Low Dimension Medical Images and Generative Deep Learning Models Can Help to Reduce X-Ray Radiation Exposure of Patients
Keywords:
Medical imaging, Medical image quality, Chest X-ray, X-ray image size, Radiation dosage, Radiation exposureAbstract
Background: X-ray and other radiation-based diagnosis form a critical step in many clinical investigations, including early detection of diseases. Deep learning based methods to derive diagnosis from medical images have been shown to be highly accurate in this regard. However, the radiological images collected for this purpose continue to be guided by what is suitable for clinical practitioners to visually interpret them, ignoring the possibility that machines can detect patterns better than the human eye, making the high dimension images unnecessary. On the other hand, image analysis studies have primarily focused on classification accuracy, ignoring the diagnostic tradeoffs with radiation exposure.
Methods: Chest X-ray images from medical datasets have been modeled using EfficientNetB0 deep learning model by reducing the images to different pixel sizes: 1 x 1, 2 x 2, 4 x 4, 8 x 8, 16 x 16, 32 x 32, 64 x 64, 128 x 128, 224 x 224, 256 x 256 and 300 x 300 pixels. The effect of increasing image size on the predictive power of the model has been studied viz-a-viz the radiation exposure of the patient for collecting a chest X-ray image of that size.
Results: In this work, we show that reduced image sizes from the original X-ray images are capable of accurate diagnosis of medical conditions with little loss in predictive power and propose that potentially lower dimensions than what is needed for visual inspection may be sufficient for the purpose, thereby substantially reducing the risks associated with high radiation dosage, currently practiced for use of images by human interpretation. We also demonstrate how reduced images can be used to generate high dimension versions suitable for visual inspection with the help of generative super-resolution techniques (SRGAN) based on deep learning.
Conclusions: In summary this paper makes a case for low dimension collection of X-ray images, with accurate clinical outcomes and thereby addresses the issue of resolution versus diagnostic accuracy.
References
. Morgan W. XIV. Electrical experiments made in order to ascertain the non-conducting power of a perfect vacuum. Philosophical Transactions of the Royal Society of London. 1785 Dec 31(75):272-8.
. Anderson JG. William Morgan and X-rays. Transactions of the Faculty of Actuaries. 1945 Jan;17:219-21.
. Zhou Y, Yang L, Han M, Huang M, Sun X, Zheng W, Han W, Wang J. Case report on early diagnosis of COVID-19. Disaster Medicine and Public Health Preparedness. 2020 Dec;14(6):e22-5.
. Wall BF, Kendall GM, Edwards AA, Bouffler S, Muirhead CR, Meara JR. What are the risks from medical X-rays and other low dose radiation?. The British journal of radiology. 2006 Apr 1;79(940):285-94.
. Zamanian A, Hardiman CJ. Electromagnetic radiation and human health: A review of sources and effects. High Frequency Electronics. 2005 Jul 20;4(3):16-26.
. Brenner DJ, Hall EJ. Computed tomography—an increasing source of radiation exposure. New England journal of medicine. 2007 Nov 29;357(22):2277-84.
. Luan FJ, Wan Y, Mak KC, Ma CJ, Wang HQ. Cancer and mortality risks of patients with scoliosis from radiation exposure: a systematic review and meta-analysis. European Spine Journal. 2020 Dec;29:3123-34.
. Gislason-Lee AJ. Patient X-ray exposure and ALARA in the neonatal intensive care unit: global patterns. Pediatrics & Neonatology. 2021 Jan 1;62(1):3-10.
. Arad I, Simanovsky N, Braunstein R. Exposure of extremely low birth weight infants to diagnostic X?Rays: a longitudinal study. Acta Paediatrica. 2009 Feb;98(2):266-9.
. Rudan I, Boschi-Pinto C, Biloglav Z, Mulholland K, Campbell H. Epidemiology and etiology of childhood pneumonia. Bulletin of the world health organization. 2008;86:408-16B.
. Hendrick RE. Radiation doses and risks in breast screening. Journal of Breast Imaging. 2020 May;2(3):188-200.
. Smith-Bindman R, Lipson J, Marcus R, Kim KP, Mahesh M, Gould R, De González AB, Miglioretti DL. Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Archives of internal medicine. 2009 Dec 14;169(22):2078-86.
. Rubai SS, Rahman MS, Purohit S, Patwary MK, Moinul AK, Meaze H, Mamun AA. Measurements of entrance surface dose and effective dose of patients in diagnostic radiography. Biomedical Journal. 2018;1:5.
