Abdominal T1 mapping is important for quantitative evaluation of various pathologies. A recent inversion recovery radial balanced-SSFP (IR-radSSFP) technique allows high resolution T1 mapping of ten slices within a single breath hold period (BHP), but requires multiple BHPs for full abdominal coverage. We propose an accelerated T1 mapping framework which utilizes deep learning to estimate T1 using a fraction of the T1 recovery curve (T1RC). In vivo experiments demonstrate that the proposed framework achieves less than 6% T1 error while using only 25% of the T1RC of the earlier IR-radSSFP technique. This enables full abdominal coverage within a single BHP.