<ul><li>Nuclear Magnetic Resonance (NMR) spectroscopy is vital for analyzing molecules' properties, but its high cost and lengthy experiments call for optimization.</li><li>Non-Uniform sampling (NUS) is commonly used to reduce acquisition times in NMR, but it can introduce artifacts. Deep learning, specifically diffusion models, are proposed to enhance NUS spectra reconstruction quality.</li><li>The study explores the use of diffusion models on NUS data, resulting in improved reconstruction of spectra from the Artina dataset.</li><li>Using time-frequency domain data with diffusion models shows promise in enhancing the efficiency and accuracy of NMR spectroscopy, paving the way for future advancements in the field.</li></ul>

DiffNMR: Advancing Inpainting of Randomly Sampled Nuclear Magnetic Resonance Signals

Discover more