. Barrett HH, Myers KJ, Hoeschen C, Kupinski MA, Little MP. Task-based measures of image quality and their relation to radiation dose and patient risk. Physics in Medicine & Biology. 2015 Jan 7;60(2):R1.
. Grodzins L. Optimum energies for x-ray transmission tomography of small samples: Applications of synchrotron radiation to computerized tomography I. Nuclear Instruments and Methods in Physics Research. 1983 Mar 1;206(3):541-5.
. Brenner DJ, Elliston CD, Hall EJ, Berdon WE. Estimated risks of radiation-induced fatal cancer from pediatric CT. American journal of roentgenology. 2001 Feb;176(2):289-96.
. Wagner LK, Eifel PJ, Geise RA. Potential biological effects following high X-ray dose interventional procedures. Journal of Vascular and Interventional Radiology. 1994 Jan 1;5(1):71-84.
. Geleijns J, Wondergem J. X-ray imaging and the skin: radiation biology, patient dosimetry and observed effects. Radiation protection dosimetry. 2005 May 17;114(1-3):121-5.
. Cherepennikov Y, Gogolev A. Method to reduce radiation exposure in the medical X-ray diagnosti?s. In2012 7th International Forum on Strategic Technology (IFOST) 2012 Sep 18 (pp. 1-5). IEEE.
. Douglas BH, Birkelo CC, Harmon GE, Vaughan HF. Use of Miniature X-Ray Films in Tuberculosis Case Finding. American Journal of Public Health and the Nations Health. 1940 Dec;30(12):1427-30.
. Tan M, Le Q. Efficientnet: Rethinking model scaling for convolutional neural networks. InInternational conference on machine learning 2019 May 24 (pp. 6105-6114). PMLR.
. Deng J, Dong W, Socher R, Li LJ, Li K, Fei-Fei L. ImageNet: A large-scale hierarchical image database. In2009 IEEE conference on computer vision and pattern recognition 2009 Jun 20 (pp. 248-255). Ieee.
. Ledig C, Theis L, Huszár F, Caballero J, Cunningham A, Acosta A, Aitken A, Tejani A, Totz J, Wang Z, Shi W. Photo-realistic single image super-resolution using a generative adversarial network. InProceedings of the IEEE conference on computer vision and pattern recognition 2017 (pp. 4681-4690).
. Tang S, Jing C, Jiang Y, Yang K, Huang Z, Wu H, Cui C, Shi S, Ye X, Tian H, Song D. The effect of image resolution on convolutional neural networks in breast ultrasound. Heliyon. 2023 Aug 1;9(8).
. Sabottke CF, Spieler BM. The effect of image resolution on deep learning in radiography. Radiology: Artificial Intelligence. 2020 Jan 22;2(1):e190015.
. Afshar P, Heidarian S, Naderkhani F, Oikonomou A, Plataniotis KN, Mohammadi A. Covid-caps: A capsule network-based framework for identification of covid-19 cases from x-ray images. Pattern Recognition Letters. 2020 Oct 1;138:638-43.
. Al-Waisy AS, Al-Fahdawi S, Mohammed MA, Abdulkareem KH, Mostafa SA, Maashi MS, Arif M, Garcia-Zapirain B. COVID-CheXNet: hybrid deep learning framework for identifying COVID-19 virus in chest X-rays images. Soft computing. 2023;27(5):2657.
. Gupta A, Gupta S, Katarya R. InstaCovNet-19: A deep learning classification model for the detection of COVID-19 patients using Chest X-ray. Applied Soft Computing. 2021 Feb 1;99:106859.
. Heidari M, Mirniaharikandehei S, Khuzani AZ, Danala G, Qiu Y, Zheng B. Improving the performance of CNN to predict the likelihood of COVID-19 using chest X-ray images with preprocessing algorithms. International journal of medical informatics. 2020 Dec 1;144:104284.
. Minaee S, Kafieh R, Sonka M, Yazdani S, Soufi GJ. Deep-COVID: Predicting COVID-19 from chest X-ray images using deep transfer learning. Medical image analysis. 2020 Oct 1;65:101794.
. Shen C, Zhang K, Tang J. A covid-19 detection algorithm using deep features and discrete social learning particle swarm optimization for edge computing devices. ACM Transactions on Internet Technology (TOIT). 2021 Dec 30;22(3):1-7
. He J, Zhu Q, Zhang K, Yu P, Tang J. An evolvable adversarial network with gradient penalty for COVID-19 infection segmentation. Applied Soft Computing. 2021 Dec 1;113:107947.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Neha Vinayak, Divyansh Pandey , Shandar Ahmad
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors who submit papers with this journal agree to the following terms.
