Deep Learning for Automated Detection and Localization of Traumatic Abdominal Solid Organ Injuries on CT Scans

Journal of Imaging Informatics in Medicine
Chi‐Tung Cheng1, Hao-Xiang Lin1, Chih-Po Hsu1, Huan‐Wu Chen2, Jen-Fu Huang1, Chi-Hsun Hsieh1, Chih-Yuan Fu1, I‐Fang Chung3, Chien‐Hung Liao1
1Department of Trauma and Emergency Surgery, Chang Gung Memorial Hospital, Linkou, Chang Gung University, Taoyuan, Taiwan
2Department of Medical Imaging & Intervention, Chang Gung Memorial Hospital, Linkou, Chang Gung University, Taoyuan, Taiwan
3Institute of Biomedical Informatics, National Yang Ming Chiao Tung University, Taipei, Taiwan

Tóm tắt

AbstractComputed tomography (CT) is the most commonly used diagnostic modality for blunt abdominal trauma (BAT), significantly influencing management approaches. Deep learning models (DLMs) have shown great promise in enhancing various aspects of clinical practice. There is limited literature available on the use of DLMs specifically for trauma image evaluation. In this study, we developed a DLM aimed at detecting solid organ injuries to assist medical professionals in rapidly identifying life-threatening injuries. The study enrolled patients from a single trauma center who received abdominal CT scans between 2008 and 2017. Patients with spleen, liver, or kidney injury were categorized as the solid organ injury group, while others were considered negative cases. Only images acquired from the trauma center were enrolled. A subset of images acquired in the last year was designated as the test set, and the remaining images were utilized to train and validate the detection models. The performance of each model was assessed using metrics such as the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value based on the best Youden index operating point. The study developed the models using 1302 (87%) scans for training and tested them on 194 (13%) scans. The spleen injury model demonstrated an accuracy of 0.938 and a specificity of 0.952. The accuracy and specificity of the liver injury model were reported as 0.820 and 0.847, respectively. The kidney injury model showed an accuracy of 0.959 and a specificity of 0.989. We developed a DLM that can automate the detection of solid organ injuries by abdominal CT scans with acceptable diagnostic accuracy. It cannot replace the role of clinicians, but we can expect it to be a potential tool to accelerate the process of therapeutic decisions for trauma care.

Từ khóa


Tài liệu tham khảo

Mehta N, Babu S, Venugopal K. An experience with blunt abdominal trauma: evaluation, management and outcome. Clin Pract 2014;4:599. https://doi.org/10.4081/cp.2014.599.

Hsieh T-M, Cheng Tsai T, Liang J-L, Che Lin C. Non-operative management attempted for selective high grade blunt hepatosplenic trauma is a feasible strategy. World J Emerg Surg 2014;9:51. https://doi.org/10.1186/1749-7922-9-51.

Kendall JL, Kestler AM, Whitaker KT, Adkisson M-M, Haukoos JS. Blunt abdominal trauma patients are at very low risk for intra-abdominal injury after emergency department observation. West J Emerg Med 2011;12:496–504. https://doi.org/10.5811/westjem.2010.11.2016.

Wiik Larsen J, Søreide K, Søreide JA, Tjosevik K, Kvaløy JT, Thorsen K. Epidemiology of abdominal trauma: An age- and sex-adjusted incidence analysis with mortality patterns. Injury 2022;53:3130–8. https://doi.org/10.1016/j.injury.2022.06.020.

Stassen NA, Bhullar I, Cheng JD, Crandall M, Friese R, Guillamondegui O, et al. Nonoperative management of blunt hepatic injury: an Eastern Association for the Surgery of Trauma practice management guideline. J Trauma Acute Care Surg 2012;73:S288–93. https://doi.org/10.1097/TA.0b013e318270160d.

Stassen NA, Bhullar I, Cheng JD, Crandall ML, Friese RS, Guillamondegui OD, et al. Selective nonoperative management of blunt splenic injury: an Eastern Association for the Surgery of Trauma practice management guideline. J Trauma Acute Care Surg 2012;73:S294–300. https://doi.org/10.1097/TA.0b013e3182702afc.

Altman AL, Haas C, Dinchman KH, Spirnak JP. Selective nonoperative management of blunt grade 5 renal injury. J Urol 2000;164:27–30; discussion 30–1.

Raza M, Abbas Y, Devi V, Prasad KVS, Rizk KN, Nair PP. Non operative management of abdominal trauma – a 10 years review. World J Emerg Surg 2013;8:14. https://doi.org/10.1186/1749-7922-8-14.

Achatz G, Schwabe K, Brill S, Zischek C, Schmidt R, Friemert B, et al. Correction to: Diagnostic options for blunt abdominal trauma. Eur J Trauma Emerg Surg 2022;48:3591. https://doi.org/10.1007/s00068-020-01456-4.

