieved without changing network architectures or loss functions, demonstrating that training dynamics alone can substantially improve anatomical representation learning. The proposed framework establishes a principled link between expert dental intuition and deep visual representations, enabling robust, data-efficient SOS staging from CBCT and offering a general strategy for modeling other continuous biological processes in medical imaging.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The progressive depth curriculum is a clean idea for tying training to biological stages, but the gains may not survive a matched training budget comparison.

read the letter

The main takeaway is that this paper introduces a training schedule that activates deeper network blocks over time to match the coarse-to-fine progression of SOS fusion. It claims this produces steadier optimization and better accuracy on the ambiguous middle stages than plain end-to-end training, all without touching the architecture or loss. That framing is the actual novelty here. It moves beyond standard curriculum learning by making network depth itself the curriculum variable, which directly echoes how clinicians scan from overall structure to subtle closure details. The abstract is clear on this point and the approach stays empirical rather than trying to derive the schedule from first principles. If the full experiments hold, it offers a lightweight way to inject domain order into representation learning for other continuous biological sequences in imaging. The paper does a decent job keeping the intervention minimal and testing it on both convolutional and transformer backbones, which avoids the usual suspicion that gains come from architecture hacks. The stress-test concern about total training steps is worth taking seriously. If each progressive stage adds epochs or changes the effective learning-rate schedule without explicit matching to the baseline runs, then the reported improvements on intermediate stages could be an artifact of extra gradient updates rather than the curriculum logic. The paper needs to show total steps, wall-clock time, or an ablation with equalized budgets to close that gap. Dataset details, exact metrics, and statistical tests are not in the abstract, so the full text must supply those to make the accuracy claims verifiable. This work is aimed at researchers in dental and craniofacial imaging who deal with staged maturation assessment. A reader focused on curriculum methods or fine-grained medical classification would get the most from it. It deserves a serious referee because the core idea is coherent and the clinical problem is real, even though the experimental controls will likely need tightening.

Referee Report

2 major / 2 minor

Summary. The manuscript presents a progressive deep learning framework for automated assessment of spheno-occipital synchondrosis (SOS) maturation from cone-beam CT (CBCT) images. It proposes sequentially growing the model by activating deeper blocks over time to mirror expert clinician reasoning, from coarse cranial base morphology to subtle fusion patterns. The central claim is that this strategy yields more stable optimization and consistently higher accuracy than standard end-to-end training across convolutional and transformer architectures, particularly for ambiguous intermediate stages, without modifying network architectures or loss functions.

Significance. If substantiated with rigorous, controlled experiments, the significance lies in demonstrating that training dynamics—specifically a curriculum over network depth—can substantially improve performance on fine-grained anatomical classification tasks involving continuous biological processes. This could advance data-efficient learning in medical imaging and establish a principled connection between clinical expertise and deep feature learning, with potential applications beyond SOS staging to other staged or transitional anatomical assessments.

major comments (2)

[Experiments] Experiments section: The comparison between the progressive activation strategy and standard end-to-end training does not specify whether total training epochs, gradient updates, or wall-clock time are matched. If each progressive stage trains for a fixed number of epochs before activating the next block, the cumulative optimization budget exceeds the baseline, providing an alternative explanation for reported gains in accuracy and stability on ambiguous intermediate stages rather than the curriculum itself.
[Results] Results section: The manuscript asserts 'consistently higher accuracy' and 'more stable optimization' but provides no quantitative metrics (e.g., per-stage accuracies, confusion matrices focused on transitional fusion stages, standard deviations over multiple runs, or statistical significance tests such as McNemar's test or paired t-tests against baselines). This leaves the central empirical claim without verifiable support.

minor comments (2)

