Cameron Ghafil
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Cameron Ghafil has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Kazuhide Matsushima
Division of Acute Care Surgery, Department of Surgery, University of Southern California, Los Angeles, USA
3 shared publications - Demetrios Demetriades
Division of Trauma and Surgical Critical Care, Department of Surgery, Los Angeles County and University of Southern California Medical Center, University of Southern California, Los Angeles, USA
2 shared publications - Kenji Inaba
University of Southern California
2 shared publications - Panagiotis Liasidis
Division of Trauma and Surgical Critical Care, Department of Surgery, Los Angeles County and University of Southern California Medical Center, University of Southern California, Los Angeles, USA
2 shared publications - Valerie Ponning Huang
Division of Trauma and Surgical Critical Care, Department of Surgery, Los Angeles County and University of Southern California Medical Center, Keck School of Medicine of the University of Southern California, Los Angeles, USA
2 shared publications - Anthony Nguyen
Daniel J. Epstein Department of Industrial and Systems Engineering University of Southern California Los Angeles CA USA
1 shared publication
Latest Publications
- Open Access
- 30 June 2023
- Published by Wiley in World Journal of Surgery
- Vol. 47 (10) , 2340-2346
Assessment of Machine Learning Methods to Predict Massive Blood Transfusion in Trauma
Background: Accurately predicting which patients are most likely to benefit from massive transfusion protocol (MTP) activation may help patients while saving blood products and limiting cost. The purpose of this study is to explore the use of modern machine learning (ML) methods to develop and validate a model that can accurately predict the need for massive blood transfusion (MBT).Methods: The institutional trauma registry was used to identify all trauma team activation cases between June 2015 and August 2019. We used an ML framework to explore multiple ML methods including logistic regression with forward and backward selection, logistic regression with lasso and ridge regularization, support vector machines (SVM), decision tree, random forest, naive Bayes, XGBoost, AdaBoost, and neural networks. Each model was then assessed using sensitivity, specificity, positive predictive value, and negative predictive value. Model performance was compared to that of existing scores including the Assessment of Blood Consumption (ABC) and the Revised Assessment of Bleeding and Transfusion (RABT).Results: A total of 2438 patients were included in the study, with 4.9% receiving MBT. All models besides decision tree and SVM attained an area under the curve (AUC) of above 0.75 (range: 0.75–0.83). Most of the ML models have higher sensitivity (0.55–0.83) than the ABC and RABT score (0.36 and 0.55, respectively) while maintaining comparable specificity (0.75–0.81; ABC 0.80 and RABT 0.83).Conclusions: Our ML models performed better than existing scores. Implementing an ML model in mobile computing devices or electronic health record has the potential to improve the usability.
- 30 June 2023
- Published by Wiley in World Journal of Surgery
- Vol. 47 (10) , 2340-2346
Open AccessAssessment of Machine Learning Methods to Predict Massive Blood Transfusion in Trauma
Background: Accurately predicting which patients are most likely to benefit from massive transfusion protocol (MTP) activation may help patients while saving blood products and limiting cost. The purpose of this study is to explore the use of modern machine learning (ML) methods to develop and validate a model that can accurately predict the need for massive blood transfusion (MBT).Methods: The institutional trauma registry was used to identify all trauma team activation cases between June 2015 and August 2019. We used an ML framework to explore multiple ML methods including logistic regression with forward and backward selection, logistic regression with lasso and ridge regularization, support vector machines (SVM), decision tree, random forest, naive Bayes, XGBoost, AdaBoost, and neural networks. Each model was then assessed using sensitivity, specificity, positive predictive value, and negative predictive value. Model performance was compared to that of existing scores including the Assessment of Blood Consumption (ABC) and the Revised Assessment of Bleeding and Transfusion (RABT).Results: A total of 2438 patients were included in the study, with 4.9% receiving MBT. All models besides decision tree and SVM attained an area under the curve (AUC) of above 0.75 (range: 0.75–0.83). Most of the ML models have higher sensitivity (0.55–0.83) than the ABC and RABT score (0.36 and 0.55, respectively) while maintaining comparable specificity (0.75–0.81; ABC 0.80 and RABT 0.83).Conclusions: Our ML models performed better than existing scores. Implementing an ML model in mobile computing devices or electronic health record has the potential to improve the usability.
