Targeted temperature management is a treatment that seeks to reduce and control the body temperature in order to improve patient outcome and is successfully used for a variety of afflictions¹. However, no consensus has been reached on the best and most cost-effective cooling technique². We demonstrate here the feasibility of a novel cooling technique using high flow cold air applied nasally and orally to the airways by monitoring the brain temperature change before, during, and after cooling. 2 healthy male volunteers (age 29 and 24) were investigated. One volunteer was scanned two times (experiment 1 and 2) and the other one time (experiment 3). Cooling was administrated using a hollow mask covering the mouth and nose of the subject, but allowing the high flow (>350 L/min) cold air (-14°C) to freely escape into the surrounding during exhalation. The resulting cold air blow into the volunteers' faces caused additional cooling. The brain temperature was monitored using gradient echo phase imaging on a 3T MRI system (SIEMENS, Erlangen Germany). The first 6 images were acquired without cooling to establish a baseline, the next 10 images acquired during cooling, and the last 9 images without cooling in order to see a return to the baseline. All subjects gave written informed consent and the measurements were approved by the local ethics committee. To correct for phase drift a baseline curve was established for each pixel using a 2-degree polynomial fit between the first 6 baseline images and the last 6 images, when temperature was assumed to have returned to normal. The temperature change was then calculated using the phase shift from the baseline³:$$∆T= ∆φ⁄(γαB_0 TE)$$with the phase difference between each pixel and its baseline curve, γ the gyromagnetic ratio, α the proton resonance frequency change coefficient, B0 the magnetic field strength, and TE the echo time. All imaging was done using a gradient echo sequence with FA = 10°, TE/TR = 4/44 ms, matrix = 128*128*32, FOV = 260*260*112 mm, and 25 acquisitions in 25 minutes. Cooling was accomplished using a cold air compressor (MECOTEC croyair MINI, Elmshorn Germany) delivering cold air measured at -14° C and a flow higher than 350 l/min⁴. The brain tissue was automatically segmented using thresholding and then further divided into four general anatomical regions: the inferior frontal lobe, superior frontal lobe, temporal and occipital lobe, and parietal lobe. A significant temperature reduction (Wilcoxon rank-sum, p<0.05) during cooling was seen in the inferior frontal lobe in all 3 experiments, with an average reduction of -0.33°C (Figure 1). The three other anatomical regions showed on average no significant changes from baseline temperature (Figure 2). The location of the observed temperature reduction is consistent with the nasal airways, indicating that the cooling effect occurred through upper airway exposure and not by reducing the temperature of the blood in the arteries in the neck. It is therefore still unclear if whole-brain cooling is possible using the proposed method. We have demonstrated, using MRI temperature measurements, that localized cooling of the brain is possible using a simple high flow cold air system. Significant temperature changes were observed within minutes of cooling, pointing to a potential use by first responders for rapid cooling in emergency situations.