Artificial intelligence in thoracic surgery: Helper or competitor?

Artificial intelligence (AI) stands out as one of the most groundbreaking technologies in modern medicine. Advances in image processing, decision support systems, and robotic applications are particularly prominent in technically complex and visually intensive disciplines. Thoracic surgery is among the fields most affected by this transformation. Today, AI algorithms can classify pulmonary nodules on low-dose chest CT scans with high accuracy and estimate malignancy risk by analyzing nodule characteristics. In addition, applications such as three-dimensional vascular modeling and automated anatomical segmentation facilitate the planning of procedures like video-assisted thoracoscopic surgery (VATS). In the field of robotic surgery, AI-supported systems assist surgeons with functions such as tissue recognition, camera guidance, and margin identification. Some systems are designed to highlight critical anatomical structures during dissection, thereby enhancing surgical safety. Furthermore, AI-based scoring systems that predict postoperative complication risks hold promise for more individualized patient management. However, these technological advancements also bring certain ethical and practical challenges. The responsibility for decisions based on algorithmic suggestions is debatable, and the fact that these systems are often trained on Western-centered datasets may limit their accuracy in local patient populations. Surgery is not only a technical but also an intuitive discipline. AI should support decision-making processes, not replace the clinician. Encouraging examples have also emerged from Türkiye. At Ankara University, under the leadership of Prof. Dr. Ayten Kayı Cangır, an AI-based model has been developed to propose treatment strategies for lung cancer patients using low- dose CT data—without requiring biopsy. Such local initiatives are crucial for the integration of AI into clinical practice. In terms of medical education, a study conducted by Mesut Buz and Prof. Dr. Recep Demirhan compared the knowledge level of medical students and ChatGPT-4 on thoracic surgery questions. The study highlighted the educational potential of AI, showing that ChatGPT-4 outperformed students with a success rate of 95%. In conclusion, AI should not be seen as a rival but rather a strategic ally in thoracic surgery. AI applications harmonized with surgical experience and clinical intuition may open the door to a safer and more effective surgical future.