[Abstract] Abstract: The description of the method and results is repetitive; the final two sentences restate the same benefits without adding new information.
[Method] Method: The progressive activation schedule lacks pseudocode or a precise definition of 'deeper blocks' for both CNN and transformer architectures, hindering reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments, which help improve the clarity and rigor of our manuscript. We address each major comment below and will make the necessary revisions to strengthen the experimental validation and reporting.

read point-by-point responses

Referee: [Experiments] Experiments section: The comparison between the progressive activation strategy and standard end-to-end training does not specify whether total training epochs, gradient updates, or wall-clock time are matched. If each progressive stage trains for a fixed number of epochs before activating the next block, the cumulative optimization budget exceeds the baseline, providing an alternative explanation for reported gains in accuracy and stability on ambiguous intermediate stages rather than the curriculum itself.

Authors: We appreciate this observation and acknowledge that the manuscript does not explicitly detail the matching of training budgets. Upon review, our experimental protocol was designed to match the total number of gradient updates and epochs across methods by proportionally allocating epochs to each progressive stage. However, to eliminate any ambiguity, we will revise the Experiments section to include a clear description of the training schedule, a table comparing total epochs and wall-clock time, and confirmation that the optimization budget is equivalent. This will demonstrate that the performance gains are attributable to the progressive strategy rather than increased training resources. revision: yes
Referee: [Results] Results section: The manuscript asserts 'consistently higher accuracy' and 'more stable optimization' but provides no quantitative metrics (e.g., per-stage accuracies, confusion matrices focused on transitional fusion stages, standard deviations over multiple runs, or statistical significance tests such as McNemar's test or paired t-tests against baselines). This leaves the central empirical claim without verifiable support.

Authors: We agree that providing these quantitative metrics would enhance the verifiability of our claims. While the manuscript reports overall accuracy improvements and qualitative observations on stability, we will expand the Results section to include per-stage accuracy breakdowns (with emphasis on intermediate stages), confusion matrices highlighting errors in transitional fusion stages, standard deviations from multiple independent runs, and results from statistical significance tests such as paired t-tests and McNemar's test. These additions will provide rigorous support for the reported benefits of the progressive framework. revision: yes

Circularity Check

0 steps flagged

No circularity: purely empirical curriculum with experimental comparisons

full rationale

The paper proposes a progressive training strategy for deep networks on SOS staging from CBCT, framing it as mirroring expert clinician reasoning via sequential block activation. No mathematical derivation, equations, or first-principles predictions are present that could reduce to inputs by construction. Claims rest on empirical accuracy gains versus standard end-to-end training across architectures, without self-citations, fitted parameters renamed as predictions, or uniqueness theorems. The curriculum is an explicit design choice, not a result derived from itself; any confounding in training budgets is an experimental validity issue, not circularity. The approach is self-contained as a standard empirical ML contribution.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim depends on the assumption that progressive depth activation will produce better anatomical representations for continuous biological processes; this rests on standard deep learning optimization behaviors and the alignment between network layers and clinical observation order, with no new entities postulated.

free parameters (1)

progressive activation schedule
The timing and criteria for activating deeper blocks during training is a tunable element required to implement the curriculum and is not derived from first principles.

axioms (1)

domain assumption Early layers encode stable cranial base morphology while deeper layers can specialize in subtle fusion patterns when activated sequentially
Invoked directly in the description of how the framework mirrors expert clinician reasoning about synchondral fusion.

pith-pipeline@v0.9.0 · 5580 in / 1376 out tokens · 48282 ms · 2026-05-10T16:28:15.533553+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

26 extracted references · 26 canonical work pages

[1]

American Journal of Orthodontics and Dentofacial Orthopedics122(4), 380–385 (2002)

Mito, T., Sato, K., Mitani, H.: Cervical vertebral bone age in girls. American Journal of Orthodontics and Dentofacial Orthopedics122(4), 380–385 (2002). https://doi.org/10.1067/mod.2002.126896

work page doi:10.1067/mod.2002.126896 2002
[2]

Journal of Children’s Orthopaedics4(5), 467–470 (2010)

