The concept of lightweight design is to minimize the weight of a structure without compromising its loading capacity. However, their improved strength-to-weight ratio entails high complexity in material composition, bringing challenges to accurately modeling and analyzing structural performance. Hence a reliable, lightweight, and in-situ structural health monitoring (SHM) system is needed to monitor real-time structural behavior. In this study a multi-walled carbon nanotube(MWNT)-based polymer paint was first developed and improved to perform spatial strain sensing over lightweight structures. To further improve its strain sensitivity, the polymer matrix was replaced by Pluronic-127 to produce more stable MWNT paint solution. In order to further improve the consistency of the carbon nanotube (CNT) thin film, a CNT-suspended ink is developed and inkjet-printed over a flexible substrate. The inkjet-printed CNT thin film consists of a uniform morphology with evenly distributed nanoparticles. Coupled with an electrical impedance tomographic(EIT) method, the CNT thin film is able to reconstruct the conductivity change of itself over a tensile loaded coupon. Moreover, the elastoresistivity of the inkjet-printed CNT thin film was characterized, correlating its electrical property change with respect to the applied strain state using tensor analysis. It is observed that the EIT reconstruction result shares a similarity with the square root of the determinant of the resistivity matrix. Finally, the inkjet-printed CNT thin film was embedded at the adherend-adhesive interface of a single-lap joint. The testing coupon is tensile loaded to develop an in-plane shear strain distribution at the interface, and the conductivity change over the thin film is reconstructed by EIT. The finite element model of the joint shows a fair similarity to the EIT reconstruction, validating its application as an effective joint monitoring sensor.