1
|
Zeng B, Wang H, Xu J, Tu P, Joskowicz L, Chen X. Two-Stage Structure-Focused Contrastive Learning for Automatic Identification and Localization of Complex Pelvic Fractures. IEEE TRANSACTIONS ON MEDICAL IMAGING 2023; 42:2751-2762. [PMID: 37030821 DOI: 10.1109/tmi.2023.3264298] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/19/2023]
Abstract
Pelvic fracture is a severe trauma with a high rate of morbidity and mortality. Accurate and automatic diagnosis and surgical planning of pelvic fracture require effective identification and localization of the fracture zones. This is a challenging task due to the complexity of pelvic fractures, which often exhibit multiple fragments and sites, large fragment size differences, and irregular morphology. We have developed a novel two-stage method for the automatic identification and localization of complex pelvic fractures. Our method is unique in that it allows to combine the symmetry properties of the pelvic anatomy and capture the symmetric feature differences caused by the fracture on both the left and right sides, thereby overcoming the limitations of existing methods which consider only image or geometric features. It implements supervised contrastive learning with a novel Siamese deep neural network, which consists of two weight-shared branches with a structural attention mechanism, to minimize the confusion of local complex structures of the pelvic bones with the fracture zones. A structure-focused attention (SFA) module is designed to capture the spatial structural features and enhances the recognition ability of fracture zones. Comprehensive experiments on 103 clinical CT scans from the publicly available dataset CTPelvic1K show that our method achieves a mean accuracy and sensitivity of 0.92 and 0.93, which are superior to those reported with three SOTA contrastive learning methods and five advanced classification networks, demonstrating the effectiveness of identifying and localizing various types of complex pelvic fractures from clinical CT images.
Collapse
|
2
|
Raja AS, Rodriguez RM, Gupta M, Isaacs ED, Kornblith LZ, Prabhakar A, Saillant N, Schmit PJ, Wei SH, Mower WR. Developing a decision instrument to guide abdominal-pelvic imaging of blunt trauma patients: Methodology and protocol of the NEXUS abdominal-pelvic imaging study. PLoS One 2022; 17:e0271070. [PMID: 35877687 PMCID: PMC9312398 DOI: 10.1371/journal.pone.0271070] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/19/2022] [Accepted: 06/22/2022] [Indexed: 11/19/2022] Open
Abstract
Although computed tomography (CT) of the abdomen and pelvis (A/P) can provide crucial information for managing blunt trauma patients, liberal and indiscriminant imaging is expensive, can delay critical interventions, and unnecessarily exposes patients to ionizing radiation. Currently no definitive recommendations exist detailing which adult blunt trauma patients should receive A/P CT imaging and which patients may safely forego CT. Considerable benefit could be realized by identifying clinical criteria that reliably classify the risk of abdominal and pelvic injuries in blunt trauma patients. Patients identified as “very low risk” by such criteria would be free of significant injury, receive no benefit from imaging and therefore could be safely spared the expense and radiation exposure associated with A/P CT. The goal of this two-phase nationwide multicenter observational study is to derive and validate the use of clinical criteria to stratify the risk of injuries to the abdomen and pelvis among adult blunt trauma patients. We estimate that nation-wide implementation of a rigorously developed decision instrument could safely reduce CT imaging of adult blunt trauma patients by more than 20%, and reduce annual radiographic charges by $180 million, while simultaneously expediting trauma care and decreasing radiation exposure with its attendant risk of radiation-induced malignancy. Prior to enrollment we convened an expert panel of trauma surgeons, radiologists and emergency medicine physicians to develop a consensus definition for clinically significant abdominal and pelvic injury. In the first derivation phase of the study, we will document the presence or absence of preselected candidate criteria, as well as the presence or absence of significant abdominal or pelvic injuries in a cohort of blunt trauma victims. Using recursive partitioning, we will examine combinations of these criteria to identify an optimal “very low risk” subset that identifies injuries with a sensitivity exceeding 98%, excludes injury with a negative predictive value (NPV) greater than 98%, and retains the highest possible specificity and potential to decrease imaging. In Phase 2 of the study we will validate the performance of a decision rule based on these criteria among a new cohort of patients to ensure that the criteria retain high sensitivity, NPV and optimal specificity. Validating the sensitivity of the decision instrument with high statistical precision requires evaluations on 317 blunt trauma patients who have significant abdominal-pelvic injuries, which will in turn require evaluations on approximately 6,340 blunt trauma patients. We will estimate potential reductions in CT imaging by counting the number of abdominal-pelvic CT scans performed on “very low risk” patients. Reductions in charges and radiation exposure will be determined by respectively summing radiographic charges and lifetime decreases in radiation morbidity and mortality for all “very low risk” cases.