Calfee, R.P., Sutter, M., Steffen, J.A., Goldfarb, C.A.: Skeletal and chronological ages in American adolescents: current findings in skeletal maturation. Journal of Children’s Orthopaedics4(5), 467–470 (2010)

work page 2010
[3]

Hormone Research57(Suppl

Karlberg, J.: Secular trends in pubertal development. Hormone Research57(Suppl. 2), 19–30 (2002)

work page 2002
[4]

Indian Journal of Endocrinology and Metabolism18(Suppl 1), S63–S71 (2014)

De Sanctis, V., Di Maio, S., Soliman, A.T., Raiola, G., Elalaily, R., Millimaggi, G.: Hand X-ray in pediatric endocrinology: Skeletal age assessment and beyond. Indian Journal of Endocrinology and Metabolism18(Suppl 1), S63–S71 (2014)

work page 2014
[5]

Australian Dental Journal62, 33–50 (2017)

Scarfe, W.C., Azevedo, B., Toghyani, S., Farman, A.G.: Cone beam computed tomographic imaging in orthodontics. Australian Dental Journal62, 33–50 (2017)

work page 2017
[6]

Journal of Applied Clinical Medical Physics22(2), 85–89 (2021)

Olch, A.J., Alaei, P.: How low can you go? A CBCT dose reduction study. Journal of Applied Clinical Medical Physics22(2), 85–89 (2021)

work page 2021
[7]

The Angle Orthodontist 91(4), 538–543 (2021)

Alhazmi,A.,Aldossary,M.,Palomo,J.M.,Hans,M.,Latimer,B.,Simpson,S.:Cor- relation of spheno-occipital synchondrosis fusion stages with a hand-wrist skeletal maturity index: a cone beam computed tomography study. The Angle Orthodontist 91(4), 538–543 (2021)

work page 2021
[8]

Angle Orthodontist94(6), 641–647 (2024)

Booth, E., Viana, G., Shirazi, S., Miller, S., Sellke, T., Elnagar, M., Viana, M., At- sawasuwan, P.: Correlations of spheno-occipital synchondrosis, cervical vertebrae, midpalatal suture, and third molar maturation stages. Angle Orthodontist94(6), 641–647 (2024). https://doi.org/10.2319/041224-295.1

work page doi:10.2319/041224-295.1 2024
[9]

PLoS ONE11(8), e0161104 (2016)

Fernández-Pérez, M.J., Alarcón, J.A., McNamara Jr., J.A., Velasco-Torres, M., Benavides, E., Galindo-Moreno, P., Catena, A.: Spheno-occipital synchondrosis fusion correlates with cervical vertebrae maturation. PLoS ONE11(8), e0161104 (2016)

work page 2016
[10]

The Anatomical Record147(1), 15–23 (1963)

Powell, T.V., Brodie, A.G.: Closure of the spheno-occipital synchondrosis. The Anatomical Record147(1), 15–23 (1963)

work page 1963
[11]

American Journal of Physical Anthropology157(1), 42–57 (2015)

Lottering, N., MacGregor, D.M., Alston, C.L., Gregory, L.S.: Ontogeny of the spheno-occipital synchondrosis in a modern Queensland, Australian population using computed tomography. American Journal of Physical Anthropology157(1), 42–57 (2015)

work page 2015
[12]

Journal of Forensic Dental Sciences5(2), 72–76 (2013)

Krishan, K., Kanchan, T.: Evaluation of spheno-occipital synchondrosis: A review of literature and considerations from forensic anthropologic point of view. Journal of Forensic Dental Sciences5(2), 72–76 (2013)

work page 2013
[13]

Turkish Journal of Orthodontics36(3), 186 (2023)

