Amir Hossein Sadeghi

UMC Utrecht

In short, Amir Hossein Sadeghi's research entails the following:

IMAGINg-based augmentation for RobotIc-assisted lung Surgery: the IMAGINERS project for automated and intelligent surgical guidance

Lung cancer, the leading cause of cancer-related deaths, includes rare neuroendocrine tumors (NETs) that have favorable outcomes when treated adequately. While (anatomical) resection, such as lobectomy or segmentectomy, remains the gold standard for treating NETs and the more prevalent early-stage non-small cell lung cancer (NSCLC), segmentectomy offers comparable survival rates to lobectomy but with improved lung function, making it potentially a better option for patients with comorbidities or future treatment needs.

Compared to lobectomies, segmentectomy presents technical challenges, particularly in robotic-assisted minimally invasive surgery, due to the need for precise identification of patient-specific segmental vascular and bronchial anatomy. Distinguishing segmental borders—which are difficult to see on conventional 2D-CT scans—is critical for appropriate patient selection and surgical planning. Conventional imaging methods struggle to account for intraoperative lung deformations (e.g. surgical manipulation), making these models unsuitable for intra-operative anatomical guidance.

The IMAGINERS project addresses these challenges by developing an AI-driven platform integrating deep learning and dynamic 3D-simulated models for preoperative planning and intraoperative guidance. By enabling real-time 3D model adjustments to match surgical views, this framework aims to enable imaging-guided robotic lung surgery. In addition, our goal is to enhance surgical precision/safety, and to expand the use of segmentectomy, ultimately improving outcomes for patients with rare lung tumors and early-stage NSCLC.

The IMAGINERS project is a multidisciplinary collaboration led by Dr. Sadeghi (cardiothoracic surgeon) at UMC Utrecht, with technical contributions from Dr. Yasmina Al Khalil and Dr. Ruisheng Su at TU Eindhoven. The initiative also involves several Dutch and European partner institutions, focusing on advancing robotic-assisted lung surgery through innovative technologies and methodologies.

Short summary of results February 2026

In the first year of the IMAGINERS project, a multidisciplinary review article has been written on how artificial intelligence (AI) can support surgeons during operations and how these systems can be validated for clinical use. This work was carried out in collaboration with Eindhoven University of Technology, Radboudumc, and UMC Utrecht, and the manuscript is nearly ready for submission.

In addition, a first proof-of-concept study on AI-based recognition of surgical phases (what the surgeon is doing during the surgery) in lung surgery has been submitted for publication. We are currently also collecting and annotating surgical videos from multiple hospitals, which will be used to train AI models to recognize both surgical steps and anatomical structures.

In 2026, the project will focus on gathering feedback from surgical teams and experts to guide future developments, as well as on advancing CT-based imaging techniques to further improve surgical planning.