Artificial intelligence CAD tools in trauma imaging: a scoping review from the American Society of Emergency Radiology (ASER) AI/ML Expert Panel

Dreizin, David; Staziaki, Pedro V.; Khatri, Garvit D.; Beckmann, Nicholas M.; Feng, Zhaoyong; Liang, Yuanyuan; DelProposto, Zachary; Klug, Maximiliano; Spann, J Stephen; Sarkar, Nathan; Fu, Yunting

doi:10.1007/s10140-023-02120-1

Cited by 16 publications

(10 citation statements)

References 103 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Radiology is a forward-thinking eld that has consistently embraced advancements in digital technology, leveraging data-driven approaches to remain at the forefront of innovation. RTAT has served as a means of objectively benchmarking improvements in radiologist e ciency with implementation of technologies including PACS, voice recognition, and more recently, scalable deep learning-based AI CAD solutions [17,18,20]. AI CAD tools may help facilitate a faster work ow in patients with known or suspected polytrauma.…”

Section: Discussionmentioning

confidence: 99%

“…Ultimately, clinical use and reimbursement by CMS will be based on studies showing improvement in patient outcomes, however outcome studies supporting FDA-approved tools are currently few in number [20]. Since RTAT is routinely collected by radiology departments, this objective parameter can be used as a preliminary indicator of potential e cacy before undertaking time, resource, and capital-intensive clinical trials and prospective outcome studies.…”

Section: Discussionmentioning

confidence: 99%

“…It continues to be used as a quality metric of diagnostic e ciency by academic radiology departments and there is pressure from referring physicians to meet faster RTATs [19]. In recent times, improved diagnostic e ciency has been an oft-cited impetus for development of diagnostic imaging CAD tools [20][21][22], and RTAT has been used as a parameter to determine improvements in e ciency related to the implementation of scalable deep learning-based computer aided detection (CAD) tools. Davis et al, reported a decrease in RTAT with the use of an intracranial hemorrhage (ICH) CAD early noti cation tool [23].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Does Acuity and Severity of Injury Affect Trauma Whole-Body CT Report Turnaround Time? A Large-scale Study

Sarkar,

Khedr,

Dreizin

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Trauma is a leading cause of death in the United States. Whole-Body CT is routinely used to screen for major injuries and hemorrhage and to plan interventions. Report Turn-Around-Time (RTAT) plays a role in the flow of patient management and is an objective marker of efficiency that could be improved upon with development and implementation of AI CAD tools for trauma patients. Currently, the degree to which this metric is affected by injury acuity and severity is poorly documented. The study included 11,251 patients who underwent admission trauma whole-body CT at two trauma centers within a single hospital system; a level I trauma center (n=9043) and a regional urban level II trauma center (n=2208). Patients were selected between July 2016 and September 2022. Clinical data and RTAT were extracted from the trauma registry and medical records. Overall median RTAT was 24 minutes (IQR: 4-48). RTAT was significantly longer in those who had hemodynamic shock (37 minutes vs 22 minutes, p<0.0001), polytrauma (ISS≥16) (34 minutes vs 21 minutes, p<0.0001), or massive transfusion (47.5 minutes vs 24 minutes, p<0.0001); those who expired (41 minutes vs 23 minutes,p<0.0001); the older cohort (age>55) (28 minutes vs 22 minutes, p<0.0001), and those with penetrating vs blunt injuries (27 minutes vs 23 minutes, p=0.001). Overall, these findings indicate that increased traumatic injury severity is associated with non-negligible increases in whole-body CT RTATs. The results provide strong justification for computer-aided detection/diagnosis (CAD) research and development to reduce cognitive load and augment diagnostic efficiency in severely injured patients who stand to benefit most from such tools.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Does Acuity and Severity of Injury Affect Trauma Whole-Body CT Report Turnaround Time? A Large-scale Study

Sarkar,

Khedr,

Dreizin

2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…As we enter an era characterized by collaboration between human expertise and computational power, DL algorithms hold the potential to revolutionize future medical practices, particularly by alleviating the workload of healthcare providers in emergency settings [ 24 ]. Despite these advancements, the availability of trauma-related algorithms to assist trauma surgeons in managing time-sensitive and life-threatening injuries is still limited [ 25 – 28 ]. Moreover, there is a clinical need for an explainable and transparent AI model to support emergency radiologists and clinicians, a need that remains unaddressed [ 29 ].…”

Section: Introductionmentioning

confidence: 99%

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

Cheng,

Lin,

Hsu

et al. 2024

J Digit Imaging. Inform. med.

View full text Add to dashboard Cite

Computed 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.

show abstract

“…These digital platforms allow for the rapid acquisition, storage, retrieval, and sharing of radiological images, facilitating remote interpretation and collaboration among healthcare providers [12]. Additionally, the integration of artificial intelligence algorithms and computer-aided detection systems has shown promise in assisting residents with image interpretation, reducing interpretation errors, and improving diagnostic accuracy [13][14][15]. This study aims to assess the competency level of EM residents in interpreting hand X-rays across three major regions in Saudi Arabia.…”

Section: Introductionmentioning

confidence: 99%

The Competency of Emergency Medicine Residents in Interpreting Hand X-rays Across the Three Major Regions of Saudi Arabia

Alsulimani,

AlRasheed,

Saeed

et al. 2024

Cureus

View full text Add to dashboard Cite

Background Interpreting hand X-rays is crucial for emergency medicine residents to accurately diagnose traumatic injuries and conditions affecting the hand. This study aimed to assess the competency of emergency medicine residents in interpreting hand X-rays across three major regions in Saudi Arabia. Methodology We conducted a cross-sectional study involving 100 emergency medicine residents from the Central, Eastern, and Western regions of Saudi Arabia. Participants were presented with 10 clinical case scenarios each accompanied by hand X-rays and were asked to provide their interpretations. Assessment scores were calculated based on the proportion of correct answers for each case. Results Half of the participants (50 residents) fell within the age range of 25 to 27 years, with 61 male and 39 female participants, respectively. Residents in the third year of training (R3) exhibited the highest mean score of 74.83% ± 20.46%. Participants using desktops to view the images achieved the highest mean score of 75% ± 10.49% compared to those using smartphones or tablets. Significant associations were found between age (F = 4.072, p = 0.020), training level (F = 3.161, p = 0.028), and choice of viewing device (F = 7.811, p = 0.001) and assessment scores. Conclusions Our study highlighted that emergency medicine residents in Saudi Arabia demonstrate competent proficiency in interpreting hand X-rays, with higher competency observed among senior residents (R3 and R4), those aged 28 to 30 years, and those using desktops for image viewing.

show abstract

Artificial intelligence CAD tools in trauma imaging: a scoping review from the American Society of Emergency Radiology (ASER) AI/ML Expert Panel

Cited by 16 publications

References 103 publications

Does Acuity and Severity of Injury Affect Trauma Whole-Body CT Report Turnaround Time? A Large-scale Study

Does Acuity and Severity of Injury Affect Trauma Whole-Body CT Report Turnaround Time? A Large-scale Study

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

The Competency of Emergency Medicine Residents in Interpreting Hand X-rays Across the Three Major Regions of Saudi Arabia

Contact Info

Product

Resources

About