Automatic planning of liver tumor thermal ablation using deep reinforcement learning

Krishna Chaitanya, Chloé Audigier, Laura Elena Balascuta, Tommaso Mansi
Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, PMLR 172:219-230, 2022.

Abstract

Thermal ablation is a promising minimally invasive intervention to treat liver tumors. It requires a meticulous planning phase where the electrode trajectory from the skin surface to the tumor inside the liver as well as the ablation protocol are defined to reach a complete tumor ablation while considering multiple clinical constraints such as avoiding too much damage to healthy tissue. The planning is usually done manually based on 2D views of pre-operative CT images and can be extremely challenging for large or irregularly shaped tumors. Conventional optimization methods have been proposed to automate this complex task, but they suffer from high computation time. To alleviate this drawback, we propose to leverage a deep reinforcement learning (DRL) approach to find the optimal electrode trajectory that satisfies all the clinical constraints and does not require any labels in training. Here, we define a custom environment as the 3D mask with tumor, surrounding organs, skin labels along with an electrode line and ablation zone. An agent, represented by a neural network, interacts with the custom environment by displacing the electrode and therefore can learn an optimal policy. The reward assignment is done based on the clinical constraints. We explore discrete and continuous action-based approaches with double deep Q networks and proximal policy optimization (PPO), respectively. We perform an evaluation on the publicly available liver tumor segmentation (LITs) challenge dataset and obtain solutions that satisfy all clinical constraints comparable to the conventional method. The DRL method does not need any post-processing steps, allowing a mean inference time of 13.3 seconds per subject compared to the conventional optimization method’s mean time of 135 seconds. Moreover, the best DRL method (PPO) yields a valid solution irrespective of the tumor location within the liver that demonstrates its robustness.

Cite this Paper

BibTeX
@InProceedings{pmlr-v172-chaitanya22a, title = {Automatic planning of liver tumor thermal ablation using deep reinforcement learning}, author = {Chaitanya, Krishna and Audigier, Chlo\'e and Balascuta, Laura Elena and Mansi, Tommaso}, booktitle = {Proceedings of The 5th International Conference on Medical Imaging with Deep Learning}, pages = {219--230}, year = {2022}, editor = {Konukoglu, Ender and Menze, Bjoern and Venkataraman, Archana and Baumgartner, Christian and Dou, Qi and Albarqouni, Shadi}, volume = {172}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v172/chaitanya22a/chaitanya22a.pdf}, url = {https://proceedings.mlr.press/v172/chaitanya22a.html}, abstract = {Thermal ablation is a promising minimally invasive intervention to treat liver tumors. It requires a meticulous planning phase where the electrode trajectory from the skin surface to the tumor inside the liver as well as the ablation protocol are defined to reach a complete tumor ablation while considering multiple clinical constraints such as avoiding too much damage to healthy tissue. The planning is usually done manually based on 2D views of pre-operative CT images and can be extremely challenging for large or irregularly shaped tumors. Conventional optimization methods have been proposed to automate this complex task, but they suffer from high computation time. To alleviate this drawback, we propose to leverage a deep reinforcement learning (DRL) approach to find the optimal electrode trajectory that satisfies all the clinical constraints and does not require any labels in training. Here, we define a custom environment as the 3D mask with tumor, surrounding organs, skin labels along with an electrode line and ablation zone. An agent, represented by a neural network, interacts with the custom environment by displacing the electrode and therefore can learn an optimal policy. The reward assignment is done based on the clinical constraints. We explore discrete and continuous action-based approaches with double deep Q networks and proximal policy optimization (PPO), respectively. We perform an evaluation on the publicly available liver tumor segmentation (LITs) challenge dataset and obtain solutions that satisfy all clinical constraints comparable to the conventional method. The DRL method does not need any post-processing steps, allowing a mean inference time of 13.3 seconds per subject compared to the conventional optimization method’s mean time of 135 seconds. Moreover, the best DRL method (PPO) yields a valid solution irrespective of the tumor location within the liver that demonstrates its robustness.} }
Endnote
%0 Conference Paper %T Automatic planning of liver tumor thermal ablation using deep reinforcement learning %A Krishna Chaitanya %A Chloé Audigier %A Laura Elena Balascuta %A Tommaso Mansi %B Proceedings of The 5th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2022 %E Ender Konukoglu %E Bjoern Menze %E Archana Venkataraman %E Christian Baumgartner %E Qi Dou %E Shadi Albarqouni %F pmlr-v172-chaitanya22a %I PMLR %P 219--230 %U https://proceedings.mlr.press/v172/chaitanya22a.html %V 172 %X Thermal ablation is a promising minimally invasive intervention to treat liver tumors. It requires a meticulous planning phase where the electrode trajectory from the skin surface to the tumor inside the liver as well as the ablation protocol are defined to reach a complete tumor ablation while considering multiple clinical constraints such as avoiding too much damage to healthy tissue. The planning is usually done manually based on 2D views of pre-operative CT images and can be extremely challenging for large or irregularly shaped tumors. Conventional optimization methods have been proposed to automate this complex task, but they suffer from high computation time. To alleviate this drawback, we propose to leverage a deep reinforcement learning (DRL) approach to find the optimal electrode trajectory that satisfies all the clinical constraints and does not require any labels in training. Here, we define a custom environment as the 3D mask with tumor, surrounding organs, skin labels along with an electrode line and ablation zone. An agent, represented by a neural network, interacts with the custom environment by displacing the electrode and therefore can learn an optimal policy. The reward assignment is done based on the clinical constraints. We explore discrete and continuous action-based approaches with double deep Q networks and proximal policy optimization (PPO), respectively. We perform an evaluation on the publicly available liver tumor segmentation (LITs) challenge dataset and obtain solutions that satisfy all clinical constraints comparable to the conventional method. The DRL method does not need any post-processing steps, allowing a mean inference time of 13.3 seconds per subject compared to the conventional optimization method’s mean time of 135 seconds. Moreover, the best DRL method (PPO) yields a valid solution irrespective of the tumor location within the liver that demonstrates its robustness.
APA
Chaitanya, K., Audigier, C., Balascuta, L.E. & Mansi, T.. (2022). Automatic planning of liver tumor thermal ablation using deep reinforcement learning. Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 172:219-230 Available from https://proceedings.mlr.press/v172/chaitanya22a.html.

Related Material