Active head motion reduction in Magnetic Resonance Imaging using tactile feedback

Abstract

Head motion is a common problem in clinical as well as empirical (functional) magnetic resonance imaging applications, as it can lead to severe artefacts that reduce image quality. The scanned individuals themselves, however, are often not aware of their head motion. The current study explored whether providing subjects with this information using tactile feedback would reduce their head motion and consequently improve image quality. In a single session that included six runs, 24 participants performed three different cognitive tasks: (a) passive viewing, (b) mental imagery, and (c) speeded responses. These tasks occurred in two different conditions: (a) with a strip of medical tape applied from one side of the magnetic resonance head coil, via the participant's forehead, to the other side, and (b) without the medical tape being applied. Results revealed that application of medical tape to the forehead of subjects to provide tactile feedback significantly reduced both translational as well as rotational head motion. While this effect did not differ between the three cognitive tasks, there was a negative quadratic relationship between head motion with and without feedback. That is, the more head motion a subject produced without feedback, the stronger the motion reduction given the feedback. In conclusion, the here tested method provides a simple and cost‐efficient way to reduce subjects' head motion, and might be especially beneficial when extensive head motion is expected a priori.

Custom-molded headcases have limited efficacy in reducing head motion during naturalistic fMRI experiments

“…Specifically, they observed that the efficacy of tactile feedback scaled with the amount of motion observed without tactile feedback, making this procedure particularly appealing when participants are likely to move a lot. Their findings also appear to be independent of behavioral tasks performed in the scanner and are efficacious even in situations when a participant is deliberately asked to move their head (Krause et al, 2019) . Another promising alternative is the use of inflatable head pillows (e.g.…”

“…The figures presented in Power et al (2019) support this notion as the largest motion reductions occurred for younger participants (7-14 years old). We speculate as to whether training procedures combined with tactile feedback might be similarly successful as participants can be taught to monitor their own movements (Krause et al, 2019) .…”

“…Audio was delivered using MR compatible in-ear headphones (Sensimetrics S14 https://www.sens.com/products/model-s14/ ). Participants were situated in the scanner with foam pillows and medical tape attached to their foreheads which provided tactile feedback (Krause et al, 2019) regarding head motion. Participants were also carefully instructed to lie as still as possible.…”

“…While some technical solutions like prospective acquisition correction (Thesen et al, 2000) have been used to correct motion in near real-time, more common solutions involve situating individuals within the scanner in a way that prevents movement of the head from occuring in the first place. Such techniques include the use of foam head-coil stabilizers and a bite-bar, thermoplastic masks over the nose and brow, general foam padding packed into the head-coil, visual feedback systems that allow participants to adjust their head position during a study, and tactile feedback using medical tape (Bettinardi et al, 1991;Fitzsimmons et al, 1997;Krause et al, 2019;Menon et al, 1997;Jonathan D. Power et al, 2019;Thulborn, 1999;Zaitsev et al, 2015) . Other techniques include training, mock-scanning, and using specialized movie stimuli to improve compliance in developmental and clinical populations (Vanderwal et al, 2019) .…”

Impact of Concatenating fMRI Data on Reliability for Functional Connectomics

“…While the need for more data per individual appears to be straightforward, how to achieve the desired amount of data is not. It is not always feasible to obtain more data due to practical limitations of a crowded scan protocol, or tolerability of the scanning environment (Krause et al, 2019;Meissner et al, 2020;Menon et al, 1997). Additionally, this solution does little to address questions about how to improve the amount of data available for analysis in existing data acquisitions.…”