Trial registration: Clinicaltrials.gov trial registration number: NCT04937868.
Collapse
Affiliation(s)
- Ali S. Raja
- Department of Emergency Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, United States of America
| | - Robert M. Rodriguez
- Department of Emergency Medicine, San Francisco General Hospital, UCSF School of Medicine, San Francisco, California, United States of America
| | - Malkeet Gupta
- Department of Emergency Medicine, Ronald Reagan UCLA Medical Center, UCLA Geffen School of Medicine, Los Angeles, California, United States of America
- Antelope Valley Hospital Emergency Department, Lancaster, California, United States of America
| | - Eric D. Isaacs
- Department of Emergency Medicine, San Francisco General Hospital, UCSF School of Medicine, San Francisco, California, United States of America
| | - Lucy Z. Kornblith
- Department of Surgery, San Francisco General Hospital, UCSF School of Medicine, San Francisco, California, United States of America
| | - Anand Prabhakar
- Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, United States of America
| | - Noelle Saillant
- Department of Surgery, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, United States of America
| | - Paul J. Schmit
- UCLA Department of Surgery, Ronald Reagan UCLA Medical Center, UCLA Geffen School of Medicine, Los Angeles, California, United States of America
| | - Sindy H. Wei
- UCLA Department of Radiological Sciences, Ronald Reagan UCLA Medical Center, Los Angeles, California, United States of America
| | - William R. Mower
- UCLA Department of Emergency Medicine, Ronald Reagan UCLA Medical Center, Los Angeles, California, United States of America
- * E-mail:
| |
Collapse
|
3
|
The trauma pelvic X-ray: Not all pelvic fractures are created equally. Am J Surg 2022; 224:489-493. [DOI: 10.1016/j.amjsurg.2022.01.009] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/20/2021] [Revised: 12/31/2021] [Accepted: 01/19/2022] [Indexed: 11/24/2022]
|
4
|
Abstract
EPIDEMIOLOGY Traumatic fractures of the pelvic ring are relatively rare, but are associated with increased risk of mortality. Depending on injury mechanism and main vector of energy impact, a distinction is made between anteroposterior compression, lateral compression, and vertical shear (Young and Burgess classification), while the stability-related classification according to Tile distinguishes between type A (stable), type B (rotationally unstable) and type C (completely unstable). A comprehensive modern classification is given by the AO/OTA. RADIOLOGICAL FRACTURE DIAGNOSIS Plain pelvis x‑rays lack sufficient sensitivity but are still used to detect highly unstable pelvic fractures. CT has superior sensitivity and specificity. In addition to fracture classification, CT allows reliable assessment of associated vascular and bladder/urethral injuries and large soft tissue hemorrhage. MRI is unparalleled in showing bone marrow edema, cauda and plexus complications, and peripelvic soft tissue damage. MRI may also prove to be a valuable diagnostic tool for pelvic ring injuries in children, adolescents and young women, provided they are hemodynamically stable. Angiography, ultrasonography and bone scintigraphy are additional important diagnostic and therapeutic options. PRACTICAL RECOMMENDATIONS Knowledge of basic pelvic trauma mechanisms is important to understand the potential severity of traumatic pelvic fractures and to classify them correctly in terms of stability. Being familiar with typical concomitant injuries in pelvic ring fractures allows reliable diagnosis and their communication with the clinician. CT remains the "diagnostic workhorse". In fragile pelvic fractures, the often prolonged course with fracture progression must be taken into account, which is why MRI is of particular importance herein.
