Abstract
We seek the development and evaluation of a fast, accurate, and consistent method for general-purpose segmentation, based on interactive machine learning (IML). To validate our method, we identified retrospective cohorts of 20 brain, 50 breast, and 50 lung cancer patients, as well as 20 spleen scans, with corresponding ground truth annotations. Utilizing very brief user training annotations and the adaptive geodesic distance transform, an ensemble of SVMs is trained, providing a patient-specific model applied to the whole image. Two experts segmented each cohort twice with our method and twice manually. The IML method was faster than manual annotation by 53.1% on average. We found significant (p < 0.001) overlap difference for spleen (Dice(IML)/Dice(Manual) = 0.91/0.87), breast tumors (Dice(IML)/Dice(Manual) = 0.84/0.82), and lung nodules (Dice(IML)/Dice(Manual) = 0.78/0.83). For intra-rater consistency, a significant (p = 0.003) difference was found for spleen (Dice(IML)/Dice(Manual) = 0.91/0.89). For inter-rater consistency, significant (p < 0.045) differences were found for spleen (Dice(IML)/Dice(Manual) = 0.91/0.87), breast (Dice(IML)/Dice(Manual) = 0.86/0.81), lung (Dice(IML)/Dice(Manual) = 0.85/0.89), the non-enhancing (Dice(IML)/Dice(Manual) = 0.79/0.67) and the enhancing (Dice(IML)/Dice(Manual) = 0.79/0.84) brain tumor sub-regions, which, in aggregation, favored our method. Quantitative evaluation for speed, spatial overlap, and consistency, reveals the benefits of our proposed method when compared with manual annotation, for several clinically relevant problems. We publicly release our implementation through CaPTk (Cancer Imaging Phenomics Toolkit) and as an MITK plugin.