Stuhlfaut JW, Anderson SW, Soto JA. Blunt abdominal trauma: current imaging techniques and CT findings in patients with solid organ, bowel, and mesenteric injury. Semin Ultrasound CT MR 2007;28:115–29. https://doi.org/10.1053/j.sult.2007.01.004.

Ingram M-CE, Siddharthan RV, Morris AD, Hill SJ, Travers CD, McKracken CE, et al. Hepatic and splenic blush on computed tomography in children following blunt abdominal trauma: Is intervention necessary? J Trauma Acute Care Surg 2016;81:266–70. https://doi.org/10.1097/TA.0000000000001114.

Gamanagatti S, Rangarajan K, Kumar A, Jineesh. Blunt abdominal trauma: imaging and intervention. Curr Probl Diagn Radiol 2015;44:321–36. https://doi.org/10.1067/j.cpradiol.2015.02.005.

Mahmood I, Tawfek Z, Abdelrahman Y, Siddiuqqi T, Abdelrahman H, El-Menyar A, et al. Significance of computed tomography finding of intra-abdominal free fluid without solid organ injury after blunt abdominal trauma: time for laparotomy on demand. World J Surg 2014;38:1411–5. https://doi.org/10.1007/s00268-013-2427-5.

Yanar H, Ertekin C, Taviloglu K, Kabay B, Bakkaloglu H, Guloglu R. Nonoperative treatment of multiple intra-abdominal solid organ injury after blunt abdominal trauma. J Trauma 2008;64:943–8. https://doi.org/10.1097/TA.0b013e3180342023.

El-Menyar A, Abdelrahman H, Al-Hassani A, Peralta R, AbdelAziz H, Latifi R, et al. Single Versus Multiple Solid Organ Injuries Following Blunt Abdominal Trauma. World J Surg 2017;41:2689–96. https://doi.org/10.1007/s00268-017-4087-3.

Safavi A, Skarsgard ED, Rhee P, Zangbar B, Kulvatunyou N, Tang A, et al. Trauma center variation in the management of pediatric patients with blunt abdominal solid organ injury: a national trauma data bank analysis. J Pediatr Surg 2016;51:499–502. https://doi.org/10.1016/j.jpedsurg.2015.08.012.

Cimbanassi S, Chiara O, Leppaniemi A, Henry S, Scalea TM, Shanmuganathan K, et al. Nonoperative management of abdominal solid-organ injuries following blunt trauma in adults: Results from an International Consensus Conference. J Trauma Acute Care Surg 2018;84:517–31. https://doi.org/10.1097/TA.0000000000001774.

van Schuppen J, Olthof DC, Wilde JCH, Beenen LFM, van Rijn RR, Goslings JC. Diagnostic accuracy of a step-up imaging strategy in pediatric patients with blunt abdominal trauma. Eur J Radiol 2014;83:206–11. https://doi.org/10.1016/j.ejrad.2013.09.024.

Pinto A, Reginelli A, Pinto F, Lo Re G, Midiri F, Muzj C, et al. Errors in imaging patients in the emergency setting. Br J Radiol 2016;89:20150914. https://doi.org/10.1259/bjr.20150914.

Liu X, Faes L, Kale AU, Wagner SK, Fu DJ, Bruynseels A, et al. A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis. Lancet Digit Health 2019;1:e271–97. https://doi.org/10.1016/S2589-7500(19)30123-2.

Cheng C-T, Wang Y, Chen H-W, Hsiao P-M, Yeh C-N, Hsieh C-H, et al. A scalable physician-level deep learning algorithm detects universal trauma on pelvic radiographs. Nat Commun 2021;12:1066. https://doi.org/10.1038/s41467-021-21311-3.

Choi J, Mavrommati K, Li NY, Patil A, Chen K, Hindin DI, et al. Scalable deep learning algorithm to compute percent pulmonary contusion among patients with rib fractures. J Trauma Acute Care Surg 2022;93:461–6. https://doi.org/10.1097/TA.0000000000003619.

Avendi MR, Kheradvar A, Jafarkhani H. A combined deep-learning and deformable-model approach to fully automatic segmentation of the left ventricle in cardiac MRI. Med Image Anal 2016;30:108–19. https://doi.org/10.1016/j.media.2016.01.005.

Liu J, Varghese B, Taravat F, Eibschutz LS, Gholamrezanezhad A. An Extra Set of Intelligent Eyes: Application of Artificial Intelligence in Imaging of Abdominopelvic Pathologies in Emergency Radiology. Diagnostics (Basel) 2022;12. https://doi.org/10.3390/diagnostics12061351.

