A1 Vertaisarvioitu alkuperäisartikkeli tieteellisessä lehdessä
Interactive Lesion Segmentation with Shape Priors From Offline and Online Learning
Tekijät: Shepherd T, Prince SJD, Alexander DC
Kustantaja: IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Julkaisuvuosi: 2012
Journal: IEEE Transactions on Medical Imaging
Tietokannassa oleva lehden nimi: IEEE TRANSACTIONS ON MEDICAL IMAGING
Lehden akronyymi: IEEE T MED IMAGING
Numero sarjassa: 9
Vuosikerta: 31
Numero: 9
Aloitussivu: 1698
Lopetussivu: 1712
Sivujen määrä: 15
ISSN: 0278-0062
DOI: https://doi.org/10.1109/TMI.2012.2196285
Tiivistelmä
In medical image segmentation, tumors and other lesions demand the highest levels of accuracy but still call for the highest levels of manual delineation. One factor holding back automatic segmentation is the exemption of pathological regions from shape modelling techniques that rely on high-level shape information not offered by lesions. This paper introduces two new statistical shape models (SSMs) that combine radial shape parameterization with machine learning techniques from the field of nonlinear time series analysis. We then develop two dynamic contour models (DCMs) using the new SSMs as shape priors for tumor and lesion segmentation. From training data, the SSMs learn the lower level shape information of boundary fluctuations, which we prove to be nevertheless highly discriminant. One of the new DCMs also uses online learning to refine the shape prior for the lesion of interest based on user interactions. Classification experiments reveal superior sensitivity and specificity of the new shape priors over those previously used to constrain DCMs. User trials with the new interactive algorithms show that the shape priors are directly responsible for improvements in accuracy and reductions in user demand.
In medical image segmentation, tumors and other lesions demand the highest levels of accuracy but still call for the highest levels of manual delineation. One factor holding back automatic segmentation is the exemption of pathological regions from shape modelling techniques that rely on high-level shape information not offered by lesions. This paper introduces two new statistical shape models (SSMs) that combine radial shape parameterization with machine learning techniques from the field of nonlinear time series analysis. We then develop two dynamic contour models (DCMs) using the new SSMs as shape priors for tumor and lesion segmentation. From training data, the SSMs learn the lower level shape information of boundary fluctuations, which we prove to be nevertheless highly discriminant. One of the new DCMs also uses online learning to refine the shape prior for the lesion of interest based on user interactions. Classification experiments reveal superior sensitivity and specificity of the new shape priors over those previously used to constrain DCMs. User trials with the new interactive algorithms show that the shape priors are directly responsible for improvements in accuracy and reductions in user demand.
Turun yliopiston Tutkimustietojärjestelmän portaali