AI-Driven Segmentation and Volumetric Response Modeling of Liver Regions to Radiotherapy

Author

• Aashish Chandra Gupta, The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical SciencesFollow

Author ORCID Identifier

0000-0002-0178-0921

Date of Graduation

8-2026

Document Type

Dissertation (PhD)

Program Affiliation

Medical Physics

Degree Name

Doctor of Philosophy (PhD)

Advisor/Committee Chair

Kristy K. Brock

Committee Member

Eugene J. Koay

Committee Member

Rebecca M. Howell

Committee Member

Jia Wu

Committee Member

James P. Long

Abstract

In liver-directed radiotherapy (RT), liver regions receiving higher doses typically undergo atrophy while contralateral/adjacent lower-dose regions may exhibit compensatory hypertrophy through regeneration of healthy tissue. Optimizing the RT plan to promote regional hypertrophy while minimizing the risk of developing atrophy has the potential to enhance post-RT liver function and long-term survivorship. However, current clinical practice largely relies on global liver dose-volume metrics during RT-planning, which may obscure favorable dose-response correlation and limit actionable guidance for clinicians. Therefore, we hypothesized that post-RT regional liver response is governed by a combination of region-specific dose-volume and patient clinical features, and that these responses can be predicted with clinically meaningful accuracy using AI-driven models to guide RT technique selection.

To address this, we developed fully automated deep-learning-based auto-contouring models using nnU-Net and patch-based Attention U-Net (paU-Net) utilizing 200 CT images to auto-contour liver segments 1, 2, 3, 4, 5-8 and spleen. nnU-Net architecture outperformed paU-Net, achieving a final mean Dice 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 test sets. Qualitative analysis showed over 85% of test cases achieved a Likert score≥3, indicating clinically acceptable contours.

Using the auto-contouring model, we curated a cohort of 207 patients treated with RT for diverse liver tumor types and quantified an uncertainty threshold of 16cc to define clinically meaningful functional regional-volume change (ΔFV); at three-months post-RT across 561 liver regions. Hypertrophy (ΔFV>15.8cc), atrophy (ΔFV< -15.8cc), and stable responses were observed in 28%, 50%, and 22% of regions respectively. A mean dose threshold of 15Gy_EQD2 distinguished hypertrophy from non-hypertrophy response with an area under receiver operating characteristic curve (AUROC) of 0.71; however, the threshold did not generalize when assessing rate of hypertrophy/atrophy below 15Gy_EQD2 across tumor types and liver regions which supported the need for region-specific predictive modelling.

We subsequently developed and validated machine learning models predicting hypertrophy and atrophy responses of right, left lateral, and left central liver regions. Cross-validation AUROCs ranged from 0.64 to 0.77. Key features included BMI, bilirubin, ALT, albumin-bilirubin score, functional volume, percent volume of a region spared from 5-55Gy_EQD2 normalized to body-surface area, and minimum doses received by more than 2% and 98% of liver regions. On a withheld test set, accuracy for right, left lateral, left central hypertrophy/not-hypertrophy models were 0.69, 0.72, 0.1 and for atrophy/not-atrophy were 0.69, 0.72, 0.72, respectively. Model interpretability analysis revealed a strong association between regional dose sparing and hypertrophy, whereas higher point dose such as D2 were associated with atrophy.

Finally, we demonstrated the clinical utility of our models by comparing predicted regional responses from 3DCRT, IMRT, and VMAT plans. Predicted atrophy showed strong overlap across plans, whereas hypertrophy exhibited greater plan-dependent variability, which suggests that selected patients may benefit from estimating treatment approach-specific regional responses.

In conclusion, our work quantified clinically meaningful relations between radiation dose and regional liver response post-RT, and demonstrated the utility of AI-driven segmentation and regional response modelling for personalization of liver-directed RT.

Recommended Citation

Gupta, Aashish Chandra, "AI-Driven Segmentation and Volumetric Response Modeling of Liver Regions to Radiotherapy" (2026). Dissertations and Theses (Open Access). 1570.
https://digitalcommons.library.tmc.edu/utgsbs_dissertations/1570

Keywords

Liver, Radiotherapy, AI, ML, artificial intelligence, segmentation, prediction, machine learning, hypertrophy, atrophy