Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jun;36(3):893-901.
doi: 10.1007/s10278-023-00774-4. Epub 2023 Jan 19.

Using Transfer Learning of Convolutional Neural Network on Neck Radiographs to Identify Acute Epiglottitis

Yang-Tse Lin  1 , Ben-Chang Shia  2 , Chia-Jung Chang  3 , Yueh Wu  4 , Jheng-Dao Yang  5 , Jiunn-Horng Kang  6   7   8
Affiliations

Using Transfer Learning of Convolutional Neural Network on Neck Radiographs to Identify Acute Epiglottitis

Yang-Tse Lin et al. J Digit Imaging. 2023 Jun.

Abstract

Acute epiglottitis (AE) is a life-threatening condition and needs to be recognized timely. Diagnosis of AE with a lateral neck radiograph yields poor reliability and sensitivity. Convolutional neural networks (CNN) are powerful tools to assist the analysis of medical images. This study aimed to develop an artificial intelligence model using CNN-based transfer learning to identify AE in lateral neck radiographs. All cases in this study are from two hospitals, a medical center, and a local teaching hospital in Taiwan. In this retrospective study, we collected 251 lateral neck radiographs of patients with AE and 936 individuals without AE. Neck radiographs obtained from patients without and with AE were used as the input for model transfer learning in a pre-trained CNN including Inception V3, Densenet201, Resnet101, VGG19, and Inception V2 to select the optimal model. We used five-fold cross-validation to estimate the performance of the selected model. The confusion matrix of the final model was analyzed. We found that Inception V3 yielded the best results as the optimal model among all pre-train models. Based on the average value of the fivefold cross-validation, the confusion metrics were obtained: accuracy = 0.92, precision = 0.94, recall = 0.90, and area under the curve (AUC) = 0.96. Using the Inception V3-based model can provide an excellent performance to identify AE based on radiographic images. We suggest using the CNN-based model which can offer a non-invasive, accurate, and fast diagnostic method for AE in the future.

Keywords: Acute epiglottitis; Artificial intelligence; Convolutional neural networks; Emergency medicine; Lateral neck radiographs; Medical errors; Transfer learning; X-ray.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interest.

Figures

Fig. 1
Fig. 1
Flowchart of data collection and procedure of data augmentation/model training
Fig. 2
Fig. 2
Procedures involved in the analysis of neck X-ray images: from left to right: raw image, region of interest (ROI) detection, resize and normalization, and class activation mapping (CAM) visualization
Fig. 3
Fig. 3
The area under the curve (AUC) of the cross-validation results
Fig. 4
Fig. 4
Comparison of different feature regions in the convolutional neural network (CNN) model by class activation mapping (CAM)
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Rafei K, Lichenstein R. Airway infectious disease emergencies. Pediatr Clin North Am. 2006;53(2):215–242. doi: 10.1016/j.pcl.2005.10.001. - DOI - PubMed
    1. Sato S, Kuratomi Y, Inokuchi A. Pathological characteristics of the epiglottis relevant to acute epiglottitis. Auris Nasus Larynx. 2012;39(5):507–511. doi: 10.1016/j.anl.2011.10.015. - DOI - PubMed
    1. Li CJ, Aronowitz P. Sore throat, odynophagia, hoarseness, and a muffled, high-pitched voice. Cleve Clin J Med. 2013;80(3):144–145. doi: 10.3949/ccjm.80a.12056. - DOI - PubMed
    1. B. Westerhuis, M.G. Bietz, J. Lindemann, Acute epiglottitis in adults: an under-recognized and life-threatening condition, S D Med 66(8) (2013) 309–11, 313. - PubMed
    1. Lee DR, Lee CH, Won YK, Suh DI, Roh EJ, Lee MH, Chung EH. Clinical characteristics of children and adolescents with croup and epiglottitis who visited 146 Emergency Departments in Korea. Korean J Pediatr. 2015;58(10):380–385. doi: 10.3345/kjp.2015.58.10.380. - DOI - PMC - PubMed

LinkOut - more resources