The SIRA (Intelligent Anatomical Recognition System for Robotic Surgery) project is developing an artificial intelligence (AI) system that will recognise organs and foreign objects in real time during robotic surgery to improve the accuracy and safety of procedures. This initiative is the result of collaboration between the University of Valencia, through the Institute on Robotics and Information and Communication Technologies (IRTIC), the Research Foundation of the General University Hospital of Valencia (FIHGUV) and the technology company Lãberit.
SIRA will be able to recognise organs, anatomical structures and foreign objects in real time during robotic surgeries such as those performed with systems like Da Vinci. The aforementioned surgical equipment consists of the console from which the expert personnel perform the surgery, the robotic arms that make the incisions in the patient's body, and the device from which the other professionals monitor the operation in the operating theatre.
SIRA will also be able to make real-time suggestions to surgeons performing the operation to optimise decision-making. The initiative has used 3.000 hours of surgical recordings inside the human body to train the AI model to successfully identify organs, tissues, and objects related to the surgical procedure.
This material covers the specialities of gynaecology, urology, thoracic surgery, and general and digestive system surgery. The videos are anonymised and only internal images of the procedures are used.
“Collaboration between the university, a public hospital and a technology company is key to improving the future of medicine in the Valencian Community”, has emphasised Professor José Vicente Riera, principal investigator of the SIRA project, when assessing the strategic nature of the initiative. In his opinion, the combined capabilities of these three Valencian entities reflect a firm commitment to healthcare innovation and will place the region at the forefront of precision surgery.
The initiative, which began development in February 2025, has a budget of close to half a million euros and is funded by grants for experimental development projects in the field of artificial intelligence and for promoting the development of sectoral data spaces by the Valencian Regional Government, within the framework of the Recovery, Transformation and Resilience Plan. This investment, co-financed with Next Generation funds from the European Union, is part of the Valencian Community's digital transformation strategy.
Development in phases
In the initial phase, the consortium has worked on a surgical labelling system for images collected during robotic operations, which is now practically complete. This is a fundamental task that will serve as the basis for training and validating the SIRA's AI algorithms. This process involves taking detailed notes on videos and still frames of surgeries to ‘teach’ the artificial intelligence model to correctly identify each organ, tissue, or object in the surgical field.
Based on these labelled images, the software will learn to distinguish between different anatomical structures and improve its recognition capabilities with each iteration.
When operational, the SIRA system will function as an intelligent assistant annexed to the robotic surgical platform used in procedures. Using advanced computer vision techniques, the software will accurately identify the organs appearing in the surgeon's field of view and highlight key structures to facilitate the specialist's spatial orientation.
It will also be able to detect the presence of foreign objects, such as gauze or misplaced instruments, immediately alerting the medical team. Likewise, the system will provide real-time alerts when surgical instruments approach vulnerable anatomical areas, such as major blood vessels, helping to prevent inadvertent damage during surgery. These functions aim to achieve more precise, safer and more effective surgeries, reducing the risk of complications for the patient.
The SIRA consortium combines the knowledge and experience of the three partners to tackle this multidisciplinary challenge. The IRTIC at the University of Valencia coordinates the project and leads the technological development of the system, applying its expertise in artificial intelligence and computer vision. For its part, FIHGUV (General Hospital) contributes clinical knowledge and takes on the crucial task of labelling the surgical images that will feed the AI.
Finally, Lãberit provides the necessary technological infrastructure and will be responsible for facilitating the subsequent implementation of the system in hospitals, ensuring data security and preparing for its large-scale clinical exploitation and use. This collaboration between universities, the public health system, and private companies allows the project to be approached from all angles, from basic research to application in the operating theatre.
Those responsible for the project highlight the significant clinical and social impact expected: once developed, SIRA will be able to be integrated as a complementary module into the robotic surgery systems of several public hospitals in the Valencian Community, contributing to the optimisation of surgical procedures and positioning the region as a global benchmark in robotic surgery and medical artificial intelligence. In addition to improving patient safety and surgical outcomes, this technology will enhance advanced medical training and research, consolidating an ecosystem of innovation in healthcare that will have a positive impact on society.
Also in:
Science Park of the University of Valencia