- 20 October 2022
- Published by Springer Nature in European Journal of Trauma and Emergency Surgery
- Vol. 49 (2) , 813-823
Falls from scaffolds: a nationwide analysis
Introduction Falls from scaffolds are a common cause of occupational injuries and fatalities, however, the data concerning these injuries are scarce. The purpose of this study was to characterize the epidemiology, injury patterns, and outcomes of falls from scaffolds. Methods Retrospective cohort study using data from the National Trauma Data Bank (2007–2017). All adult patients who sustained injuries following falls from scaffolds and as a comparator, falls from ladders, were identified using the external cause of injury codes. The primary outcome was differences in the type and severity of injuries between the groups. Results A total of 183,853 patients were analyzed. Of these, 18,189 (9.9%) fell from scaffolds and 165,664 (90.1%) from ladders. Patients who fell from scaffolds were younger, more often male, and more likely to sustain severe trauma (ISS > 15: 24.1% vs 17.3%, p < 0.001). Falls from scaffolds resulted in more severe head injuries (head AIS ≥ 3: 18.6% vs 14.7%, p < 0.001) and more spinal fractures (30.3% vs 25.2%, p < 0.001). Falls from scaffolds were associated with higher mortality (2.5% vs 1.8%, p < 0.001), higher ICU admission rate (25.0% vs 21.5%, p < 0.001), and longer hospitalization. On multivariable analysis, the strongest predictors of mortality were GCS < 9 and hypotension on admission, severe (AIS ≥ 3) head injury, and age > 65 years. Conclusion Falls from scaffolds are associated with more severe injuries and worse outcomes compared to ladder falls. Males in the fourth decade of life were disproportionally affected. Further research on fall prevention is warranted to decrease this important cause of death and disability.
- 20 October 2022
- Published by Springer Nature in European Journal of Trauma and Emergency Surgery
- Vol. 49 (2) , 813-823
Falls from scaffolds: a nationwide analysis
Introduction Falls from scaffolds are a common cause of occupational injuries and fatalities, however, the data concerning these injuries are scarce. The purpose of this study was to characterize the epidemiology, injury patterns, and outcomes of falls from scaffolds. Methods Retrospective cohort study using data from the National Trauma Data Bank (2007–2017). All adult patients who sustained injuries following falls from scaffolds and as a comparator, falls from ladders, were identified using the external cause of injury codes. The primary outcome was differences in the type and severity of injuries between the groups. Results A total of 183,853 patients were analyzed. Of these, 18,189 (9.9%) fell from scaffolds and 165,664 (90.1%) from ladders. Patients who fell from scaffolds were younger, more often male, and more likely to sustain severe trauma (ISS > 15: 24.1% vs 17.3%, p < 0.001). Falls from scaffolds resulted in more severe head injuries (head AIS ≥ 3: 18.6% vs 14.7%, p < 0.001) and more spinal fractures (30.3% vs 25.2%, p < 0.001). Falls from scaffolds were associated with higher mortality (2.5% vs 1.8%, p < 0.001), higher ICU admission rate (25.0% vs 21.5%, p < 0.001), and longer hospitalization. On multivariable analysis, the strongest predictors of mortality were GCS < 9 and hypotension on admission, severe (AIS ≥ 3) head injury, and age > 65 years. Conclusion Falls from scaffolds are associated with more severe injuries and worse outcomes compared to ladder falls. Males in the fourth decade of life were disproportionally affected. Further research on fall prevention is warranted to decrease this important cause of death and disability.
