Acoustic noise during MR scans can cause discomfort for both patients and technicians. Instead of trying to eliminate the noise from the scanner, we hypothesize that making the sounds pleasant would also increase patient comfort. In this study, we take advantage of the flexibility of Magnetic Resonance Fingerprinting (MRF)(1,2) to directly convert music to encoding gradients, which were then used with varying flip angles and TRs to quantify multiple tissue parameters simultaneously. This MRF-Music acquisition will significantly improve patients’ experience during MR scans while maintaining high scan efficiency and diagnostic quality. The main considerations of converting music to gradients were to 1) meet the hardware requirements of the scanner system, 2) preserve the sound of the music, 3) combine gradient design with basic imaging criteria such as sequence timing, image resolution and field of view, and 4) have sufficient sampling density to generate accurate maps. Figure 1 illustrates the flow chart of the music conversion. The audio waveform (Fig 1a) was first pre-processed with a conventional audio compression filter that constrains the waveform to a band of amplitude, and then low pass filtered to remove high frequency oscillation and resampled to match the gradient raster time(Fig 1b). We have observed that one can maintain sound quality while making relatively significant deviations from the actual music waveform as long as the basic timing of the music is preserved. In particular, it is important to preserve the inherent zero crossings of the music waveform. Therefore, acquisition blocks (TR) were defined in time by the zero crossings. These were then grouped such that each TR had a fixed number of music segments, each of sufficient duration that starts and ends at zero. Fig 1c shows an example of a SSFP-based sequence with four segments in each TR, with each segment being assigned to a specific encoding gradient. In Figure 2, half-sine waves were used to design slice selection gradients in order to make RF pulse design easy and robust. For the RF excitation, VERSE RF pulses(3,4) were designed to achieve the same slice profile for all TRs. A scaled music waveform formed the readout segment in Gx and a three-lobed balanced waveform was added in Gy to increase k-space coverage. The amplitudes of these waveforms were derived from Table 1. Figure 2b shows example k-space trajectories from ‘Bach Cello Suite No.1’ as performed by YoYo Ma. MRF-Music has also been extended to 3D acquisition with isotropic image resolution(5). The total conversion time for a 2D MRF-Music sequence was 2 minutes. All scans were acquired on a 3T scanner (Siemens Skyra). Acquisition parameters were: FOV of 300x300 mm2, image resolution of 1.2x1.2 mm2, slice thickness of 5 mm, and 4000 undersampled images were acquired in 68 seconds. The resultant time series of images were used to determine the values for T1, T2, and M0. A multi-scale reconstruction(6) was used to iteratively improve map quality. The experiments included phantom studies to estimate accuracy and scan efficiency, 2D and 3D in vivo studies to evaluate image quality. Additionally we performed a comfort level survey in 10 volunteers who haven’t had MRI before. These volunteers were asked to rated their comfort levels immediately after a one-minute acquisition from 1-10 (1=most uncomfortable and 10=most comfortable) for each of four different acquisitions (TSE, EPI, MRF-Music and no scan), presented in random order with 3 repetitions.The ANOVA analysis of the survey results showed a significant improvement of the comfort level from MRF-Music scan in comparison to both EPI and TSE, with an average scores of 7.2±1.2 from MRF-Music and 4.5±1.4 and 4.9±1.5 from EPI and TSE, respectively. Figure 3a,b compare T1 and T2 values of a phantom study from MRF-Music with those from MRF and standard spin-echo methods. The concordance correlation coefficients (CCC) indicated that both methods were in good agreement with the standard methods. Figure 3c,d compare the scan efficiency between MRF and MRF-Music. Both methods achieve high precision in a short acquisition time, with conventional MRF only outperforming MRF-Music by an average factor of 1.11 and 1.32 for T1 and T2, respectively. Figure 4 compares the maps obtained from MRF-Music with those from MRF scan. Both quantitative values and tissue appearance are in good agreement. This study demonstrated that MRF-Music can combine data acquisition with a musical sound to quantify multiple tissue parameters in a clinically acceptable time. Thus we believe that MRF-Music represents a broadly applicable, flexible framework to improve patient care in MRI.