1
|
Tan H, Xu H, Yu N, Yu Y, Duan H, Fan Q, Zhanyu T. The value of deep learning-based computer aided diagnostic system in improving diagnostic performance of rib fractures in acute blunt trauma. BMC Med Imaging 2023; 23:55. [PMID: 37055752 PMCID: PMC10099632 DOI: 10.1186/s12880-023-01012-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/20/2021] [Accepted: 04/04/2023] [Indexed: 04/15/2023] Open
Abstract
BACKGROUND To evaluate the value of a deep learning-based computer-aided diagnostic system (DL-CAD) in improving the diagnostic performance of acute rib fractures in patients with chest trauma. MATERIALS AND METHODS CT images of 214 patients with acute blunt chest trauma were retrospectively analyzed by two interns and two attending radiologists independently firstly and then with the assistance of a DL-CAD one month later, in a blinded and randomized manner. The consensusdiagnosis of fib fracture by another two senior thoracic radiologists was regarded as reference standard. The rib fracture diagnostic sensitivity, specificity, positive predictive value, diagnostic confidence and mean reading time with and without DL-CAD were calculated and compared. RESULTS There were 680 rib fracture lesions confirmed as reference standard among all patients. The diagnostic sensitivity and positive predictive value of interns weresignificantly improved from (68.82%, 84.50%) to (91.76%, 93.17%) with the assistance of DL-CAD, respectively. Diagnostic sensitivity and positive predictive value of attendings aided by DL-CAD (94.56%, 95.67%) or not aided (86.47%, 93.83%), respectively. In addition, when radiologists were assisted by DL-CAD, the mean reading time was significantly reduced, and diagnostic confidence was significantly enhanced. CONCLUSIONS DL-CAD improves the diagnostic performance of acute rib fracture in chest trauma patients, which increases the diagnostic confidence, sensitivity, and positive predictive value for radiologists. DL-CAD can advance the diagnostic consistency of radiologists with different experiences.
Collapse
Affiliation(s)
- Hui Tan
- Department of Radiology, Affiliated Hospital of Shaanxi University of Chinese Medicine, Xianyang, China
| | - Hui Xu
- Peter Boris Centre for Addiction Research, McMaster University & St. Joseph's Health Care Hamilton, 100 West 5th Street, Hamilton, ON, L8P 3R2, Canada.
| | - Nan Yu
- Department of Radiology, Affiliated Hospital of Shaanxi University of Chinese Medicine, Xianyang, China
| | - Yong Yu
- Department of Radiology, Affiliated Hospital of Shaanxi University of Chinese Medicine, Xianyang, China
| | - Haifeng Duan
- Department of Radiology, Affiliated Hospital of Shaanxi University of Chinese Medicine, Xianyang, China
| | - Qiuju Fan
- Department of Radiology, Affiliated Hospital of Shaanxi University of Chinese Medicine, Xianyang, China.
| | - Tian Zhanyu
- Institute of Medical Technology, Shaanxi University of Chinese Medicine, Xianyang, China
| |
Collapse
|
2
|
Ebrahimian R, Souri Z, Feizkhah A, Mobayen M, Eslami Kenarsari H, Esmailzadeh M, Ghorbani M, Mirhedayati S, Bagheri Toolaroud P. Evaluation of the Spiral Chest CT Scan Findings in Patients with Multiple Trauma. Bull Emerg Trauma 2023; 11:19-25. [PMID: 36818057 PMCID: PMC9923035 DOI: 10.30476/beat.2023.97214.1402] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/13/2022] [Revised: 12/22/2023] [Accepted: 12/25/2023] [Indexed: 02/24/2023] Open
Abstract
Objective To evaluate the spiral chest computed tomography (CT) scan findings in patients with multiple trauma during the COVID-19 pandemic. Methods This retrospective study was performed on multiple trauma patients admitted to a tertiary hospital in the north of Iran in 2020. All patients with multiple trauma who had undergone a chest spiral CT were included in this study. Furthermore, the data analysis was performed through descriptive and analytical statistics using SPSS software. Results A total of 600 patients were included over the study period. The mean age of patients was 48.2±20.3 years. Of the total, 496 (65.3%) patients had blunt chest injuries, and 104 (34.7%) had penetrating chest injuries. Falling was the most common mechanical cause of chest trauma in 270 patients (45%). Surgical interventions were performed in 110 (18.3%) patients. A total of 276 (46%) patients had chest injuries identified by CT scans. Many patients (15.6%) had ground-glass lung opacity in the CT scan reports. Lung consolidation, pneumothorax, lung contusion, hemothorax, and rib fractures were the most common. Conclusion Due to the high frequency of typical findings in spiral CT scan examinations, obtaining a reliable history of trauma severity, injury mechanism, and a detailed physical examination is recommended before prescribing a CT scan for patients.