Dreizin D, Chen T, Liang Y, Zhou Y, Paes F, Wang Y, et al. Added value of deep learning-based liver parenchymal CT volumetry for predicting major arterial injury after blunt hepatic trauma: a decision tree analysis. Abdom Radiol (NY) 2021;46:2556–66. https://doi.org/10.1007/s00261-020-02892-x.

Chiu I-M, Lin C-HR, Yau F-FF, Cheng F-J, Pan H-Y, Lin X-H, et al. Use of a Deep-Learning Algorithm to Guide Novices in Performing Focused Assessment With Sonography in Trauma. JAMA Netw Open 2023;6:e235102. https://doi.org/10.1001/jamanetworkopen.2023.5102.

Choi J, Alawa J, Tennakoon L, Forrester JD. DeepBackRib: Deep learning to understand factors associated with readmissions after rib fractures. J Trauma Acute Care Surg 2022;93:757–61. https://doi.org/10.1097/TA.0000000000003791.

Dreizin D, Staziaki PV, Khatri GD, Beckmann NM, Feng Z, Liang Y, et al. Artificial intelligence CAD tools in trauma imaging: a scoping review from the American Society of Emergency Radiology (ASER) AI/ML Expert Panel. Emerg Radiol 2023;30:251–65. https://doi.org/10.1007/s10140-023-02120-1.

Agrawal A, Khatri GD, Khurana B, Sodickson AD, Liang Y, Dreizin D. A survey of ASER members on artificial intelligence in emergency radiology: trends, perceptions, and expectations. Emerg Radiol 2023;30:267–77. https://doi.org/10.1007/s10140-023-02121-0.

Harris RJ, Kim S, Lohr J, Towey S, Velichkovich Z, Kabachenko T, et al. Classification of Aortic Dissection and Rupture on Post-contrast CT Images Using a Convolutional Neural Network. J Digit Imaging 2019;32:939–46. https://doi.org/10.1007/s10278-019-00281-5.

Huang S, Zhou Z, Qian X, Li D, Guo W, Dai Y. Automated quantitative assessment of pediatric blunt hepatic trauma by deep learning-based CT volumetry. Eur J Med Res 2022;27:305. https://doi.org/10.1186/s40001-022-00943-1.

Wang J, Wood A, Gao C, Najarian K, Gryak J. Automated Spleen Injury Detection Using 3D Active Contours and Machine Learning. Entropy 2021;23. https://doi.org/10.3390/e23040382.

Zhou Y, Dreizin D, Wang Y, Liu F, Shen W, Yuille AL. External Attention Assisted Multi-Phase Splenic Vascular Injury Segmentation With Limited Data. IEEE Trans Med Imaging 2022;41:1346–57. https://doi.org/10.1109/TMI.2021.3139637.

Dreizin D, Zhou Y, Chen T, Li G, Yuille AL, McLenithan A, et al. Deep learning-based quantitative visualization and measurement of extraperitoneal hematoma volumes in patients with pelvic fractures: Potential role in personalized forecasting and decision support. J Trauma Acute Care Surg 2020;88:425. https://doi.org/10.1097/TA.0000000000002566.

Dreizin D, Nixon B, Hu J, Albert B, Yan C, Yang G, et al. A pilot study of deep learning-based CT volumetry for traumatic hemothorax. Emerg Radiol 2022. https://doi.org/10.1007/s10140-022-02087-5.

Dreizin D, Zhou Y, Fu S, Wang Y, Li G, Champ K, et al. A Multiscale Deep Learning Method for Quantitative Visualization of Traumatic Hemoperitoneum at CT: Assessment of Feasibility and Comparison with Subjective Categorical Estimation. Radiol Artif Intell 2020;2:e190220. https://doi.org/10.1148/ryai.2020190220.

Dreizin D, Zhou Y, Zhang Y, Tirada N, Yuille AL. Performance of a Deep Learning Algorithm for Automated Segmentation and Quantification of Traumatic Pelvic Hematomas on CT. J Digit Imaging 2020;33:243–51. https://doi.org/10.1007/s10278-019-00207-1.

Cheng C-T, Lin H-S, Hsu C-P, Chen H-W, Huang J-F, Fu C-Y, et al. The three-dimensional weakly supervised deep learning algorithm for traumatic splenic injury detection and sequential localization: an experimental study. Int J Surg 2023;109:1115. https://doi.org/10.1097/JS9.0000000000000380.

Wasserthal J, Meyer M, Breit H-C, Cyriac J, Yang S, Segeroth M. TotalSegmentator: robust segmentation of 104 anatomical structures in CT images. arXiv [eessIV] 2022.

Kozar RA, Crandall M, Shanmuganathan K, Zarzaur BL, Coburn M, Cribari C, et al. Organ injury scaling 2018 update: Spleen, liver, and kidney. J Trauma Acute Care Surg 2018;85:1119–22. https://doi.org/10.1097/TA.0000000000002058.

