Ewha Med J.  2024 Apr;47(2):e23. 10.12771/emj.2024.e23.

An accurate pediatric bone age prediction model using deep learning and contrast conversion

• Choi DH
• Ahn SH
• Lee R

• KMID: 2556312
• DOI: http://doi.org/10.12771/emj.2024.e23

Abstract

Objectives
This study aimed to develop an accurate pediatric bone age prediction model by utilizing deep learning models and contrast conversion techniques, in order to improve growth assessment and clinical decision-making in clinical practice.
Methods
The study employed a variety of deep learning models and contrast conversion techniques to predict bone age. The training dataset consisted of pediatric left-hand X-ray images, each annotated with bone age and sex information. Deep learning models, including a convolutional neural network , Residual Network 50 , Visual Geometry Group 19, Inception V3, and Xception were trained and assessed using the mean absolute error (MAE). For the test data, contrast conversion techniques including fuzzy contrast enhancement, contrast limited adaptive histogram equalization (HE) , and HE were implemented. The quality of the images was evaluated using peak signal-to-noise ratio (SNR), mean squared error, SNR, coefficient of variation, and contrast-to-noise ratio metrics. The bone age prediction results using the test data were evaluated based on the MAE and root mean square error, and the t-test was performed.
Results
The Xception model showed the best performance (MAE=41.12). HE exhibited superior image quality, with higher SNR and coefficient of variation values than other methods. Additionally, HE demonstrated the highest contrast among the techniques assessed, with a contrast-to-noise ratio value of 1.29. Improvements in bone age prediction resulted in a decline in MAE from 2.11 to 0.24, along with a decrease in root mean square error from 0.21 to 0.02.
Conclusion
This study demonstrates that preprocessing the data before model training does not significantly affect the performance of bone age prediction when comparing contrast-converted images with original images.

Keyword

Bone age measurement; X-ray image; Deep learning

Figure

• 

  Fig. 1. Histograms depicting (A) the sex distribution and (B) the monthly age distribution for males and females.

• 

  Fig. 2. Comparison of bone age and model predictions. (A) CNN, (B) ResNet 50, (C) VGG 19, (D) Inception V3, and (E) Xception. The blue line represents the actual bone age, while the red dot represents the predicted result. CNN, convolutional neural network; ResNet 50, Residual Network 50; VGG 19, Visual Geometry Group 19.

• 

  Fig. 3. The original left-hand X-ray image and the image after applying each contrast conversion algorithm. (A) Original image, (B) FCE algorithm applied, (C) HE algorithm applied, (D) CLAHE algorithm applied. HE, histogram equalization; CLAHE, contrast limited adaptive histogram equalization; FCE, fuzzy contrast enhancement.

• 

  Fig. 4. Quantitative analysis results of images obtained using the contrast conversion algorithm. (A) PSNR and MSE results; (B) SNR, COV, and CNR results. FCE, fuzzy contrast enhancement; HE, histogram equalization; CLAHE, contrast limited adaptive histogram equalization; PSNR, peak signal-to-noise ratio; MSE, mean squared error; SNR, signal-to-noise ratio; COV, coefficient of variation; CNR, contrast-to-noise ratio.

Reference

References

1. Cavallo F, Mohn A, Chiarelli F, Giannini C. Evaluation of bone age in children: a mini-review. Frontiers Pediatr. 2021; 9:580314. DOI: 10.3389/fped.2021.580314. PMID: 33777857. PMCID: PMC7994346.

2. Wang S, Wang X, Shen Y, He B, Zhao X, Cheung PWH, et al. An ensemble-based densely-connected deep learning system for assessment of skeletal maturity. IEEE Trans Syst Man Cybern Syst. 2020; 52(1):426–437. DOI: 10.1109/TSMC.2020.2997852.

3. Ferrillo M, Curci C, Roccuzzo A, Migliario M, Invernizzi M, de Sire A. Reliability of cervical vertebral maturation compared to hand-wrist for skeletal maturation assessment in growing subjects: a systematic review. J Back Musculoskelet Rehabil. 2021; 34(6):925–936. DOI: 10.3233/BMR-210003. PMID: 33998532.

4. Satoh M, Hasegawa Y. Factors affecting prepubertal and pubertal bone age progression. Front Endocrinol. 2022; 13:967711. DOI: 10.3389/fendo.2022.967711. PMID: 36072933. PMCID: PMC9441639.

5. Ahn KS, Bae B, Jang WY, Lee JH, Oh S, Kim BH, et al. Assessment of rapidly advancing bone age during puberty on elbow radiographs using a deep neural network model. Eur Radiol. 2021; 31(12):8947–8955. DOI: 10.1007/s00330-021-08096-1. PMID: 34115194.

6. Maratova K, Zemkova D, Sedlak P, Pavlikova M, Amaratunga SA, Krasnicanova H, et al. A comprehensive validation study of the latest version of BoneXpert on a large cohort of Caucasian children and adolescents. Front Endocrinol. 2023; 14:1130580. DOI: 10.3389/fendo.2023.1130580. PMID: 37033216. PMCID: PMC10079872.

7. Son SJ, Song Y, Kim N, Do Y, Kwak N, Lee MS, et al. TW3-based fully automated bone age assessment system using deep neural networks. IEEE Access. 2019; 7:33346–33358. DOI: 10.1109/ACCESS.2019.2903131.

8. Gilsanz V, Ratib O. Hand bone age: a digital atlas of skeletal maturity. Berlin: Springer;2005.

9. Kim PH, Yoon HM, Kim JR, Hwang JY, Choi JH, Hwang J, et al. Bone age assessment using artificial intelligence in Korean pediatric population: a comparison of deep-learning models trained with healthy chronological and Greulich-Pyle ages as labels. Korean J Radiol. 2023; 24(11):1151–1163. DOI: 10.3348/kjr.2023.0092. PMID: 37899524. PMCID: PMC10613838.