Collapse
Affiliation(s)
- Thomas Grieser
- Klinik für Diagnostische und Interventionelle Radiologie, Universitätsklinikum Augsburg, Stenglinstr. 2, 86156, Augsburg, Deutschland.
| |
Collapse
|
5
|
Shyu JY, Khurana B, Soto JA, Biffl WL, Camacho MA, Diercks DB, Glanc P, Kalva SP, Khosa F, Meyer BJ, Ptak T, Raja AS, Salim A, West OC, Lockhart ME. ACR Appropriateness Criteria® Major Blunt Trauma. J Am Coll Radiol 2020; 17:S160-S174. [PMID: 32370960 DOI: 10.1016/j.jacr.2020.01.024] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/16/2020] [Accepted: 01/22/2020] [Indexed: 11/20/2022]
Abstract
This review assesses the appropriateness of various imaging studies for adult major blunt trauma or polytrauma in the acute setting. Trauma is the leading cause of mortality for people in the United States <45 years of age, and the fourth leading cause of death overall. Imaging, in particular CT, plays a critical role in the management of these patients, and a number of indications are discussed in this publication, including patients who are hemodynamically stable or unstable; patients with additional injuries to the face, extremities, chest, bowel, or urinary system; and pregnant patients. Excluded from consideration in this review are penetrating traumatic injuries, burns, and injuries to pediatric patients. Patients with suspected injury to the head and spine are also discussed more specifically in other appropriateness criteria documents. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.
Collapse
Affiliation(s)
- Jeffrey Y Shyu
- Research Author, Brigham & Women's Hospital, Boston, Massachusetts
| | - Bharti Khurana
- Principal Author, Brigham & Women's Hospital, Boston, Massachusetts.
| | - Jorge A Soto
- Research Author, Boston University School of Medicine, Boston, Massachusetts
| | - Walter L Biffl
- Scripps Memorial Hospital La Jolla, La Jolla, California; American Association for the Surgery of Trauma
| | - Marc A Camacho
- The University of South Florida Morsani College of Medicine, Tampa, Florida
| | - Deborah B Diercks
- University of Texas Southwestern Medical Center, Dallas, Texas; American College of Emergency Physicians
| | - Phyllis Glanc
- University of Toronto and Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
| | | | - Faisal Khosa
- Vancouver General Hospital, Vancouver, British Columbia, Canada
| | | | - Thomas Ptak
- University of Maryland Medical Center, Baltimore, Maryland
| | - Ali S Raja
- Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; Society for Academic Emergency Medicine
| | - Ali Salim
- Brigham & Women's Hospital, Boston, Massachusetts; American College of Surgeons
| | - O Clark West
- UTHealth McGovern Medical School, Houston, Texas
| | - Mark E Lockhart
- Specialty Chair, University of Alabama at Birmingham, Birmingham, Alabama
| |
Collapse
|
6
|
van Trigt J, Schep NWL, Peters RW, Goslings JC, Schepers T, Halm JA. Routine pelvic X-rays in asymptomatic hemodynamically stable blunt trauma patients: A meta-analysis. Injury 2018; 49:2024-2031. [PMID: 30220636 DOI: 10.1016/j.injury.2018.09.009] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/21/2018] [Revised: 07/30/2018] [Accepted: 09/05/2018] [Indexed: 02/02/2023]
Abstract
INTRODUCTION There is no consensus on how pelvic X-rays should be ordered selectively in blunt trauma patients which may save time, reduce radiation exposure and costs. The aim of this systematic review and meta-analysis was to assess the need for routine pelvic X-rays in awake, respiratory and hemodynamically (HD) stable blunt trauma patients without signs of pelvic fracture. Criteria to identify patients who could safely forgo pelvic X-ray were evaluated. METHODS A literature search was performed for prospective comparative cohort studies. Inclusion criteria were: blunt force trauma, hemodynamically and respiratory stable and awake patients, physical examination (PE) for pelvic fractures was adequately described, and the reliability of negative PE findings could be evaluated. Primary outcome was the negative predictive value (NPV) of PE for all and for clinically relevant pelvic fractures. Additionally sensitivity, specificity and positive predictive value (PPV) were calculated. RESULTS Ten studies were included; yielding a total of 11,423 patients. The NPV of PE for all pelvic fractures ranged from 0.96 to 1.00 with a median of 0.996. Combining studies, total NPV was 0.991. For clinically relevant fractures, the NPV of PE ranged from 0.996 to 1.00 with a median of 1.00. In patients with negative findings during PE, 0.9% had fractures, and 0.1% had clinically relevant fractures, none requiring surgical management. CONCLUSIONS In awake, hemodynamically and respiratory stable blunt trauma patients, PE could identify those patients who could safely forgo pelvic X-ray. Selective ordering of pelvic X-ray may lead to a decrease in patient work-up time, lower radiation exposure, and reduce costs. A decision making flow chart is proposed..