Tashayyodi, N., Kajan, Z.D., Ostovarrad, F., Khosravifard, N.: Relationship of the Fusion Stage of Spheno-Occipital Synchondrosis with Midpalatal and Zygomatico- maxillary Sutures on Cone-Beam Computed Tomography Scans of Patients Aged Between 7 and 21 Years. Turkish Journal of Orthodontics36(3), 186 (2023)

work page 2023
[14]

PLoS ONE12(8), e0183305 (2017)

Alhazmi, A., Vargas, E., Palomo, J.M., Hans, M., Latimer, B., Simpson, S.: Timing and rate of spheno-occipital synchondrosis closure and its relationship to puberty. PLoS ONE12(8), e0183305 (2017)

work page 2017
[15]

The Angle Orthodontist80(1), 106–110 (2010)

Leonardi, R., Cutrera, A., Barbato, E.: Rapid maxillary expansion affects the spheno-occipital synchondrosis in youngsters: a study with low-dose computed to- mography. The Angle Orthodontist80(1), 106–110 (2010)

work page 2010
[16]

The Angle Orthodontist91(6), 822–829 (2021)

Leonardi, R., Ronsivalle, V., Lagravere, M.O., Barbato, E., Isola, G., Lo Giudice, A.: Three-dimensional assessment of the spheno-occipital synchondrosis and clivus Title Suppressed Due to Excessive Length 15 aftertooth-borneandbone-bornerapidmaxillaryexpansion:aretrospectiveCBCT study using voxel-based superimposition. The Angle Orthodontist91(6), 822–829 (2021)

work page 2021
[17]

Plastic and Reconstructive Surgery134(3), 504–510 (2014)

Goldstein, J.A., Paliga, J.T., Wink, J.D., Bartlett, S.P., Nah, H.-D., Taylor, J.A.: Earlier evidence of spheno-occipital synchondrosis fusion correlates with severity of midface hypoplasia in patients with syndromic craniosynostosis. Plastic and Reconstructive Surgery134(3), 504–510 (2014)

work page 2014
[18]

Journal of Oral and Maxillofacial Surgery72(6), 1173–1179 (2014)

Tahiri, Y., Paliga, J.T., Vossough, A., Bartlett, S.P., Taylor, J.A.: The spheno- occipital synchondrosis fuses prematurely in patients with Crouzon syndrome and midface hypoplasia compared with age- and gender-matched controls. Journal of Oral and Maxillofacial Surgery72(6), 1173–1179 (2014)

work page 2014
[19]

Dentomaxillofacial Radiology44(2015)

Kapila, S.D., Nervina, J.M.: CBCT in orthodontics: Assessment of treatment out- comes and indications for its use. Dentomaxillofacial Radiology44(2015). Preprint at https://doi.org/10.1259/dmfr.20140282

work page doi:10.1259/dmfr.20140282 2015
[20]

Korean Journal of Radiology18(4), 570–584 (2017)

Lee, J.-G., Jun, S., Cho, Y.-W., Lee, H., Kim, G.B., Seo, J.B., Kim, N.: Deep learning in medical imaging: general overview. Korean Journal of Radiology18(4), 570–584 (2017)

work page 2017
[21]

American Journal of Orthodontics and Dentofacial Orthopedics160(2), 170–192.e4 (2021)

Mohammad-Rahimi, H., Nadimi, M., Rohban, M.H., Shamsoddin, E., Lee, V.Y., Motamedian, S.R.: Machine learning and orthodontics, current trends and the future opportunities: A scoping review. American Journal of Orthodontics and Dentofacial Orthopedics160(2), 170–192.e4 (2021). https://doi.org/10.1016/j.ajodo.2021.02.013

work page doi:10.1016/j.ajodo.2021.02.013 2021
[22]

Graphical Models109, 101071 (2020)

Zhang, J., Li, C., Song, Q., Gao, L., Lai, Y.-K.: Automatic 3D tooth segmenta- tion using convolutional neural networks in harmonic parameter space. Graphical Models109, 101071 (2020)

work page 2020
[23]

