A Comparative Analysis of Diagnostic Accuracy of Focused Assessment With Sonography for Trauma Performed by Emergency Medicine and Radiology Residents

Applications of Artificial Intelligence in Acute Abdominal Imaging

Artificial intelligence (AI) is a rapidly growing field with significant implications for radiology. Acute abdominal pain is a common clinical presentation that can range from benign conditions to life-threatening emergencies. The critical nature of these situations renders emergent abdominal imaging an ideal candidate for AI applications. CT, radiographs, and ultrasound are the most common modalities for imaging evaluation of these patients. For each modality, numerous studies have assessed the performance of AI models for detecting common pathologies, such as appendicitis, bowel obstruction, and cholecystitis. The capabilities of these models range from simple classification to detailed severity assessment. This narrative review explores the evolution, trends, and challenges in AI applications for evaluating acute abdominal pathologies. We review implementations of AI for non-traumatic and traumatic abdominal pathologies, with discussion of potential clinical impact, challenges, and future directions for the technology.

Diagnostic Accuracy and Prognostic Significance of Point-Of-Care Ultrasound (POCUS) for Traumatic Cervical Spine in Emergency care setting: A Comparison of clinical outcomes between POCUS and Computed Tomography on a Cohort of 284 Cases and Review of Literature

Background:

The cervical spine is injured in approximately 3% of major trauma patients, and 10% of patients with serious head injury. Therefore, clearance of the cervical spine in multitrauma patients is a critically important task. This is particularly important, considering that there is a positive correlation between a Glasgow Coma Scale of <14 and cervical spine injury. Radiography is not sensitive enough to rule out cervical spine injury, especially as radiography done in the trauma setting is usually technically unsatisfactory.

Objective:

The current study aims to assess the diagnostic accuracy and prognostic significance of using bedside point-of-care ultrasound (POCUS) in traumatic cervical spine injuries compared to computed tomography (CT) as the reference standard.

Materials and Methods:

This comparative study enrolled 284 patients with severe multiple trauma at a tertiary care center between July 2017 and March 2020. The inclusion criteria included an indication of cervical spine CT scan, satisfaction of patients with participation in the study, and the lack of history of injury and severe traumatic events. The exclusion criteria were the history of a previous cervical spinal trauma, spondylosis, scoliosis, spinal tuberculosis, degenerative vertebral changes, and patients who refused to give consent to participate in research or CT scanning. The data were analyzed by SPSS software, and sensitivity, specificity, and positive predictive value (PPV)/negative predictive value (NPV) were determined based on CT findings.

Results:

The best window for the cervical spine was through the anterior triangle using the linear array probe (6–13 MHz). POCUS had a sensitivity of 78.5%, specificity of 98.4%, PPV of 93.2%, NPV of 92.8%, and accuracy of 93.2% in detecting all types of spinal injuries in comparison with CT scan as the standard modality. POCUS had a sensitivity of 100%, specificity of 92.3%, PPV of 62.3%, NPV of 100%, and accuracy of 91.7% in cases with the movement of injured particles. POCUS had a sensitivity of 32.2%, specificity of 100%, PPV of 100%, NPV of 91.4%, and accuracy of 90.8% in detecting the fracture of transverse process. POCUS had a sensitivity of 36.1%, specificity of 100%, PPV of 100%, NPV of 98.1%, and accuracy of 98.4% in ≤14-year age multitrauma patients. In comparison, the current study achieved a sensitivity of 79.4%, specificity of 95.7%, PPV of 92.1%, NPV of 86.3%, and accuracy of 88.6% in >14-year age multitrauma patients.

Conclusion:

POCUS for cervical spine is feasible using portable ultrasound machine and by neurosurgeons/radiologists/emergency physicians with basic training. It holds great potential in resource-starved settings and in unstable patients for ruling out unstable cervical spine injuries and injuries associated with the movement of fractured or dislocated particles. POCUS examination of the cervical spine was possible in the emergency setting and even in unstable patients and could be done without moving the neck. Future studies, ideally conducted as randomized control trials, are required to establish training and education standards, and to assess the feasibility and safety of POCUS as an alternative to radiography.

Management of pelvic injuries in hemodynamically unstable polytrauma patients - Challenges and current updates

Pelvic injuries are notorious for causing rapid exsanguination, and also due to concomitant injuries and complications, they have a relatively higher mortality rate. Management of pelvic fractures in hemodynamically unstable patients is a challenging task and has been variably approached. Over the years, various concepts have evolved, and different guidelines and protocols were established in regional trauma care centers based mainly on their previous experience, outcomes, and availability of resources. More recently, damage control resuscitation, pelvic angioembolization, and acute definitive internal fixation are being employed in the management of these unstable injuries, without clear consensus or guidelines. In this background, we have performed a computerized search using the Cochrane Database of Systematic Reviews, Scopus, Embase, Web of Science, and PubMed databases on studies published over the past 30 years. This comprehensive review aims to consolidate available literature on the current epidemiology, diagnostics, resuscitation, and management options of pelvic fractures in polytraumatized patients with hemodynamic instability with particular focus on damage control resuscitation, pelvic angioembolization, and acute definitive internal fixation.

