PROJECT SUMMARY Stroke is the leading cause of disability, the second leading cause of dementia, and the second leading cause of death worldwide, affecting nearly 800,000 patients per annum in the U.S. alone. Cerebrovascular ischemia is rapidly potentiated by rising body temperatures, occurring even with mild and sub-febrile hyperthermia. The potency of hyperthermic damage in ischemic brain arises due to impaired cooling in hypoperfused brain tissues. Consequently, localized brain hyperthermia can precede, and commonly exceeds, systemic temperature changes following ischemic injury, and is well-recognized as accelerating stroke progression and consumption of at-risk tissues. Despite severe worsening of clinical outcomes among febrile stroke patients, the development of prognostic biomarkers and robust treatment targets in peri-ischemic cerebral hyperthermia remains unrealized due to lack of pragmatic means to acquire accurate spatially-resolved cerebral thermographs. MR thermometry (MRT) using the proton resonance frequency (PRF) chemical shift with 1H-MR spectroscopy (MRS) has been explored for non-invasive cerebral thermography. Despite promising results, several fundamental limitations undermine estimation of absolute temperature and thus impede development of meaningful clinical paradigms, including: 1) the local environment in tissues can significantly alter performance and affect temperature estimation; 2) magnetic field fluctuations caused by physiological- and hardware-related noise introduce large errors in temperature estimation; and 3) low temperature sensitivity and signal-to-noise of spectroscopic imaging approaches hinder in vivo accuracy. This application addresses such challenges to fulfill the elusive aim of absolute MR thermometry using novel multinuclear thermometry developed by the investigators. Initial results suggest significantly heightened immunity to B0 drift, susceptibility, and pH, and higher SNR per-unit time, relative to MRS approache