Fully automated deep learning based auto-contouring of liver segments and spleen on contrast-enhanced CT images.

Manual delineation of liver segments on computed tomography (CT) images for primary/secondary liver cancer (LC) patients is time-intensive and prone to inter/intra-observer variability. Therefore, we developed a deep-learning-based model to auto-contour liver segments and spleen on contrast-enhanced CT (CECT) images. We trained two models using 3d patch-based attention U-Net ([Formula: see text] and 3d full resolution of nnU-Net ([Formula: see text] to determine the best architecture ([Formula: see text]. BA was used with vessels ([Formula: see text] and spleen ([Formula: see text] to assess the impact on segment contouring. Models were trained, validated, and tested on 160 ([Formula: see text]), 40 ([Formula: see text]), 33 ([Formula: see text]), 25 (CCH) and 20 (CPVE) CECT of LC patients. [Formula: see text] outperformed [Formula: see text] across all segments with median differences in Dice similarity coefficients (DSC) ranging 0.03-0.05 (p < 0.05). [Formula: see text], and [Formula: see text] were not statistically different (p > 0.05), however, both were slightly better than [Formula: see text] by DSC up to 0.02. The final model, [Formula: see text], showed a mean DSC of 0.89, 0.82, 0.88, 0.87, 0.96, and 0.95 for segments 1, 2, 3, 4, 5-8, and spleen, respectively on entire test sets. Qualitatively, more than 85% of cases showed a Likert score [Formula: see text] 3 on test sets. Our final model provides clinically acceptable contours of liver segments and spleen which are usable in treatment planning.

Analysis and prediction of liver volume change maps derived from computational tomography scans acquired pre- and post-radiation therapy.

External beam radiation therapy (EBRT) of liver cancers can cause local liver atrophy as a result of tissue damage or hypertrophy as a result of liver regeneration. Predicting those volumetric changes would enable new strategies for liver function preservation during treatment planning. However, understanding of the spatial dose/volume relationship is still limited. This study leverages the use of deep learning-based segmentation and biomechanical deformable image registration (DIR) to analyze and predict this relationship. Pre- and Post-EBRT imaging data were collected for 100 patients treated for hepatocellular carcinomas, cholangiocarcinoma or CRC with intensity-modulated radiotherapy (IMRT) with prescription doses ranging from 50 to 100 Gy delivered in 10-28 fractions. For each patient, DIR between the portal and venous (PV) phase of a diagnostic computed tomography (CT) scan acquired before radiation therapy (RT) planning, and a PV phase of a diagnostic CT scan acquired after the end of RT (on average 147 ± 36 d) was performed to calculate Jacobian maps representing volume changes in the liver. These volume change maps were used: (i): to analyze the dose/volume relationship in the whole liver and individual Couinaud's segments; and (ii): to investigate the use of deep-learning to predict a Jacobian map solely based on the pre-RT diagnostic CT and planned dose distribution. Moderate correlations between mean equivalent dose in 2 Gy fractions (EQD2) and volume change was observed for all liver sub-regions analyzed individually with Pearson correlationrranging from -0.36 to -067. The predicted volume change maps showed a significantly stronger voxel-wise correlation with the DIR-based volume change maps than when considering the original EQD2 distribution (0.63 ± 0.24 versus 0.55 ± 23, respectively), demonstrating the ability of the proposed approach to establish complex relationships between planned dose and liver volume response months after treatment, which represents a promising prediction tool for the development of future adaptive and personalized liver radiation therapy strategies.