Skip to main content

Advertisement

SpringerLink
Log in

A Patch-Based Deep Learning Approach for Detecting Rib Fractures on Frontal Radiographs in Young Children

  • Original Paper
  • Published:
Journal of Digital Imaging Aims and scope Submit manuscript

Abstract

Chest radiography is the modality of choice for the identification of rib fractures in young children and there is value for the development of computer-aided rib fracture detection in this age group. However, the automated identification of rib fractures on chest radiographs can be challenging due to the need for high spatial resolution in deep learning frameworks. A patch-based deep learning algorithm was developed to automatically detect rib fractures on frontal chest radiographs in children under 2 years old. A total of 845 chest radiographs of children 0–2 years old (median: 4 months old) were manually segmented for rib fractures by radiologists and served as the ground-truth labels. Image analysis utilized a patch-based sliding-window technique, to meet the high-resolution requirements for fracture detection. Standard transfer learning techniques used ResNet-50 and ResNet-18 architectures. Area-under-curve for precision-recall (AUC-PR) and receiver-operating-characteristic (AUC-ROC), along with patch and whole-image classification metrics, were reported. On the test patches, the ResNet-50 model showed AUC-PR and AUC-ROC of 0.25 and 0.77, respectively, and the ResNet-18 showed an AUC-PR of 0.32 and AUC-ROC of 0.76. On the whole-radiograph level, the ResNet-50 had an AUC-ROC of 0.74 with 88% sensitivity and 43% specificity in identifying rib fractures, and the ResNet-18 had an AUC-ROC of 0.75 with 75% sensitivity and 60% specificity in identifying rib fractures. This work demonstrates the utility of patch-based analysis for detection of rib fractures in children under 2 years old. Future work with large cohorts of multi-institutional data will improve the generalizability of these findings to patients with suspicion of child abuse.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Abbreviations

IRB:

Institutional review board

HIPAA:

Health Insurance Portability and Accountability Act

CNN:

Convolutional neural network

OTS:

Off-the-shelf

OTSFT:

Off-the-shelf fine-tuning

IQR:

Inter-quartile range

AUC-PR:

Area under the precision-recall curve

AUC-ROC:

Area under the receiver operator characteristic curve

NICU:

Neonatal intensive care unit

References

  1. Sartorelli KH, Vane DW. The diagnosis and management of children with blunt injury of the chest. Semin Pediatr Surg. 2004 May;13(2):98–105.

    Article  PubMed  Google Scholar 

  2. Ruest S, Kanaan G, Moore JL, Goldberg AP. Pediatric rib fractures identified by chest radiograph: A comparison between accidental and nonaccidental trauma. Pediatr Emerg Care. 2021 Dec 1;37(12):e1409–15.

    Article  PubMed  Google Scholar 

  3. Expert Panel on Pediatric Imaging:, Wootton-Gorges SL, Soares BP, Alazraki AL, Anupindi SA, Blount JP, et al. ACR Appropriateness Criteria® Suspected Physical Abuse-Child. J Am Coll Radiol. 2017 May;14(5S):S338–49.

  4. Meyer JS, Gunderman R, Coley BD, Bulas D, Garber M, Karmazyn B, et al. ACR Appropriateness Criteria(®) on suspected physical abuse-child. J Am Coll Radiol. 2011 Feb;8(2):87–94.

    Article  PubMed  Google Scholar 

  5. Marine MB, Corea D, Steenburg SD, Wanner M, Eckert GJ, Jennings SG, et al. Is the new ACR-SPR practice guideline for addition of oblique views of the ribs to the skeletal survey for child abuse justified? AJR Am J Roentgenol. 2014 Apr;202(4):868–71.

    Article  PubMed  Google Scholar 

  6. Andriole KP, Wolfe JM, Khorasani R, Treves ST, Getty DJ, Jacobson FL, et al. Optimizing analysis, visualization, and navigation of large image data sets: one 5000-section CT scan can ruin your whole day. Radiology. 2011 May;259(2):346–62.

    Article  PubMed  Google Scholar 

  7. Lakhani P, Sundaram B. Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology. 2017 Aug;284(2):574–82.

    Article  PubMed  Google Scholar 

  8. Zhang R, Tie X, Qi Z, Bevins NB, Zhang C, Griner D, et al. Diagnosis of coronavirus disease 2019 pneumonia by using chest radiography: value of artificial intelligence. Radiology. 2021 Feb;298(2):E88–97.

