Back

Cancer Cell Removing Using a Reinforcement Learning Agent

Mousavi Fard, S. A.

2024-09-03 cancer biology
10.1101/2024.09.01.610680 bioRxiv
Show abstract

Cancer cell is a deadly problem which is the main cause of global death. Unfortunately, the conventional therapies like chemo/radio therapy are not viable ways to remove all of the cancer cells. Although Robotic achievements have been increased in cancer therapy, these devices do not have the decision-making ability to grasp their environment like biologists. In this paper, a cancer cell removing method based on Artificial Intelligence techniques is introduced. The proposed idea adopts a combination of object detection and reinforcement model in order to detect the cancer cells and take some actions to remove them. To implement this idea, YOLOv9 is trained on a cancer cell image dataset to detect and segment the cancer cell and create a set point for RL model then in the next step, Soft Actor Critic (SAC) is considered as a RL model to grasp the desired environment and take some appropriate actions to reach the target. The experimental result of this model shows that the proposed model can be adopted in different cancer therapy robots like micro/wireless soft robots to boost their performance in terms of their decision-making ability.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.