In: Healthcare, vol

Thurzo, A., Kosnáčová, H.S., Kurilová, V., Kosmel’, S., Beňuš, R., Moravansk` y, N., Kováč, P., Kuracinová, K.M., Palkovič, M., Varga, I.: Use of advanced artificial intelligence in forensic medicine, forensic anthropology and clinical anatomy. In: Healthcare, vol. 9(11), p. 1545 (2021)

work page 2021
[24]

Journal of Pharmacy and Bioallied Sciences16(Suppl 3), S2585–S2587 (2024)

Kumar, N., Malik, B., Dubey, A., Kaur, H., Mujoo, S., Jugade, S.C., Gulia, S.K.: Long-Term Trends in the Utilization of Cone Beam Computed Tomography in Oral and Maxillofacial Radiology. Journal of Pharmacy and Bioallied Sciences16(Suppl 3), S2585–S2587 (2024). https://doi.org/10.4103/jpbs.jpbs_295_24

work page doi:10.4103/jpbs.jpbs_295_24 2024
[25]

Bassed, R.B., Briggs, C., Drummer, O.H.: Analysis of time of closure of the spheno- occipitalsynchondrosisusingcomputedtomography.ForensicScienceInternational 200(1–3), 161–164 (2010)

work page 2010
[26]

PLOS ONE17(7), e0269198 (2022)

Atici, S.F., Ansari, R., Allareddy, V., Suhaym, O., Cetin, A.E., Elnagar, M.H.: Fully automated determination of the cervical vertebrae maturation stages using deep learning with directional filters. PLOS ONE17(7), e0269198 (2022)

work page 2022

[1] [1]

American Journal of Orthodontics and Dentofacial Orthopedics122(4), 380–385 (2002)

Mito, T., Sato, K., Mitani, H.: Cervical vertebral bone age in girls. American Journal of Orthodontics and Dentofacial Orthopedics122(4), 380–385 (2002). https://doi.org/10.1067/mod.2002.126896

work page doi:10.1067/mod.2002.126896 2002

[2] [2]

Journal of Children’s Orthopaedics4(5), 467–470 (2010)

Calfee, R.P., Sutter, M., Steffen, J.A., Goldfarb, C.A.: Skeletal and chronological ages in American adolescents: current findings in skeletal maturation. Journal of Children’s Orthopaedics4(5), 467–470 (2010)

work page 2010

[3] [3]

Hormone Research57(Suppl

Karlberg, J.: Secular trends in pubertal development. Hormone Research57(Suppl. 2), 19–30 (2002)

work page 2002

[4] [4]

Indian Journal of Endocrinology and Metabolism18(Suppl 1), S63–S71 (2014)

De Sanctis, V., Di Maio, S., Soliman, A.T., Raiola, G., Elalaily, R., Millimaggi, G.: Hand X-ray in pediatric endocrinology: Skeletal age assessment and beyond. Indian Journal of Endocrinology and Metabolism18(Suppl 1), S63–S71 (2014)

work page 2014

[5] [5]

Australian Dental Journal62, 33–50 (2017)

Scarfe, W.C., Azevedo, B., Toghyani, S., Farman, A.G.: Cone beam computed tomographic imaging in orthodontics. Australian Dental Journal62, 33–50 (2017)

work page 2017

[6] [6]

Journal of Applied Clinical Medical Physics22(2), 85–89 (2021)

Olch, A.J., Alaei, P.: How low can you go? A CBCT dose reduction study. Journal of Applied Clinical Medical Physics22(2), 85–89 (2021)

work page 2021

[7] [7]

The Angle Orthodontist 91(4), 538–543 (2021)

Alhazmi,A.,Aldossary,M.,Palomo,J.M.,Hans,M.,Latimer,B.,Simpson,S.:Cor- relation of spheno-occipital synchondrosis fusion stages with a hand-wrist skeletal maturity index: a cone beam computed tomography study. The Angle Orthodontist 91(4), 538–543 (2021)

work page 2021

[8] [8]

