Spinal cord injury (SCI) affects millions globally, with half a million new incidents each year, resulting in considerable challenges. Traditional rehabilitation approaches frequently do not address personal requirements and can be exhausting for physiotherapists, so this is where this project comes in. This thesis analyses the effects of gait-based rehabilitation using robotic systems, specifically using the Lokomat system, in patients with incomplete spinal cord injury (iSCI). The main objective is to objectively evaluate how this therapy affects gait quality and balance function by collecting and analysing biomechanical data in PRE/MID and POST treatment conditions, so it could be a building block towards the future personalisation of robotic therapies for SCI patients. For this purpose, a study was conducted with four patients undergoing a 40-session training programme with the aid of a walker, complemented with motion capture (performed by the National Hospital for Paraplegics of Toledo (HNP)) and data processed with biomechanical models in OpenSim to calculate and visualise parameters such as joint angles, forces and moments. In addition, analyses of spatio-temporal, kinematic, kinetic and balance metrics were performed to identify significant changes in motor function, gait efficiency and postural stability. The results showed substantial improvements in gait parameters, such as increased stride length and gait speed, as well as increased mobility in key joints (hip and ankle). There was also evidence of a reduction in pathological patterns and increased coordination between body segments, reflecting a more efficient and safe gait. In terms of balance, postural pathologies decreased, and the trajectory of the centre of mass became more linear and stable, with less dependence on external support. These findings support the incorporation of robotic therapy in rehabilitation programmes for patients with iSCI, highlighting its ability to promote neuroplasticity and improve motor and postural function. It is recommended to expand the sample of participants, to perform long-term follow-ups and to deepen neuromuscular analysis, including electromyography, to better understand the underlying physiological mechanisms. In addition, the development of personalised predictive models using biomechanical simulations could further enhance the individualisation of treatments and maximise the benefits for each patient.