Woo S, Park J, Lee J-Y, Kweon IS. CBAM: Convolutional Block Attention Module. Computer Vision – ECCV 2018, Cham: Springer International Publishing; 2018, p. 3–19. https://doi.org/10.1007/978-3-030-01234-2_1.

Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-cam: Visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE international conference on computer vision, 2017, p. 618–26.

Salahuddin Z, Woodruff HC, Chatterjee A, Lambin P. Transparency of deep neural networks for medical image analysis: A review of interpretability methods. Comput Biol Med 2022;140:105111. https://doi.org/10.1016/j.compbiomed.2021.105111.

Malhotra AK, Fabian TC, Croce MA, Gavin TJ, Kudsk KA, Minard G, et al. Blunt hepatic injury: a paradigm shift from operative to nonoperative management in the 1990s. Ann Surg 2000;231:804–13. https://doi.org/10.1097/00000658-200006000-00004.

Coccolini F, Coimbra R, Ordonez C, Kluger Y, Vega F, Moore EE, et al. Liver trauma: WSES 2020 guidelines. World J Emerg Surg 2020;15:24. https://doi.org/10.1186/s13017-020-00302-7.

Sharma OP, Oswanski MF, Singer D. Role of repeat computerized tomography in nonoperative management of solid organ trauma. Am Surg 2005;71:244–9.

Lee JT, Slade E, Uyeda J, Steenburg SD, Chong ST, Tsai R, et al. American Society of Emergency Radiology Multicenter Blunt Splenic Trauma Study: CT and Clinical Findings. Radiology 2021;299:122–30. https://doi.org/10.1148/radiol.2021202917.

Zhao Y, Wang X, Che T, Bao G, Li S. Multi-task deep learning for medical image computing and analysis: A review. Comput Biol Med 2023;153:106496. https://doi.org/10.1016/j.compbiomed.2022.106496.

Monteiro M, Newcombe VFJ, Mathieu F, Adatia K, Kamnitsas K, Ferrante E, et al. Multiclass semantic segmentation and quantification of traumatic brain injury lesions on head CT using deep learning: an algorithm development and multicentre validation study. The Lancet Digital Health 2020;2:e314–22. https://doi.org/10.1016/S2589-7500(20)30085-6.

Farzaneh N, Stein EB, Soroushmehr R, Gryak J, Najarian K. A deep learning framework for automated detection and quantitative assessment of liver trauma. BMC Med Imaging 2022;22:39. https://doi.org/10.1186/s12880-022-00759-9.

Chen H, Unberath M, Dreizin D. Toward automated interpretable AAST grading for blunt splenic injury. Emerg Radiol 2023;30:41–50. https://doi.org/10.1007/s10140-022-02099-1.

Farzaneh N, Reza Soroushmehr SM, Patel H, Wood A, Gryak J, Fessell D, et al. Automated Kidney Segmentation for Traumatic Injured Patients through Ensemble Learning and Active Contour Modeling. Conf Proc IEEE Eng Med Biol Soc 2018;2018:3418–21. https://doi.org/10.1109/EMBC.2018.8512967.

Tulum G, Teomete U, Cuce F, Ergin T, Koksal M, Dandin O, et al. Automated segmentation of the injured kidney due to abdominal trauma. J Med Syst 2019;44:5. https://doi.org/10.1007/s10916-019-1476-1.

Hamghalam M, Moreland R, Gomez D, Simpson A, Lin HM, Jandaghi AB, et al. Machine Learning Detection and Characterization of Splenic Injuries on Abdominal Computed Tomography. Can Assoc Radiol J 2024:8465371231221052. https://doi.org/10.1177/08465371231221052.

Esteva A, Chou K, Yeung S, Naik N, Madani A, Mottaghi A, et al. Deep learning-enabled medical computer vision. NPJ Digit Med 2021;4:5. https://doi.org/10.1038/s41746-020-00376-2.

Egger J, Gsaxner C, Pepe A, Pomykala KL, Jonske F, Kurz M, et al. Medical deep learning—A systematic meta-review. Comput Methods Programs Biomed 2022;221:106874. https://doi.org/10.1016/j.cmpb.2022.106874.

Adams-McGavin RC, Tafur M, Vlachou PA, Wu M, Brassil M, Crivellaro P, et al. Interrater Agreement of CT Grading of Blunt Splenic Injuries: Does the AAST Grading Need to Be Reimagined? Can Assoc Radiol J 2023:8465371231184425. https://doi.org/10.1177/08465371231184425.

Thông tin
Thông tin xuất bản

Journal of Imaging Informatics in Medicine

Thông tin tác giả