Collapse
Affiliation(s)
- Jessica van Trigt
- Trauma Unit, Amsterdam UMC, Location AMC, Amsterdam, The Netherlands
| | - Niels W L Schep
- Department of Surgery, Maasstad Hospital, Rotterdam, The Netherlands
| | - Rolf W Peters
- Trauma Unit, Amsterdam UMC, Location AMC, Amsterdam, The Netherlands
| | - J Carel Goslings
- Department of General and Trauma Surgery, Onze Lieve Vrouwe Gasthuis, Amsterdam, The Netherlands
| | - Tim Schepers
- Trauma Unit, Amsterdam UMC, Location AMC, Amsterdam, The Netherlands
| | - Jens A Halm
- Trauma Unit, Amsterdam UMC, Location AMC, Amsterdam, The Netherlands.
| |
Collapse
|
7
|
Abstract
BACKGROUND This study aimed to develop a tool in identifying traumatized children at low risk of pelvic fracture and to determine the sensitivity of this low-risk model for pelvic fractures. We hypothesized that the proportion of children without predictors with pelvic fracture is less than 1%. METHODS This is a retrospective trauma registry analysis of previously healthy children 1 year to 17 years old presenting to the pediatric emergency department with blunt trauma. Postulated predictors of pelvic fracture on radiograph or computed tomography included pain/abnormal examination result of the pelvis/hip, femur deformity, hematuria, abdominal pain/tenderness, Glasgow Coma Scale (GCS) score of 13 or lower, and hemodynamic instability. We used multivariable logistic regression to identify independent predictors of fracture. RESULTS Of 1,121 eligible patients (mean [SD] age, 8.5 [4.6] years), 87 (7.8%) had pelvic fracture. Independent predictors included pain/abnormal examination result of the pelvis/hip (odds ratio [OR], 16.7; 95% confidence interval [CI], 9.6-29.1), hematuria (OR, 6.6; 95% CI, 3.0-14.6), femoral deformity (OR, 5.9; 95% CI, 3.1-11.3), GCS score of 13 or lower (OR, 2.4; 95% CI, 1.3-4.3), and hemodynamic instability (OR, 3.4; 95% CI, 1.7-6.9). One of 590 children (0.2%; 95% CI, 0-0.5%) without predictors had pelvic fractures versus 86 (16.2%) of 531 in those with one or more predictors (OR, 119; 95% CI, 16.6-833). One of 87 children with pelvic fractures had no predictors (1.1%; 95% CI, 0-3%). When assuming a 100% radiography rate, this tool saves 53% pelvic radiographs. CONCLUSION Children with multiple blunt trauma without pain/abnormal examination result of the pelvis/hip, femur deformity, hematuria, abdominal pain/tenderness, GCS score of 13 or lower, or hemodynamic instability constitute a low-risk population for pelvic fracture, with less than 0.5% risk rate. This population does not require routine pelvic imaging. LEVEL OF EVIDENCE Therapeutic study, level IV.
Collapse
|
8
|
Kwok MY, Yen K, Atabaki S, Adelgais K, Garcia M, Quayle K, Kooistra J, Bonsu BK, Page K, Borgialli D, Kuppermann N, Holmes JF. Sensitivity of Plain Pelvis Radiography in Children With Blunt Torso Trauma. Ann Emerg Med 2015; 65:63-71.e1. [DOI: 10.1016/j.annemergmed.2014.06.017] [Citation(s) in RCA: 6] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/06/2014] [Revised: 06/10/2014] [Accepted: 06/16/2014] [Indexed: 11/15/2022]
|