Comparison between mixed reality with artificial algorithms and ultrasound in localization of anterior thigh flap...

This study explores the efficacy of integrating mixed reality (MR) technology with artificial algorithms for locating vessels during anterolateral thigh perforator flap surgeries, comparing it with color Doppler ultrasonography (CDU) to provide clinical insights.

Eighty patients were randomly assigned to the MR group or the CDU group, with 40 patients in each. In the MR group, a localization device was attached to the lower limb, and CT angiography (CTA) data were converted into a 3D model. An artificial algorithm matched the device with the 3D model to overlay perforating vessels. The CDU group used traditional preoperative localization. Primary outcomes included the number of identified perforators and the distance between marked and actual vessel exit points. Secondary outcomes were flap harvest time and flap survival.

Recognition rates were 94.3% in the MR group and 82.0% in the CDU group (P = 0.008). The average distance between marked and actual exit points was 1.5 mm vs. 2.7 mm (P < 0.0001). Flap harvest times averaged 52 and 68 min, respectively (P < 0.0001). In the MR group, one flap developed an infection and another necrosis, while in the CDU group, one flap had a crisis, and two experienced necrosis.

Mixed reality combined with artificial algorithms offers superior vessel localization compared to ultrasound and holds promise for multidisciplinary perforator flap surgeries.

This study was approved by the Ethics Committee of Chongqing University Cancer Hospital (Ethical Approval Number: CZLS2021177-A) and retrospectively registered at the Chinese Clinical Trial Register (registration number: CHiCTR2400087615, date of registration: 2024-07-31).