Abstract The most common malignancy for men in the western world is Prostate cancer (PCa), with a predicted 288,300 new cases and 34,700 specific deaths in 2023. With recent developments in comprehensive screening and biopsy strategies, the ten-year survival rate of PCa is largely improved to 98% and the disease specific mortality is reduced to only 4-8%. However, the widespread adoption of these strategies has also led to significant overtreatment of the disease, leading to a significant loss in patient quality of life with an added financial burden of 22 billion dollars annually in the U.S. alone. To better detect and guide biopsies, non- invasive magnetic resonance imaging (MRI) has emerged as an informative, accompanying tool along with the Prostate Imaging Reporting and Data System (PI-RADS), an internationally established scoring system for characterizing the risk of clinically significant PCa (csPCa) in focal lesions detected on MRI. While aiming to defer biopsies for low-risk patients whenever possible, ~30% of patients with a negative MRI still end up proceeding to biopsy due to suboptimal negative predictive values with current MRI techniques (~90%). This leads to unnecessary biopsies and post-procedure complications for a population in which the prevalence of csPCa is only around 8%. The majority of detected PCa (70%) is localized in the peripheral zone of the prostate. Thus, there is an urgent need for novel, non-invasive imaging techniques to improve our capability to more definitively rule out csPCa in the peripheral zone of the prostate to avoid unnecessary biopsies, complications, and costs. Our team has pioneered the prostate Magnetic Resonance Fingerprinting (MRF) technique, which simultaneously quantifies T1 and T2 in ~40 sec per slice. We propose to develop a rapid and reproducible MRF method to quantitatively and more accurately characterize prostatic peripheral zone tissue in order to limit overdiagnosis and overtreatment for patients with no csPCa. We will develop novel prostate MRF techniques to provide simultaneous and motion-robust T1, T2, and diffusion quantification (Aim 1). Rapid and whole-gland imaging will be achieved by leveraging novel deep learning techniques and multi-slice imaging. Deep-learning-based prostate segmentation derived from MRF signal evolutions will be further developed to automatically extract quantitative metrics from the peripheral zone for post-processing (Aim 2). These developed methods will be applied in a diagnostic study of a population with clinical indication for a biopsy to assess its capability to more accurately inform biopsy decision (Aim 3). Upon successful development, this MRF-based method will provide more quantitative tissue assessment of the prostate, optimizing biopsy avoidance in patients with a negative MRI.