The standard treatment workflow for treating cervical cancer involves external radiation therapy and concurrent chemotherapy, followed by brachytherapy. Throughout the course of treatment, the patient has to undergo multiple computed tomography (CT) scans. This leads to an additional burden to the patients, especially in rural areas, owing to logistics-related cost issues, apart from exposure to additional radiation related risks. An alternative solution to address this issue is to simulate pseudo-CT images instead of acquiring actual CT images. In this study, we presented a methodology for simulating pseudo-CT based on deformable image registration (DIR) between the ultrasound images acquired before radiotherapy and after brachytherapy applicator insertion stages. The deformation field obtained after the registration was applied to the corresponding CT images acquired before radiotherapy to simulate the pseudo-CT image corresponding to post applicator insertion that is used for dosimetry planning. Preliminary studies were performed on a commercially available multimodal abdominal phantom data and the performance of five open-source DIR algorithms was evaluated to test the feasibility of the methodology. Several quantitative similarity metrics are used to compare the simulated pseudo-CT with the ground truth CT. The algorithms were found to perform equally well on the phantom data and the mean target registration error (TRE) between the lesion locations was found to be approximately 3.8 mm. The study was then performed on the data from three cervical cancer patients. In addition to the similarity metrics, distance between the applicators as well as target registration error (TRE) between the applicator positions were measured in both images. The mean values of distance measurement error ranged from 0.88–2.48 mm, 0.93–3.6 mm, and 0.5–4.8 mm for patients 1 to 3, respectively. All the algorithms were found to perform better in determining tandem applicator positions than ovoid applicator positions, with the demons algorithm and MSIDIR being more consistent than the others. It can be concluded from the study that it is feasible to simulate pseudo-CT from previously acquired ultrasound and CT data. This can reduce the number of CT scans required, especially for cancer patients undergoing treatment