Electrical Impedance Tomography (EIT) is a non-invasive imaging modality that reconstructs internal conductivity distributions from boundary voltage measurements. This study presents a 3D EIT reconstruction framework for detecting and localizing single and multiple internal targets in a cylindrical saline-water tank under controlled experimental conditions, with the longer-term aim of supporting biomedical EIT applications such as heart-lung imaging. The reconstruction framework was implemented in MATLAB using EIDORS with a 3D cylindrical forward model and 32 electrodes arranged in two axial rings, enabling volumetric data acquisition in the region between the electrode planes. A difference imaging approach was used based on a one-step Gauss–Newton inverse solver with NOSER regularization and Jacobian-based sensitivity normalization. Spatial smoothing was applied to reduce artifacts and improve visualization of the reconstructed conductivity distributions. Experiments were carried out in a tank with a height 25 cm and radius 9.6 cm using fruit targets, including apple, lemon, and orange to represent differences in conductivity, shape, and size. The targets were arranged in both single-target and multiple-target configurations and were placed in the central sensing region between the two electrode rings. The reconstructed images successfully localized the inclusions and preserved their relative positions in both single and multiple target cases. In the multiple-target experiments, the method distinguished separate conductivity perturbations and provided reasonable spatial localization. However, a limited resolution in the z-direction was observed, resulting in partially merged reconstructed regions rather than clearly separated depth-wise targets. Although the reconstructed targets appeared slightly enlarged compared to the actual objects due to the inherent spatial smoothing of EIT, the overall spatial distribution remained consistent with the true object arrangement. Overall, the results demonstrate the feasibility of EIT for volumetric detection of multiple internal targets in tank experiments and provide a practical foundation for future biomedical 3D EIT research, particularly toward heart-lung imaging.