Collapse
Affiliation(s)
- Ramin Ebrahimian
- Clinical Research Development Unit of Poursina Hospital, Guilan University of Medical Sciences, Rasht, Iran,Department of Surgery, Faculty of Medicine, Guilan University of Medical Sciences, Rasht, Iran
| | - Zoubin Souri
- Department of Radiology, Faculty of Medicine, Guilan University of Medical Sciences, Rasht, Iran
| | - Alireza Feizkhah
- Burn and Regenerative Medicine Research Center, Guilan University of Medical Sciences, Rasht, Iran,Department of Medical Physics, School of Medicine, Guilan University of Medical Sciences, Rasht, Iran
| | - Mohammadreza Mobayen
- Burn and Regenerative Medicine Research Center, Guilan University of Medical Sciences, Rasht, Iran
| | - Habib Eslami Kenarsari
- Clinical Research Development Unit of Poursina Hospital, Guilan University of Medical Sciences, Rasht, Iran
| | - Mojdeh Esmailzadeh
- Burn and Regenerative Medicine Research Center, Guilan University of Medical Sciences, Rasht, Iran
| | - Mohsen Ghorbani
- Clinical Research Development Unit of Poursina Hospital, Guilan University of Medical Sciences, Rasht, Iran
| | - Soroush Mirhedayati
- Clinical Research Development Unit of Poursina Hospital, Guilan University of Medical Sciences, Rasht, Iran
| | - Parissa Bagheri Toolaroud
- Burn and Regenerative Medicine Research Center, Guilan University of Medical Sciences, Rasht, Iran,Health Information Management Research Center, Kashan University of Medical Sciences, Kashan, Iran
| |
Collapse
|
3
|
Jin L, Yang J, Kuang K, Ni B, Gao Y, Sun Y, Gao P, Ma W, Tan M, Kang H, Chen J, Li M. Deep-learning-assisted detection and segmentation of rib fractures from CT scans: Development and validation of FracNet. EBioMedicine 2020; 62:103106. [PMID: 33186809 PMCID: PMC7670192 DOI: 10.1016/j.ebiom.2020.103106] [Citation(s) in RCA: 57] [Impact Index Per Article: 14.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/30/2020] [Revised: 10/17/2020] [Accepted: 10/19/2020] [Indexed: 11/24/2022] Open
Abstract
BACKGROUND Diagnosis of rib fractures plays an important role in identifying trauma severity. However, quickly and precisely identifying the rib fractures in a large number of CT images with increasing number of patients is a tough task, which is also subject to the qualification of radiologist. We aim at a clinically applicable automatic system for rib fracture detection and segmentation from CT scans. METHODS A total of 7,473 annotated traumatic rib fractures from 900 patients in a single center were enrolled into our dataset, named RibFrac Dataset, which were annotated with a human-in-the-loop labeling procedure. We developed a deep learning model, named FracNet, to detect and segment rib fractures. 720, 60 and 120 patients were randomly split as training cohort, tuning cohort and test cohort, respectively. Free-Response ROC (FROC) analysis was used to evaluate the sensitivity and false positives of the detection performance, and Intersection-over-Union (IoU) and Dice Coefficient (Dice) were used to evaluate the segmentation performance of predicted rib fractures. Observer studies, including independent human-only study and human-collaboration study, were used to benchmark the FracNet with human performance and evaluate its clinical applicability. A annotated subset of RibFrac Dataset, including 420 for training, 60 for tuning and 120 for test, as well as our code for model training and evaluation, was open to research community to facilitate both clinical and engineering research. FINDINGS Our method achieved a detection sensitivity of 92.9% with 5.27 false positives per scan and a segmentation Dice of 71.5%on the test cohort. Human experts achieved much lower false positives per scan, while underperforming the deep neural networks in terms of detection sensitivities with longer time in diagnosis. With human-computer collobration, human experts achieved higher detection sensitivities than human-only or computer-only diagnosis. INTERPRETATION The proposed FracNet provided increasing detection sensitivity of rib fractures with significantly decreased clinical time consumed, which established a clinically applicable method to assist the radiologist in clinical practice. FUNDING A full list of funding bodies that contributed to this study can be found in the Acknowledgements section. The funding sources played no role in the study design; collection, analysis, and interpretation of data; writing of the report; or decision to submit the article for publication .
