Towards Safe Mechanical Ventilation Treatment Using Deep Offline Reinforcement Learning
This work addresses the problem of reducing the time-consuming and complex task of manual ventilator adjustments for healthcare workers, though it appears incremental as it builds on existing offline RL methods applied to a specific medical domain.
The paper tackled the challenge of automating mechanical ventilation settings for patients with pulmonary impairment by developing DeepVent, an offline deep reinforcement learning agent based on Conservative Q-Learning, which recommends safe ventilator parameters and outperforms physicians from the MIMIC-III dataset in promoting 90-day survival.
Mechanical ventilation is a key form of life support for patients with pulmonary impairment. Healthcare workers are required to continuously adjust ventilator settings for each patient, a challenging and time consuming task. Hence, it would be beneficial to develop an automated decision support tool to optimize ventilation treatment. We present DeepVent, a Conservative Q-Learning (CQL) based offline Deep Reinforcement Learning (DRL) agent that learns to predict the optimal ventilator parameters for a patient to promote 90 day survival. We design a clinically relevant intermediate reward that encourages continuous improvement of the patient vitals as well as addresses the challenge of sparse reward in RL. We find that DeepVent recommends ventilation parameters within safe ranges, as outlined in recent clinical trials. The CQL algorithm offers additional safety by mitigating the overestimation of the value estimates of out-of-distribution states/actions. We evaluate our agent using Fitted Q Evaluation (FQE) and demonstrate that it outperforms physicians from the MIMIC-III dataset.