Head motion during imaging remains a major hurdle in brain imaging. Problems associated with motion can be severe particularly among pediatric patients, stroke patients and in the elderly patients ^[1]. Motion during imaging translates to phase errors in k-space which results in ghosting or blurring in the reconstructed images. Although several motion correction and prevention techniques have been explored with different level of success, very few techniques that provide prospective correction techniques have resulted with limited applications ^[2]. Here we present a prospective motion correction (PMC) technique using a miniature passive magnetic sensor. The motion sensor can virtually work with any imaging technique that is sensitive to motion. Tracking sensor: The tracking sensor (Robin Medical, Inc. Baltimore, USA) (Fig 1) consists of six coil loops which are placed in pairs orthogonal to each other. They are designed to pick up the voltages induced by the tracking gradients magnetic fields generated by static bipolar gradients embedded in the imaging sequence.The induced voltages into these coils are compared with pre-recorded calibration chart of voltage map of the MR space which reports the position and orientation of the sensor in the magnet. MRI methods: MRI scans were performed using 3T TIM TRIO (Siemens Healthcare, Erlangen, Germany) equipped with a 32 channel head coil (Siemens). A series of high resolution gradient echo (GRE) axial slice brain scans were acquired from a normal healthy volunteer. The GRE sequence was initially modified by adding three bipolar gradients (total duration = 1.5 ms) along the read, phase, and slice-selection directions respectively (Fig 2). A dynamic feedback loop mechanism was implemented into the sequence to receive the sensor position and orientation for real-time update of the imaging slice using an in-house developed application [3]. Phase correction is then performed based on the motion information before reconstructing the k-space data. To test the accuracy of our PMC technique, we conducted three experiments. In the first experiment, the subject with the tracking sensor taped on the left side of his head was asked to remain as still as possible to acquire the base line images which were used as reference. In the second experiment, the subject was instructed to move during the scan whenever prompted by the MRI technician while the PMC was turned off. The third experiment was same as the second, except now the PMC was turned on. In all three cases the sensor’s relative position and orientation were recorded in time. The data consisted of 30 time frames of a single 5 mm axial slice. The image matrix was 256x256 pixels with a resolution of 1.4x1.4 mm2. The acquisition parameters were: TR/TE = 10.7/3.19 ms, α = 15o, bandwidth = 330 Hz/pixel.The standard deviations of the position and orientation coordinates are listed in Table 1 and demonstrate very little variation in position and orientation coordinates while the subject head is still. Fig 3 shows five time frames of the axial images of the subject brain: a) with no motion b) with PMC off and c) with PMC on. To estimate the error from motion, we subtracted images of time series from the first image in each of the above two time series b) and c). Row (c) shows the difference images series with PMC turned off and Row (d) shows the difference images with PMC turned on. Comparison of images in d) and e) shows that motion artifacts are significantly reduced when the PMC is turned on. However images in e) show some residual errors which can be caused by the field distortions during motion and the gradient nonlinearities which resulted in images misalignment with those acquired at rest.We demonstrate that our PMC method in brain imaging using a passive magnetic sensor is capable of tracking the patient head motion with high accuracy. Our results suggest that our PMC technique can correct rigid body motion that is normally encountered in routine clinical scanning. The small footprint of the sensor makes it an ideal tool for tracking motion and can be positioned on the subject at a location out of line of sight unlike optical tracking systems [4]. Furthermore, the method is compatible with virtually any imaging sequence with long acquisition time. Especially techniques such as fMRI, DTI and spectroscopy sequences are likely to benefit greatly when dealing with non-cooperative subjects. Further investigation is necessary to reduce the errors that result from field distortions and to understand the extent of motion that can be effectively corrected by our PMC technique.