- 15 August 2022
- Published by Elsevier in Journal of Gastrointestinal Surgery
- Vol. 27 (2) , 411-412
Small Bowel Obstruction in the Virgin Abdomen
- 15 August 2022
- Published by Elsevier in Journal of Gastrointestinal Surgery
- Vol. 27 (2) , 411-412
Small Bowel Obstruction in the Virgin Abdomen
- 22 September 2021
- Published by Wiley in World Journal of Surgery
- Vol. 46 (1) , 91-97
Performance of Focused Assessment with Sonography for Trauma Following Resuscitative Thoracotomy for Traumatic Cardiac Arrest
Background: In patients undergoing resuscitative thoracotomy (RT) for traumatic cardiac arrest, focused assessment with sonography for trauma (FAST) is often used to look for intraperitoneal fluid. These findings can help determine whether abdominal exploration is warranted once return of spontaneous circulation is achieved; however, the diagnostic accuracy of FAST in this clinical scenario has yet to be evaluated. The purpose of this study was to assess the performance of FAST in identifying intra‐abdominal hemorrhage following RT.Methods: We performed a 3‐year retrospective study at a high‐volume level 1 trauma center from 2014 to 2016. We included patients who underwent RT in the Emergency Department. All FAST examinations were performed by non‐radiologists. Operative findings, computed tomography reports, diagnostic peritoneal aspirate (DPA) results, and autopsy findings were used as reference standards to calculate the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of the FAST.Results: A total of 158 patients met our inclusion criteria. The median age was 35 years (interquartile range [IQR]: 23–53), 86.1% were male, and 60.1% sustained blunt trauma. Most patients suffered severe injuries with a median injury severity score of 27 (IQR: 18–38). The sensitivity, specificity, PPV, NPV, and accuracy of FAST for identifying intra‐abdominal hemorrhage were 66.0%, 84.8%, 68.6%, 83.2%, and 78.5%, respectively. Among the 107 patients with a negative FAST, 22 (20.6%) underwent DPA, which was positive in 5 patients.Conclusions: FAST can be utilized in the diagnostic workup of trauma patients after RT. In patients with a positive FAST, exploratory laparotomy is warranted, whereas other diagnostic adjuncts such as DPA or mandatory abdominal exploration may be considered in patients with a negative FAST.
- 22 September 2021
- Published by Wiley in World Journal of Surgery
- Vol. 46 (1) , 91-97
Performance of Focused Assessment with Sonography for Trauma Following Resuscitative Thoracotomy for Traumatic Cardiac Arrest
Background: In patients undergoing resuscitative thoracotomy (RT) for traumatic cardiac arrest, focused assessment with sonography for trauma (FAST) is often used to look for intraperitoneal fluid. These findings can help determine whether abdominal exploration is warranted once return of spontaneous circulation is achieved; however, the diagnostic accuracy of FAST in this clinical scenario has yet to be evaluated. The purpose of this study was to assess the performance of FAST in identifying intra‐abdominal hemorrhage following RT.Methods: We performed a 3‐year retrospective study at a high‐volume level 1 trauma center from 2014 to 2016. We included patients who underwent RT in the Emergency Department. All FAST examinations were performed by non‐radiologists. Operative findings, computed tomography reports, diagnostic peritoneal aspirate (DPA) results, and autopsy findings were used as reference standards to calculate the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of the FAST.Results: A total of 158 patients met our inclusion criteria. The median age was 35 years (interquartile range [IQR]: 23–53), 86.1% were male, and 60.1% sustained blunt trauma. Most patients suffered severe injuries with a median injury severity score of 27 (IQR: 18–38). The sensitivity, specificity, PPV, NPV, and accuracy of FAST for identifying intra‐abdominal hemorrhage were 66.0%, 84.8%, 68.6%, 83.2%, and 78.5%, respectively. Among the 107 patients with a negative FAST, 22 (20.6%) underwent DPA, which was positive in 5 patients.Conclusions: FAST can be utilized in the diagnostic workup of trauma patients after RT. In patients with a positive FAST, exploratory laparotomy is warranted, whereas other diagnostic adjuncts such as DPA or mandatory abdominal exploration may be considered in patients with a negative FAST.