1. Setup study method for super high resolution functional and anatomical imaging in cerebellum area. 2. Acquiring seed based fcMRI using seed from pure sensory stimulation in cerebellum. 3. Studying the whole brain scale functional network with cerebellum and medulla involvement in sensory/motor system.Male Sprague-Dawley rats weighing around 400 g was used in this study. The ear tubing procedure was performed on both ears, and Fomblin Y was filled into the external and middle ear. This middle ear Fomblin filling (MEFF) technique greatly reduces susceptibility induced artifact1. The MEFF procedure was performed under 2% isoflurane inhalation anesthesia. A 9.4 Tesla MRI system with a 31 cm horizontal bore was used for scanning. A Bruker linear transmit-coil was used with the center of the coil located 3 mm anterior to the external ear canal. Signal acquisition was achieved using our self-designed 15 mm receive coil with low-noise-amplifier (LNA) on board. The coupling circuit uses an American Technical Ceramics 800R series, high-Q, non-magnetic capacitor, a Rogers RT/duroid 5880 low-loss circuit board with 1-ounce copper trace on both sides, and a WanTCom WMA9RA LNA, with an input impedance of 1.5 Ohms and an overall gain of 28 dB. A partial k-space sequence with 20 over-scan lines was used for this high resolution EPI acquisition. TE=10.78 ms, TR=2 s, FOV=28.8 mm, slice thickness=0.3 mm, matrix size=96×96. For seed based fcMRI analysis, the seeds were chosen from fMRI activation of forepaw stimulation. Data analysis was performed using AFNI. The seed region was chosen from the cerebellum during forepaw A band-pass filter was used for all resting-state EPI acquisitions with a low-pass filter of 0.1 Hz and a high-pass filter of 0.01 Hz on a voxel-by-voxel basis covering the entire brain. Results were smoothed with 0.3 mm FWHM.Fig. 1 shows the results of anatomical imaging and EPI image acquired with our surface coil comparing to the atlas. It is clear that the anatomical structures can be clearly identified at this resolution with our self-designed receive coil. Due to MEFF procedure, the coverage of our EPI image is highly comparable to the anatomical image and atlas. There is no significant signal dropout in the results. Fig. 2 shows an example of fcMRI result in cerebellum and medulla area with seed chosen from forepaw stimulation. In the cerebellum region, two groups of neurons involve in the sensory network. These two groups of neurons are located in Crus 1 lobule. More prominent sensory networks are observed in medulla area. Spinaocerebellar tracts (ventral) are constantly demonstrated across slices. While Oliveocerebellar tract and intermediate reticular nucleus (IRt) are also visualized in Figure 2C.One of the major hurdle for animal true whole brain scale functional imaging is that the reduced SNR in deep area due to various reasons especially under high resolution. This study proves that with the combination of MEFF method and a self-designed single loop receive coil, one can get great anatomical and EPI images from the cerebellum and medullar region. The SNR remains very high for functional analysis despite that we used 200 micron cubic voxel for imaging. The depth sensitivity of the 15mm coil is good enough to cover the entire cerebellum and medulla. The image quality can be further improved by applying a bigger coil with similar configuration. Neurons in cerebellum react to somatosensory input. This has been proved by previous studies2,3. Due to the ultra-high resolution, especially in the Z direction, we are able to isolate individual neuronal groups that are related to sensory input. Special attention will be put to the IRt area. This area is closely related to noxious sensation and is also involved in pain. The function of cerebellum is very complicated since it is a key relay point for both sensory and motor network. These two networks can work independently or together. It becomes a much complicated task to further isolated each network and study their interaction. From the current result we got, we observed localized functional network within Crus I and paramedian lobule (Fig.2A, B), and inter-hemispherical connectivity between spinocerebellar tract and olivocerebellar tract of both sides (Fig.2C). With our powerful MEFF technique and coil development, we are trying to isolate each neuronal group and study the resting state functional connectivity network they involved at a true whole brain scale.