BACKGROUND: Corrective surgery for pectus excavatum often relies on the Haller index (HI), derived from chest roentgenograms or computed tomography; however, this exposes children to potentially unnecessary radiation. Our aim was to develop a novel three-dimensional (3D) optical imaging technique to accurately measure chest wall dimensions in a clinically relevant manner. METHODS: Patients with pectus excavatum were imaged using a 3D structured light scanner. Patient characteristics, including height, weight, body mass index, and radiographic HIs (rHI) were recorded. We defined the optical index (OI) as the ratio of the lateral to anterior-posterior measurements obtained from the 3D optical images and compared those to patients' rHIs. Two-thirds of the patients' images were used to develop a predictive model of the rHI, using their OI and biometric data in multilinear regression modeling. The predictive model was applied to the remaining images, and the predicted HIs (pHI) were compared to the rHIs. RESULTS: Optical imaging was performed in 42 patients (ages, 5-35 years) with pectus excavatum; of these, 31 had recent chest roentgenograms, with rHIs ranging from 2.00 to 7.20. The OIs derived from the images correlated closely with rHIs (r = 0.850). Our predictive model, using patients' OI, height, and weight, was able to accurately estimate their rHIs with a median error of 8.11% (interquartile range, 3.5%-17.4%). CONCLUSIONS: 3D optical imaging of patients with pectus excavatum is emerging as an alternative method to assess HIs without the use of ionizing radiation. Additional studies will focus on volumetric quantification of chest wall deformities, using the 3D capabilities of this technology.
BACKGROUND: Corrective surgery for pectus excavatum often relies on the Haller index (HI), derived from chest roentgenograms or computed tomography; however, this exposes children to potentially unnecessary radiation. Our aim was to develop a novel three-dimensional (3D) optical imaging technique to accurately measure chest wall dimensions in a clinically relevant manner. METHODS:Patients with pectus excavatum were imaged using a 3D structured light scanner. Patient characteristics, including height, weight, body mass index, and radiographic HIs (rHI) were recorded. We defined the optical index (OI) as the ratio of the lateral to anterior-posterior measurements obtained from the 3D optical images and compared those to patients' rHIs. Two-thirds of the patients' images were used to develop a predictive model of the rHI, using their OI and biometric data in multilinear regression modeling. The predictive model was applied to the remaining images, and the predicted HIs (pHI) were compared to the rHIs. RESULTS: Optical imaging was performed in 42 patients (ages, 5-35 years) with pectus excavatum; of these, 31 had recent chest roentgenograms, with rHIs ranging from 2.00 to 7.20. The OIs derived from the images correlated closely with rHIs (r = 0.850). Our predictive model, using patients' OI, height, and weight, was able to accurately estimate their rHIs with a median error of 8.11% (interquartile range, 3.5%-17.4%). CONCLUSIONS: 3D optical imaging of patients with pectus excavatum is emerging as an alternative method to assess HIs without the use of ionizing radiation. Additional studies will focus on volumetric quantification of chest wall deformities, using the 3D capabilities of this technology.