41 New methods for diagnosis of thoracic trauma in prehospital care
2017
Aim Trauma to the thorax is common and can be life-threatening. Prehospital diagnosis of thoracic injuries is challenging. Ultrasound is a promising technology; however, its accuracy is operator dependent. Methods not requiring operator image interpretation would be beneficial. The aim of this study is to evaluate electrical bioimpedance (EBI) and microwave technology (MWT). Both technologies are non-operator dependent, non-invasive, harmless, cost efficient, rapid and portable. Methods Two complementary lines of research are pursued. A clinical study aiming to differentiate EBI measurements of thoracic trauma patients (n=20) and healthy controls (n=20), using diagnostic mathematical algorithms, has been completed. Clinical trials are complemented by experiments on realistic porcine models of pneumothorax (PTX) and hemothorax (HTX). 1–2 These experiments enable analysis of EBI and MWT with well-defined injuries. A pilot study on two pigs with unilateral PTX from small (50 mL) to large (2000 mL) sizes, and large HTX, was performed. Diagnostic performance is evaluated using cross-validation to derive the area under the ROC curve (AUC), and confusion matrices. Results The clinical study achieved AUC=0.87. The pilot porcine study showed that EBI parameters evolved as expected with increasing PTX/HTX; EBI theory predicts presence of air should increase resistivity and fluid decrease it. The MWT classification accuracy for predicting size of PTX was 100% and 98% for each pig, respectively. Conclusion EBI and MW are promising technologies for prehospital diagnosis of thoracic injuries. References Buendia, et al. Electrical bioimpedance for diagnosing thoracic injuries: Test on a porcine pneumothorax and hemothorax model. (2016) Short Communications From AAAM’s 60th Annual Scientific Conference, Traffic Injury Prevention, 17:supp 1, 175–218. Oveland, et al.: A wearable microwave detector for diagnosing thoracic injuries-test on a porcine pneumothorax model. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine2015;23(Suppl 2):A20. Conflict of interest None declared. Funding The authors acknowledge funding received from the Swedish research centre MedTech West and Folksam insurance company.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
0
References
0
Citations
NaN
KQI