Collapse
Affiliation(s)
- Liang Jin
- Radiology Department, Huadong Hospital, affiliated to Fudan University, Shanghai, China
| | - Jiancheng Yang
- Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai, P.R. China; MoE Key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University, Shanghai, P.R. China; Dianei Technology, Shanghai, P.R. China
| | | | - Bingbing Ni
- Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai, P.R. China; MoE Key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University, Shanghai, P.R. China; Huawei Hisilicon, Shanghai, P.R. China
| | - Yiyi Gao
- Radiology Department, Huadong Hospital, affiliated to Fudan University, Shanghai, China
| | - Yingli Sun
- Radiology Department, Huadong Hospital, affiliated to Fudan University, Shanghai, China
| | - Pan Gao
- Radiology Department, Huadong Hospital, affiliated to Fudan University, Shanghai, China
| | - Weiling Ma
- Radiology Department, Huadong Hospital, affiliated to Fudan University, Shanghai, China
| | - Mingyu Tan
- Radiology Department, Huadong Hospital, affiliated to Fudan University, Shanghai, China
| | - Hui Kang
- Dianei Technology, Shanghai, P.R. China
| | | | - Ming Li
- Radiology Department, Huadong Hospital, affiliated to Fudan University, Shanghai, China; Institute of Functional and Molecular Medical Imaging, Fudan University, Shanghai, China.
| |
Collapse
|
4
|
The Use of Chest Computed Tomographic Angiography in Blunt Trauma Pediatric Population. Pediatr Emerg Care 2020; 36:e682-e685. [PMID: 29406478 DOI: 10.1097/pec.0000000000001422] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/25/2022]
Abstract
INTRODUCTION Blunt chest trauma in children is common. Although rare, associated major thoracic vascular injuries (TVIs) are lethal potential sequelae of these mechanisms. The preferred study for definitive diagnosis of TVI in stable patients is computed tomographic angiography imaging of the chest. This imaging modality is, however, associated with high doses of ionizing radiation that represent significant carcinogenic risk for pediatric patients. The aim of the present investigation was to define the incidence of TVI among blunt pediatric trauma patients in an effort to better elucidate the usefulness of computed tomographic angiography use in this population. METHODS A retrospective cohort study was conducted including all blunt pediatric (age < 14 y) trauma victims registered in Israeli National Trauma Registry maintained by Gertner Institute for Epidemiology and Health Policy Research between the years 1997 and 2015. Data collected included age, sex, mechanism of injury, Glasgow Coma Scale, Injury Severity Score, and incidence of chest named vessel injuries. Statistical analysis was performed using SAS statistical software version 9.2 (SAS Institute Inc, Cary, NC). RESULTS Among 433,325 blunt trauma victims, 119,821patients were younger than 14 years. Twelve (0.0001%, 12/119821) of these children were diagnosed with TVI. The most common mechanism in this group was pedestrian hit by a car. Mortality was 41.7% (5/12). CONCLUSIONS Thoracic vascular injury is exceptionally rare among pediatric blunt trauma victims but does contribute to the high morbidity and mortality seen with blunt chest trauma. Computed tomographic angiography, with its associated radiation exposure risk, should not be used as a standard tool after trauma in injured children. Clinical protocols are needed in this population to minimize radiation risk while allowing prompt identification of life-threatening injuries.