Angle Orthodontist94(6), 641–647 (2024)

Booth, E., Viana, G., Shirazi, S., Miller, S., Sellke, T., Elnagar, M., Viana, M., At- sawasuwan, P.: Correlations of spheno-occipital synchondrosis, cervical vertebrae, midpalatal suture, and third molar maturation stages. Angle Orthodontist94(6), 641–647 (2024). https://doi.org/10.2319/041224-295.1

work page doi:10.2319/041224-295.1 2024

[9] [9]

PLoS ONE11(8), e0161104 (2016)

Fernández-Pérez, M.J., Alarcón, J.A., McNamara Jr., J.A., Velasco-Torres, M., Benavides, E., Galindo-Moreno, P., Catena, A.: Spheno-occipital synchondrosis fusion correlates with cervical vertebrae maturation. PLoS ONE11(8), e0161104 (2016)

work page 2016

[10] [10]

The Anatomical Record147(1), 15–23 (1963)

Powell, T.V., Brodie, A.G.: Closure of the spheno-occipital synchondrosis. The Anatomical Record147(1), 15–23 (1963)

work page 1963

[11] [11]

American Journal of Physical Anthropology157(1), 42–57 (2015)

Lottering, N., MacGregor, D.M., Alston, C.L., Gregory, L.S.: Ontogeny of the spheno-occipital synchondrosis in a modern Queensland, Australian population using computed tomography. American Journal of Physical Anthropology157(1), 42–57 (2015)

work page 2015

[12] [12]

Journal of Forensic Dental Sciences5(2), 72–76 (2013)

Krishan, K., Kanchan, T.: Evaluation of spheno-occipital synchondrosis: A review of literature and considerations from forensic anthropologic point of view. Journal of Forensic Dental Sciences5(2), 72–76 (2013)

work page 2013

[13] [13]

Turkish Journal of Orthodontics36(3), 186 (2023)

Tashayyodi, N., Kajan, Z.D., Ostovarrad, F., Khosravifard, N.: Relationship of the Fusion Stage of Spheno-Occipital Synchondrosis with Midpalatal and Zygomatico- maxillary Sutures on Cone-Beam Computed Tomography Scans of Patients Aged Between 7 and 21 Years. Turkish Journal of Orthodontics36(3), 186 (2023)

work page 2023

[14] [14]

PLoS ONE12(8), e0183305 (2017)

Alhazmi, A., Vargas, E., Palomo, J.M., Hans, M., Latimer, B., Simpson, S.: Timing and rate of spheno-occipital synchondrosis closure and its relationship to puberty. PLoS ONE12(8), e0183305 (2017)

work page 2017

[15] [15]

The Angle Orthodontist80(1), 106–110 (2010)

Leonardi, R., Cutrera, A., Barbato, E.: Rapid maxillary expansion affects the spheno-occipital synchondrosis in youngsters: a study with low-dose computed to- mography. The Angle Orthodontist80(1), 106–110 (2010)

work page 2010

[16] [16]

The Angle Orthodontist91(6), 822–829 (2021)

Leonardi, R., Ronsivalle, V., Lagravere, M.O., Barbato, E., Isola, G., Lo Giudice, A.: Three-dimensional assessment of the spheno-occipital synchondrosis and clivus Title Suppressed Due to Excessive Length 15 aftertooth-borneandbone-bornerapidmaxillaryexpansion:aretrospectiveCBCT study using voxel-based superimposition. The Angle Orthodontist91(6), 822–829 (2021)

work page 2021

[17] [17]

Plastic and Reconstructive Surgery134(3), 504–510 (2014)