    Article  PubMed  Google Scholar 

  9. Jin L, Yang J, Kuang K, Ni B, Gao Y, Sun Y, et al. Deep-learning-assisted detection and segmentation of rib fractures from CT scans: Development and validation of FracNet. EBioMedicine. 2020 Dec;62:103106.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Wu M, Chai Z, Qian G, Lin H, Wang Q, Wang L, et al. Development and Evaluation of a Deep Learning Algorithm for Rib Segmentation and Fracture Detection from Multicenter Chest CT Images. Radiol Artif Intell. 2021 Sep;3(5):e200248.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Meng XH, Wu DJ, Wang Z, Ma XL, Dong XM, Liu AE, et al. A fully automated rib fracture detection system on chest CT images and its impact on radiologist performance. Skeletal Radiol. 2021 Feb 18;

  12. Zhou QQ, Wang J, Tang W, Hu ZC, Xia ZY, Li XS, et al. Automatic detection and classification of rib fractures on thoracic CT using convolutional neural network: accuracy and feasibility. Korean J Radiol. 2020;21(7):869–79.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Yao L, Guan X, Song X, Tan Y, Wang C, Jin C, et al. Rib fracture detection system based on deep learning. Sci Rep. 2021 Dec 6;11(1):23513.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Abdelhafiz D, Yang C, Ammar R, Nabavi S. Deep convolutional neural networks for mammography: advances, challenges and applications. BMC Bioinformatics. 2019 Jun 6;20(Suppl 11):281.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Lotter W, Sorensen G, Cox D. A Multi-scale CNN and Curriculum Learning Strategy for Mammogram Classification. In: Cardoso MJ, Arbel T, Carneiro G, Syeda-Mahmood T, Tavares JMRS, Moradi M, et al., editors. Deep learning in medical image analysis and multimodal learning for clinical decision support. Cham: Springer International Publishing; 2017. p. 169–77.

    Chapter  Google Scholar 

  16. Sabottke CF, Spieler BM. The effect of image resolution on deep learning in radiography. Radiol Artif Intell. 2020 Jan 22;2(1):e190015.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Shen L, Margolies LR, Rothstein JH, Fluder E, McBride R, Sieh W. Deep learning to improve breast cancer detection on screening mammography. Sci Rep. 2019 Aug 29;9(1):12495.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Xi P, Shu C, Goubran R. Abnormality Detection in Mammography using Deep Convolutional Neural Networks. 2018 IEEE International Symposium on Medical Measurements and Applications (MeMeA). IEEE; 2018. p. 1–6.

    Google Scholar 

  19. Hou L, Samaras D, Kurc TM, Gao Y, Davis JE, Saltz JH. Patch-based Convolutional Neural Network for Whole Slide Tissue Image Classification. Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2016 Jul;2016:2424–33.

    PubMed  PubMed Central  Google Scholar 

  20. Sanchez TR, Nguyen H, Palacios W, Doherty M, Coulter K. Retrospective evaluation and dating of non-accidental rib fractures in infants. Clin Radiol. 2013 Aug;68(8):e467-71.

    Article  CAS  PubMed  Google Scholar 

  21. ImageNet Benchmark (Image Classification) | Papers With Code [Internet]. [cited 2022 Feb 14]. Available from: https://paperswithcode.com/sota/image-classification-on-imagenet

  22. Huang S-T, Huang M-Y, Liu L-R, Tsai M-F, Chiu H-W. Recognition of rib fracture on chest X-ray images with deep learning: a pilot study. Res Sq. 2022 Jul 25;

  23. Tsai A-C, Ou Y-Y, Lin C-H, Chen C-W, Wang J-F. Rib Fracture Diagnosis System on Chest X-Rays with Deep Learning. 2021 9th International Conference on Orange Technology (ICOT). IEEE; 2021. p. 1–4.

  24. Smith RL, Ackerley IM, Wells K, Bartley L, Paisey S, Marshall C. Reinforcement learning for object detection in PET imaging. 2019 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC). IEEE; 2019. p. 1–4.

    Google Scholar 

  25. Ackerley I, Spezi E, Prakash V, Smith RL, Scuffham JW, Lewis E, et al. Using deep machine learning to detect esophageal lesions in PET-CT scans. In: Gimi B, Krol A, editors. Medical Imaging 2019: Biomedical Applications in Molecular, Structural, and Functional Imaging. SPIE; 2019. p. 26.

  26. Naik N, Madani A, Esteva A, Keskar NS, Press MF, Ruderman D, et al. Deep learning-enabled breast cancer hormonal receptor status determination from base-level H&E stains. Nat Commun. 2020 Nov 16;11(1):5727.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Yao J, Zhu X, Jonnagaddala J, Hawkins N, Huang J. Whole slide images based cancer survival prediction using attention guided deep multiple instance learning networks. Med Image Anal. 2020 Oct;65:101789.