Collapse
|
5
|
Abstract
The aim of this study was to assess the applicability of low-dose thoracic computed tomography (CT) in the diagnosis of rib fractures.A total of 37 trauma patients were selected for CT scanning using a noise index (NI) model. Each patient was scanned at both NI = 11 and NI = 26, while the other scanning parameters were kept the same. The scanning dose length product (DLP) and effective dose (ED) were recorded after each examination. Two radiologists diagnosed the rib fractures by degree (I, II, III, and IV) using Bone Reading software and axial images. Image quality was scored by 2 experienced radiologists using a 5-point scale. The numbers and degrees of rib fractures for different NIs were recorded and tested using the Chi-squared test. The interobserver differences were determined by kappa statistics.The CTDIvols and EDs for NI = 11 and NI = 26 were 9.82 ± 4.78, 5.75 ± 2.75, and 2.14 ± 1.19 and 1.24 ± 0.73, respectively; the latter was decreased by 78.2% and 78.4% relative to the former. Low-dose thoracic CT was feasible for the auxiliary diagnosis of rib fractures using Bone Reading software (P > .05). There was perfect interobserver concordance in terms of diagnostic acceptability (kappa = 0.931, 0.905).The use of an appropriate low-dose CT scanning technique is satisfactory for the assessment and diagnosis of rib fractures.
Collapse
|
6
|
Abstract
Introduction Chest trauma may be blunt or penetrating and the chest is the third most common trauma region. It is a significant cause of mortality. Multidetector computed tomography (MDCT) has been an increasingly used method to evaluate chest trauma because of its high success in detecting tissue and organ injuries. Herein, we aimed to present MDCT findings in patients with blunt and penetrating chest trauma admitted to our department. Methods A total of 240 patients admitted to the emergency department of our hospital between April 2012 and July 2013 with a diagnosis of chest trauma who underwent MDCT evaluations were included. Most of the patients were male (83.3%) and victims of a blunt chest trauma. The images were analyzed with respect to the presence of fractures of bony structures, hemothorax, pneumothorax, mediastinal organ injury, and pulmonary and vascular injuries. Results MDCT images of the 240 patients yielded a prevalence of 41.7% rib fractures, 11.2% scapular fractures, and 7.5% clavicle fractures. The prevalence of thoracic vertebral fracture was 13.8% and that of sternal fracture was 3.8%. The prevalence of hemothorax, pneumothorax, pneumomediastinum, and subcutaneous emphysema was 34.6%, 62.1%, 9.6%, and 35.4%, respectively. The prevalence of rib, clavicle, and thoracic vertebral fractures and pulmonary contusion was higher in the blunt trauma group, whereas the prevalence of hemothorax, subcutaneous emphysema, diaphragmatic injury, and other vascular lacerations was significantly higher in the penetrating trauma group than in the blunt trauma group (p<0.05). Conclusion MDCT images may yield a high prevalence of fracture of bony structures, soft tissue lacerations, and vascular lesions, which should be well understood by radiologists dealing with trauma.
Collapse
Affiliation(s)
| | - Serdar Onat
- Department of Chest Surgery, Dicle University School of Medicine, Diyarbakir
| | | |
Collapse
|
7
|
Erratum: Usefulness of low dose chest CT for initial evaluation of blunt chest trauma: Erratum. Medicine (Baltimore) 2017; 96:e7234. [PMID: 31305677 PMCID: PMC5466247 DOI: 10.1097/md.0000000000007234] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/04/2022] Open
Abstract
[This corrects the article DOI: 10.1097/MD.0000000000005888.].
Collapse
|