Kleine-Levin Syndrome (KLS) is a rare neurological disorder characterized by recurrent episodes of excessive sleepiness and other symptoms listed in the ICSD Diagnostic Criteria for KLS. To date, the etiology of KLS is still unknown. The most consistent finding in KLS is abnormal thalamic function. Thalamic abnormality was reported from task fMRI of working memory as well as from SPECT studies. During working memory tasks, KLS patients consistently show hyperactivation in the left thalamus as compared to healthy controls. They also show less activation in the anterior cingulate cortex¹ (ACC). Besides, one research suggests that thalamus may represent the functional interface between the arousal and the attentional systems². Another research further relates such attention to ACC³. Therefore, the purpose of this study is to examine functional connectivity of the thalamus in resting state fMRI. Specifically, we are interested in functional connectivity between the thalamus and ACC.One male patient and one female patient with KLS were recruited in the study. They were brother (28 years old, disease duration = 11 years) and sister (30 years old, disease duration = 13 years). Aside from hypersomnia, the male patient’s attention and reading ability were also compromised. The female patient was found to have decreased cognitive ability, working memory and attention. Both patients received MRI scans in aggravated and sober conditions. The aggravated condition was considered when patients’ sleep exceeded 15 hours per day. The sober condition was considered when patients resumed normal sleep duration. For comparison, ten healthy people (mean age = 25.2 ± 7.2 years) were also recruited to the study. All participants received MRI scans on a 3 Tesla MRI system (Tim Trio, Siemens, Erlangen, Germany) with a 32-channel phased array head coil. T1-weighted imaging was performed using a 3D MPRAGE sequence: TR / TE = 2000 ms / 3 ms, flip angle = 9°, FOV = 256 × 192 × 208 mm^3, acquisition matrix = 256 × 192 × 208. Multi-echo resting state fMRI was performed using a 2D gradient echo EPI sequence⁴: TR / TE = 2550 ms / 12, 28, 44, 60 ms, flip angle = 90°, FOV = 240 × 240 mm^2, slice thickness = 2.5 mm, and acquisition matrix = 64 × 64. The anatomical image was first skull-stripped and then warped nonlinearly to MNI anatomical template using FSL FNIRT. Denoising was applied after optimal combination of echoes and regressing out motion and high-frequency signals using AFNI tool 3dbandpass. Regression models for motion artifact included rigid body parameters for the alignment of all fMRI volumes to a reference volume. Before regression, data were despiked with a tanh function. After preprocessing, the data were decomposed with FastICA to remove non- BOLD components. Then, we extracted time courses to obtain seed region, nuisance regressors using spm8. The bilateral thalamus seeding ROI was generated using WFU_PickAtlas^5-7. All these regressors were organized into a design matrix of a general linear model^8,9. Using this design matrix, we computed a contrast image that represented signals coherent with seed region signal. Figure 1a shows the contrast of bilateral thalamic seeding in the male patient. The activation of ACC is decreased as compared to normal controls. In addition, the activation in bilateral insula (t-value = 3.78) is increased. These results indicate that the connection between the thalamus and ACC and the connection between the thalamus and insula have been altered in opposite directions. Figure 1b shows the contrast of bilateral thalamic seeding in the female patient. The activation of ACC is decreased as compared to normal controls. The result indicates that the connection between the thalamus and ACC is probably compromised. Figure 2 shows the contrast of bilateral thalamic seeding between normal control and the two patients in sagittal view. Both patients show less activation in ACC compared to normal control.In our patients with KLS, we found that their connections between the thalamus and ACC were decreased compared to normal controls. This malfunctioning connection might explain their symptoms of hypersomnia. Incidentally, we found hyperactivation in the thalamic-insular connection in the male patient. It has been reported that insula plays a role between sleep and conscious¹⁰. However, we found insular hyperactivation only in male patient, but no obvious activation in female patient. This might result from the heterogeneity of the disease.The future work will involve investigation of the mechanism of the decreased connection between ACC and thalamus. The unique increase in the connection between the thalamus and insula in our male patient also warrants further investigation.