Role of emergency physician–performed ultrasound in the differential diagnosis of abdominal pain

Objectives:

This prospective study was performed to evaluate the diagnostic accuracy of bedside point-of-care abdominal ultrasonography performed by emergency physician in patients with non-traumatic acute abdominal pain.

Methods:

The patients, who were admitted to emergency department due to abdominal pain, were included in this study. The emergency physician obtained a routine history, physical examination, blood draws, and ordered diagnostic imaging. After the initial clinical examinations, all the patients underwent ultrasonography for abdominal pathologies by emergency physician and radiologist, respectively. Point-of-care abdominal ultrasonography compared with abdominal ultrasonography performed by radiologist as the gold standard.

Results:

The study included 122 patients. Gallbladder and appendix pathologies were the most commonly detected in the abdominal ultrasonography. Compared with abdominal ultrasonography, point-of-care abdominal ultrasonography was found to have 89% sensitivity and 94% specificity in gallbladder pathologies; 91% sensitivity and 91% specificity in acute appendicitis; 79% sensitivity and 97% specificity in abdominal free fluid; 83% sensitivity and 96% specificity in ovarian pathologies. Compared to final diagnosis, preliminary diagnoses of emergency physicians were correct in 92 (75.4%) patients.

Conclusion:

This study showed that emergency physicians were successful in identifying abdominal organ pathologies with point-of-care abdominal ultrasonography after training.

Diagnostic accuracy of the Extended Focused Abdominal Sonography for Trauma (E-FAST) performed by emergency physicians compared to CT

INTRODUCTION

The diagnostic accuracy of the FAST exam performed by EM residents were shown to be similar to radiology residents. However, in the last 2 decades, an extended-FAST (E-FAST) protocol including thoracic examination to exclude pneumo- and hemothorax was introduced. The accuracy of emergency physicians (EPs) while performing E-FAST is a less studied area, especially in Europe. The aim of this study was to compare the diagnostic accuracy of the E-FAST exam performed by EM residents with the results of CT scan as a gold standard.

METHODS

This was a prospective, observational, diagnostic accuracy study conducted at the ED of a Level 1 Trauma Center. All consecutive adult multiple trauma patients were eligible, and any patient in whom thoraco-abdominal CT was ordered were recruited. Unstable and unavailable patients were excluded. E-FAST examination was performed by EPs as the index test, and CT examinations reported by a blinded academic radiology faculty was the gold standard.

RESULTS

A total of 140 patients were recruited from eligible 144 patients. The final study population was 132 for abdominal and 130 for thorax examinations. In this study, AUC of E-FAST was 0.71 for abdominal free fluid, 0.87 for pneumothorax and 1.00 for pleural effusion. The sensitivity was 42.9% and specificity was 98.4%. The +LR for abdominal free fluid was 26.8 and -LR was 0.58.

CONCLUSION

E-FAST examination has an excellent specificity. However, the sensitivity of the test is not high enough to rule-out thoraco-abdominal injuries in trauma patients when performed by EPs.

Should we perform a FAST exam in haemodynamically stable patients presenting after blunt abdominal injury: a retrospective cohort study

Background

Focussed Assessment with Sonography for Trauma (FAST) is a bedside ultrasonography technique used to detect free intraperitoneal fluid in patients presenting with blunt abdominal trauma (BAT) in the emergency department.

Methods

In this retrospective cohort study we investigated the potential of FAST as a risk stratification instrument in haemodynamically (HD) stable patients presenting after BAT by establishing the association between the FAST exam result and final outcome. An adverse outcome was defined in this context as the need for either a laparoscopy/laparotomy or an angiographic embolization or death due to abdominal injuries).

Results

A total of 421 patients with BAT were included, of which nine had an adverse outcome (2%). FAST was negative in 407 patients. Six of them turned out to have free intraperitoneal fluid (sensitivity 67 [41–86]%). FAST was positive in 14 patients, 12 of whom had free intraperitoneal fluid (specificity 99 [98–100]%). A positive FAST (positive likelihood ratio 34.3 [15.1–78.5]) was stronger associated with an adverse outcome than Injury Severity Score (ISS) or any individual clinical- or biochemical variables measured at presentation in the ED.

Discussion

The FAST exam can provide valuable prognostic information at minimal expenses during the early stages of resuscitation in haemodynamically stable patients presenting with BAT.

Conclusions

FAST exam should not be omitted in patients with BAT.

Electronic supplementary material

The online version of this article (doi:10.1186/s13049-016-0342-0) contains supplementary material, which is available to authorized users.