Microwave-based stroke diagnosis making global pre-hospital thrombolytic treatment possible

Microwave-based stroke diagnosis making global pre-hospital thrombolytic treatment possible 2480 1293 IEEE Transactions on Biomedical Engineering (TBME)

Persson, M. ; Fhager, A. ; Trefna, H. ; Yu, Y. ; McKelvey, T. ; Pegenius, G. ; Karlsson, Jan-Erik ; Elam, M.
Chalmers University of Technology, SWEDEN, Volume 61, Issue 11, Page:2806-2817

In this paper we present a microwave-based system suitable for pre-hospital diagnosis of stroke (Please see a video). The motivation behind the work is to help the 15 million people who suffer a stroke each year to a better treatment.
Today the situation is that most patients are not receiving optimal treatment. This is because regional ischemia following thrombosis underlies most strokes, but a significant proportion is caused by haemorrhage calling for different treatment. Ischemic stroke patients may receive acute thrombolytic treatment, dramatically reducing or abolishing symptoms. However, such treatment must be initiated within 4.5h after stroke onset, preceded by safe differentiation of haemorrhagic vs. ischemic stroke since thrombolysis can severely aggravate a haemorrhagic condition. Today this requires brain scans, calling for hospitalization before deciding on treatment strategy. Consequently, a minority of patients that could benefit from thrombolytic treatment reach hospital and diagnosis within this time frame.  Here we present results from two initial clinical studies using two different brain diagnostic devices based on microwave technology. The method is based on scattering measurements with an antenna system worn on the head. Measurement data is analyzed with a machine-learning algorithm that is based on training using data from patients with a known condition. CT images were used as reference. The detection methodology is evaluated with the leave-one-out validation method combined with a Monte Carlo based bootstrap step. The results show that these devices and the associated algorithms can differentiate haemorrhagic from ischemic stroke patients. These findings set the stage for pre-hospital stroke diagnosis and treatment.

Keywords: Antenna measurements, Antennas, Biomedical measurement, Frequency measurement, Hemorrhaging, Microwave measurement, Microwave theory and techniques