采用三维数据分析拍摄角度对保乳手术美容效果评价的影响

Effect of camera angle on cosmetic assessment of breast conservative surgery with three dimensional data

目的采用三维(3D)激光扫描数据研究使用基于二维(2D)图片的乳腺癌保乳手术美容效果客观评价方法时,拍摄图片的角度对评价结果一致性的影响。方法获取两种不同图源的2D图片:采用索尼数码单反相机α77拍摄获得接受保乳手术治疗的30位患者乳房区域的2D数码正面照片;同时使用激光扫描仪获取患者乳房的完整3D几何模型和表面纹理,并对沿身体中线旋转后的3D模型进行屏幕截图,生成不同视角的2D图片来模拟不同的拍摄角度。邀请3位专家依据Harris主观标准对两种图源的正面照片进行美容效果评价;并采用客观评价方法来评价所有2D图片;分析拍摄角度对总体客观评价结果及乳房对称性、颜色和瘢痕三个客观评价指标测量值的影响。结果两种图源的2D图片对主观评价结果的影响差异无统计学意义(P>0. 05)。使用客观评价方法依次评价两种图源的2D图片,也显示出图源对评价的影响差异无统计学意义(P>0. 05);采用客观评价方法进行美容效果评价,结果表明不同旋转角度对总体评价结果有明显的影响(P<0. 05),同时乳房对称性、颜色和瘢痕三个特征指标的测量值随拍摄角度的变化发生明显的改变(P<0. 05)。结论基于2D图片的评价易受拍摄角度的影响,基于3D数据的评价可给出更为客观的评价结果 ,具有很好的应用前景。

Objective To investigate the effect of camera angles on consistency of two-dimension (2D) picture-based objective evaluation for cosmetic outcome of breast conservative surgery (BCT) with three-dimensional (3D) scanning data. Methods The face-view digital photographs were taken from 30 patients treated with BCT and their 3D digital breast models with high quality color textures were acquired using a 3D laser scanner.The 2D snapshots of 3D models were obtained after rotation along the midline of the body ranging from -10° to 10° to stimulate different camera angles.The cosmetic outcome was assessed on both face-view photographs and 2D snapshots,subjectively and objectively.The subjective assessment was performed by three experts using the Harris scale (a subjective method) on face-view photographs,while the objective assessment was undertaken on both face-view photographs and 2D snapshots with BCCT.core,a representative 2D-picture based multi-features objective method and software tool.The effects of camera angles on the overall cosmetic assessment and the measurements of indices to for assessing the features of breast symmetry,color and scar visibility were also investigated. Results There was no significant difference in subjective assessment result when two different types of 2D face-view pictures (2D snapshot and 2D picture) from the same patient were used for subjective assessment by three experts ( P >0.05);there was also no significant difference in objective assessment result when BCCT.core was used to assess the two types of 2D face-view pictures ( P >0.05).The objective assessment on 2D snapshots showed that the rotation angle did affect both the overall cosmetic assessment result and the measurements indices for assessing the breast features ( P <0.05). Conclusion The 2D-picture based cosmetic evaluation is susceptible to camera angle,and more consistent objective evaluation results can be achieved with 3D scanning data.3D scanning data has a great potential to be used for more consistent and objective assessment of cosmetic outcomes in patients after breast conservative surgery.