Goldstein, J.A., Paliga, J.T., Wink, J.D., Bartlett, S.P., Nah, H.-D., Taylor, J.A.: Earlier evidence of spheno-occipital synchondrosis fusion correlates with severity of midface hypoplasia in patients with syndromic craniosynostosis. Plastic and Reconstructive Surgery134(3), 504–510 (2014)

work page 2014

[18] [18]

Journal of Oral and Maxillofacial Surgery72(6), 1173–1179 (2014)

Tahiri, Y., Paliga, J.T., Vossough, A., Bartlett, S.P., Taylor, J.A.: The spheno- occipital synchondrosis fuses prematurely in patients with Crouzon syndrome and midface hypoplasia compared with age- and gender-matched controls. Journal of Oral and Maxillofacial Surgery72(6), 1173–1179 (2014)

work page 2014

[19] [19]

Dentomaxillofacial Radiology44(2015)

Kapila, S.D., Nervina, J.M.: CBCT in orthodontics: Assessment of treatment out- comes and indications for its use. Dentomaxillofacial Radiology44(2015). Preprint at https://doi.org/10.1259/dmfr.20140282

work page doi:10.1259/dmfr.20140282 2015

[20] [20]

Korean Journal of Radiology18(4), 570–584 (2017)

Lee, J.-G., Jun, S., Cho, Y.-W., Lee, H., Kim, G.B., Seo, J.B., Kim, N.: Deep learning in medical imaging: general overview. Korean Journal of Radiology18(4), 570–584 (2017)

work page 2017

[21] [21]

American Journal of Orthodontics and Dentofacial Orthopedics160(2), 170–192.e4 (2021)

Mohammad-Rahimi, H., Nadimi, M., Rohban, M.H., Shamsoddin, E., Lee, V.Y., Motamedian, S.R.: Machine learning and orthodontics, current trends and the future opportunities: A scoping review. American Journal of Orthodontics and Dentofacial Orthopedics160(2), 170–192.e4 (2021). https://doi.org/10.1016/j.ajodo.2021.02.013

work page doi:10.1016/j.ajodo.2021.02.013 2021

[22] [22]

Graphical Models109, 101071 (2020)

Zhang, J., Li, C., Song, Q., Gao, L., Lai, Y.-K.: Automatic 3D tooth segmenta- tion using convolutional neural networks in harmonic parameter space. Graphical Models109, 101071 (2020)

work page 2020

[23] [23]

In: Healthcare, vol

Thurzo, A., Kosnáčová, H.S., Kurilová, V., Kosmel’, S., Beňuš, R., Moravansk` y, N., Kováč, P., Kuracinová, K.M., Palkovič, M., Varga, I.: Use of advanced artificial intelligence in forensic medicine, forensic anthropology and clinical anatomy. In: Healthcare, vol. 9(11), p. 1545 (2021)

work page 2021

[24] [24]

Journal of Pharmacy and Bioallied Sciences16(Suppl 3), S2585–S2587 (2024)

Kumar, N., Malik, B., Dubey, A., Kaur, H., Mujoo, S., Jugade, S.C., Gulia, S.K.: Long-Term Trends in the Utilization of Cone Beam Computed Tomography in Oral and Maxillofacial Radiology. Journal of Pharmacy and Bioallied Sciences16(Suppl 3), S2585–S2587 (2024). https://doi.org/10.4103/jpbs.jpbs_295_24

work page doi:10.4103/jpbs.jpbs_295_24 2024

[25] [25]

Bassed, R.B., Briggs, C., Drummer, O.H.: Analysis of time of closure of the spheno- occipitalsynchondrosisusingcomputedtomography.ForensicScienceInternational 200(1–3), 161–164 (2010)

work page 2010

[26] [26]

PLOS ONE17(7), e0269198 (2022)

Atici, S.F., Ansari, R., Allareddy, V., Suhaym, O., Cetin, A.E., Elnagar, M.H.: Fully automated determination of the cervical vertebrae maturation stages using deep learning with directional filters. PLOS ONE17(7), e0269198 (2022)

work page 2022