TY - GEN
T1 - A Maximum Entropy Deep Reinforcement Learning Neural Tracker
AU - Balaram, Shafa
AU - Arulkumaran, Kai
AU - Dai, Tianhong
AU - Bharath, Anil Anthony
PY - 2019
Y1 - 2019
N2 - Tracking of anatomical structures has multiple applications in the field of biomedical imaging, including screening, diagnosing and monitoring the evolution of pathologies. Semi-automated tracking of elongated structures has been previously formulated as a problem suitable for deep reinforcement learning (DRL), but it remains a challenge. We introduce a maximum entropy continuous-action DRL neural tracker capable of training from scratch in a complex environment in the presence of high noise levels, Gaussian blurring and detractors. The trained model is evaluated on two-photon microscopy images of mouse cortex. At the expense of slightly worse robustness compared to a previously applied DRL tracker, we reach significantly higher accuracy, approaching the performance of the standard hand-engineered algorithm used for neuron tracing. The higher sample efficiency of our maximum entropy DRL tracker indicates its potential of being applied directly to small biomedical datasets.
AB - Tracking of anatomical structures has multiple applications in the field of biomedical imaging, including screening, diagnosing and monitoring the evolution of pathologies. Semi-automated tracking of elongated structures has been previously formulated as a problem suitable for deep reinforcement learning (DRL), but it remains a challenge. We introduce a maximum entropy continuous-action DRL neural tracker capable of training from scratch in a complex environment in the presence of high noise levels, Gaussian blurring and detractors. The trained model is evaluated on two-photon microscopy images of mouse cortex. At the expense of slightly worse robustness compared to a previously applied DRL tracker, we reach significantly higher accuracy, approaching the performance of the standard hand-engineered algorithm used for neuron tracing. The higher sample efficiency of our maximum entropy DRL tracker indicates its potential of being applied directly to small biomedical datasets.
UR - http://dx.doi.org/10.1007/978-3-030-32692-0_46
U2 - 10.1007/978-3-030-32692-0_46
DO - 10.1007/978-3-030-32692-0_46
M3 - Published conference contribution
SN - 9783030326913
SN - 9783030326920
SP - 400
EP - 408
BT - International Workshop on Machine Learning in Medical Imaging
ER -