MRI is the ideal imaging techniques for children, offering significant improvement in soft tissue contrast and lack of ionizing radiation compared to CT. However, the slow imaging speed and low performance of conventional motion compensation techniques require the use of deep sedation or general anesthesia, which can potentially have serious cognitive effects and makes the imaging study very complex. Our lab is developing motion-resistant MRI techniques for children using auto-navigated radial acquisition and deep learning reconstruction. Figure 4 shows an example of free-breathing imaging on an 8-year-old patient, where our motion-resistant MRI with 1-minute scan time improved image quality over the clinical standard that uses respiratory gating with 2-minute scan time. The reduction in scan time would reduce anesthesia depth or even remove anesthesia in pediatric MRI.