Preoperative visualization of congenital lung abnormalities: hybridizing artificial intelligence and virtual reality

Abstract

OBJECTIVES When surgical resection is indicated for a congenital lung abnormality, lobectomy is often preferred over segmentectomy, mostly because the latter is associated with more residual disease. Presumably, this occurs in children because sublobar surgery often does not adhere to anatomical borders (wedge resection instead of segmentectomy), thus increasing the risk of residual disease. This study investigated the feasibility of identifying eligible cases for anatomical segmentectomy by combining virtual reality and artificial intelligence. METHODS Semi-automated segmentation of bronchovascular structures and lesions were visualized with virtual reality and artificial intelligence technology. Two specialists independently evaluated via a questionnaire the informative value of regular computed tomography versus three-dimensional-virtual reality images. RESULTS Five asymptomatic, non-operated cases were selected. Bronchovascular segmentation, volume calculation, and image visualization in the VR environment was successful in all cases. Based on the computed tomography images, assignment of the congenital lung abnormality lesion to specific lung segments matched between the consulted specialists in only one out of the cases. Based on the three three-dimensional-virtual reality images, however, the localization matched in three of the five cases. If the patients would have been operated, adding the three-dimensional-virtual reality tool to the pre-operative workup would have resulted in changing the surgical strategy (i.e., lobectomy versus segmentectomy) in four cases. CONCLUSIONS This study demonstrated the technical feasibility of a hybridized artificial intelligence –virtual reality visualization of segment-level lung anatomy in patients with congenital lung abnormalities. Further exploration of the value of three-dimensional-virtual reality in identifying eligible cases for anatomical segmentectomy is therefore warranted.

Anatomy of the lung revisited by 3D-CT imaging

“…A decade has passed since the initial reports on 3D-CT image-guided lung resection ( 54 – 59 ) and 3D-CT imaging has become widely implemented with results of prospective multicenter studies now being reported ( 60 ). The current surge of 3D-CT imaging analysis shows that the field is still growing, with the technology continuing to improve and now even being combined with virtual reality and artificial intelligence ( 61 , 62 ). Future studies using these new and innovative methodologies will continue to refine our understanding of lung anatomy while enhancing our ability to perform safe and effective surgical resections.…”

Artificial Intelligence in Cardiac Surgery: A Systematic Review

“…The potential clinical feasibility of a combined VR/AR and AI approach has also been demonstrated in fairly recent studies done by Sadeghi et al and Bakhuis et al which showcased the use of a combined VR and AI strategy in providing better visualization of their patient’s’ pulmonary anatomy. (54,55)…”

Prognosis of infants with congenital pulmonary airway malformations after surgery: a short and mid-term evaluation

“…Thoracoscopic anatomical segmentectomy is the ideal surgical option for removing the lesion while preserving normal lung tissue within the lobes and protecting lung function. Nevertheless, non-anatomic lung preservation surgery is more likely to have residual lesions than lobectomy [10,11].…”