    Article  PubMed  Google Scholar 

  28. Mobadersany P, Cooper LAD, Goldstein JA. GestAltNet: aggregation and attention to improve deep learning of gestational age from placental whole-slide images. Lab Invest. 2021 Jul;101(7):942–51.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Lu MY, Williamson DFK, Chen TY, Chen RJ, Barbieri M, Mahmood F. Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Images. arXiv. 2020;

  30. Candemir S, Nguyen XV, Folio LR, Prevedello LM. Training Strategies for Radiology Deep Learning Models in Data-limited Scenarios. Radiol Artif Intell. 2021 Nov;3(6):e210014.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Ruest S, Kanaan G, Moore JL, Goldberg AP. The Prevalence of Rib Fractures Incidentally Identified by Chest Radiograph among Infants and Toddlers. J Pediatr. 2019 Jan;204:208–13.

    Article  PubMed  Google Scholar 

  32. Zhang W, Deng L, Zhang L, Wu D. A survey on negative transfer. IEEE/CAA J Autom Sinica. 2023 Feb;10(2):305–29.

    Article  Google Scholar 

  33. www.acr.org/-/media/ACR/Files/Practice-Parameters/Skeletal-Survey.pdf.

  34. https://www.rcr.ac.uk/system/files/publication/field_publication_files/bfcr174_suspected_physical_abuse.pdf [Internet]. [cited 2023 Jan 2]. Available from: https://www.rcr.ac.uk/system/files/publication/field_publication_files/bfcr174_suspected_physical_abuse.pdf

Download references

Author information

Author notes

  1. Adarsh Ghosh and Daniella Patton contributed equally to this work.

  2. Susan Sotardi and Michael Francavilla contributed equally to this work as senior authors.

Authors and Affiliations

  1. Department of Radiology, Children’s Hospital of Philadelphia, Philadelphia, PA, USA

    Adarsh Ghosh, Daniella Patton, Saurav Bose, M. Katherine Henry, Minhui Ouyang, Hao Huang, Arastoo Vossough, Raymond Sze, Susan Sotardi & Michael Francavilla

  2. Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA

    M. Katherine Henry, Minhui Ouyang, Hao Huang, Arastoo Vossough, Raymond Sze, Susan Sotardi & Michael Francavilla

  3. Safe Place: Center for Child Protection and Health, Division of General Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, PA, USA

    M. Katherine Henry

  4. Department of Radiology, Cincinnati Children’s Hospital and Medical Center, Cincinnati, OH, USA

    Adarsh Ghosh, Susan Sotardi & Michael Francavilla

  5. Cincinnati Children’s Burnet Campus, 3333 Burnet Avenue, Cincinnati, OH, 45229, USA

    Adarsh Ghosh

Authors
  1. Adarsh Ghosh
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Daniella Patton
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Saurav Bose
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. M. Katherine Henry
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Minhui Ouyang
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Hao Huang
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Arastoo Vossough
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Raymond Sze
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Susan Sotardi
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Michael Francavilla
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Adarsh Ghosh, Daniella Patton, and Saurav Bose. The first draft of the manuscript was written by Adarsh Ghosh and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Adarsh Ghosh.

Ethics declarations

Ethics Approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Children’s Hospital of Philadelphia.

Consent to Participate

Requirement for informed consent was waived.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose. Children’s Hospital of Philadelphia has received payment for the expert testimony of Dr. Henry when subpoenaed to provide testimony in cases of suspected abuse.

Additional information

Publisher's Note

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

Summary Statement: A patch-based deep learning approach can identify rib fractures in pediatric chest radiographs and can be used as an aid to detect rib fractures in this vulnerable population.

Key Results:

∎ In a retrospective study of 815 young children < 2 years of age, patch-based deep learning algorithms flagged rib fractures in pediatric chest radiographs (accuracy on patch level: 91.02%).

∎ The performance of the artificial intelligence algorithm for fracture detection had an area under the receiver operating characteristic curve of 0.87 and 0.75 on validation and test sets, respectively.

∎ On an independent test set, the ResNet-50 model obtained a sensitivity of 88.41% for rib fractures.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ghosh, A., Patton, D., Bose, S. et al. A Patch-Based Deep Learning Approach for Detecting Rib Fractures on Frontal Radiographs in Young Children. J Digit Imaging 36, 1302–1313 (2023). https://doi.org/10.1007/s10278-023-00793-1

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10278-023-00793-1

Keywords

Search

Navigation