We demonstrate visualization of lipid distribution in in-vitro artery model by 1.7-μm spectroscopic spectral-domain optical coherence tomography (SD-OCT). In the demonstration, we measure spectral fringes by a spectrometer with an extended InGaAs line sensor and a super-continuum (SC) light source whose spectrum is arranged to have its maximum intensity in 1.7-μm band. The OCT system has an axial resolution of 21μm, a measurement range of 5mm and a sensitivity of 108dB with an A-scan rate of 0.96kHz, which is limited by the noise of the available SC light source. The in-vitro model is made by injecting lipid into swine carotid artery, which is compared to intact artery. We perform Bscan of the model in water by connecting an OCT probe to the OCT system and pulling the probe back at 0.027mm/sec with a rotation rate of 112rpm. For visualizing lipid distribution, we adopt a spectroscopic OCT algorism where the detected spectral fringe is divided into six sub-bands, the set of the sub-band A-scans are fitted to a model accounting absorption characteristics of lipid with its peak at 1726nm, and the content of lipid is estimated as lipid score. As a result, the p-value of the lipid score between normal artery and plaque one is less than 1E-10 in 1-mm depth from the surface, which is significant of visualization of lipid distribution.