Learning from a Few Shots: Data-efficient Cervical Vertebral Maturation Assessment

Helen Schneider, Aditya Parikh, Priya Priya, Maximilian Broß, Tom Verhofstadt, Anna Konermann, Rafet Sifa
Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, PMLR 301:1403-1417, 2026.

Abstract

The timing of treatment is a crucial decision in orthodontics. Initiating treatment duringthe appropriate growth phase leads to optimal patient outcomes and can prevent prolongedtreatment durations. The most commonly used method for classifying growth phases iscervical vertebral maturation (CVM) assessment, which categorizes CVM into six stagesbased on the shape and size of the cervical vertebrae. Due to the complexity of manual CVManalysis, it often falls short in performance when assessed visually. Deep learning methodscan assist physicians in classifying CVM stages, thus improving orthodontic workflows andtreatments. However, a significant challenge in deep learning-based CVM assessment isthe limited dataset volume, resulting from difficulties in data collection and annotation.While small training datasets can greatly hinder the model’s generalization performance,research on data-efficient training methods for CVM assessment is still lacking. To the bestof our knowledge, this paper is the first to evaluate the potential of few-shot learning and in-domain transfer learning for CVM assessment. Specifically, we investigate the architecturesResNet18 and MedSam-2D. Few-shot learning enhances classification performance by upto 9%. Additionally, in-domain pre-training (using chest X-ray data) results in a significantperformance increase of up to 4%.

Cite this Paper

BibTeX
@InProceedings{pmlr-v301-schneider26a, title = {Learning from a Few Shots: Data-efficient Cervical Vertebral Maturation Assessment}, author = {Schneider, Helen and Parikh, Aditya and Priya, Priya and Bro{\ss}, Maximilian and Verhofstadt, Tom and Konermann, Anna and Sifa, Rafet}, booktitle = {Proceedings of The 8th International Conference on Medical Imaging with Deep Learning}, pages = {1403--1417}, year = {2026}, editor = {Tasdizen, Tolga and Elhabian, Shireen and Summers, Ronald and Chen, Chen and Koch, Lisa and Zhuang, Yan}, volume = {301}, series = {Proceedings of Machine Learning Research}, month = {09--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v301/main/assets/schneider26a/schneider26a.pdf}, url = {https://proceedings.mlr.press/v301/schneider26a.html}, abstract = {The timing of treatment is a crucial decision in orthodontics. Initiating treatment duringthe appropriate growth phase leads to optimal patient outcomes and can prevent prolongedtreatment durations. The most commonly used method for classifying growth phases iscervical vertebral maturation (CVM) assessment, which categorizes CVM into six stagesbased on the shape and size of the cervical vertebrae. Due to the complexity of manual CVManalysis, it often falls short in performance when assessed visually. Deep learning methodscan assist physicians in classifying CVM stages, thus improving orthodontic workflows andtreatments. However, a significant challenge in deep learning-based CVM assessment isthe limited dataset volume, resulting from difficulties in data collection and annotation.While small training datasets can greatly hinder the model’s generalization performance,research on data-efficient training methods for CVM assessment is still lacking. To the bestof our knowledge, this paper is the first to evaluate the potential of few-shot learning and in-domain transfer learning for CVM assessment. Specifically, we investigate the architecturesResNet18 and MedSam-2D. Few-shot learning enhances classification performance by upto 9%. Additionally, in-domain pre-training (using chest X-ray data) results in a significantperformance increase of up to 4%.} }
Endnote
%0 Conference Paper %T Learning from a Few Shots: Data-efficient Cervical Vertebral Maturation Assessment %A Helen Schneider %A Aditya Parikh %A Priya Priya %A Maximilian Broß %A Tom Verhofstadt %A Anna Konermann %A Rafet Sifa %B Proceedings of The 8th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Tolga Tasdizen %E Shireen Elhabian %E Ronald Summers %E Chen Chen %E Lisa Koch %E Yan Zhuang %F pmlr-v301-schneider26a %I PMLR %P 1403--1417 %U https://proceedings.mlr.press/v301/schneider26a.html %V 301 %X The timing of treatment is a crucial decision in orthodontics. Initiating treatment duringthe appropriate growth phase leads to optimal patient outcomes and can prevent prolongedtreatment durations. The most commonly used method for classifying growth phases iscervical vertebral maturation (CVM) assessment, which categorizes CVM into six stagesbased on the shape and size of the cervical vertebrae. Due to the complexity of manual CVManalysis, it often falls short in performance when assessed visually. Deep learning methodscan assist physicians in classifying CVM stages, thus improving orthodontic workflows andtreatments. However, a significant challenge in deep learning-based CVM assessment isthe limited dataset volume, resulting from difficulties in data collection and annotation.While small training datasets can greatly hinder the model’s generalization performance,research on data-efficient training methods for CVM assessment is still lacking. To the bestof our knowledge, this paper is the first to evaluate the potential of few-shot learning and in-domain transfer learning for CVM assessment. Specifically, we investigate the architecturesResNet18 and MedSam-2D. Few-shot learning enhances classification performance by upto 9%. Additionally, in-domain pre-training (using chest X-ray data) results in a significantperformance increase of up to 4%.
APA
Schneider, H., Parikh, A., Priya, P., Broß, M., Verhofstadt, T., Konermann, A. & Sifa, R.. (2026). Learning from a Few Shots: Data-efficient Cervical Vertebral Maturation Assessment. Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 301:1403-1417 Available from https://proceedings.mlr.press/v301/schneider26a.html.

Related Material