Dynamic Projection of Medication Nonpersistence and Nonadherence Among Patients With Early Breast Cancer.

Importance: Oral endocrine treatments have been shown to be effective when carefully adhered to. However, in patients with early breast cancer, adherence challenges are notable, with 17% experiencing nonpersistence and 41% nonadherence at least once. Objective: To model the persistence of and adherence to oral anticancer treatment of a patient with localized breast cancer. Design, Setting, and Participants: This cohort study was conducted using anonymous reimbursement data belonging to French female patients with breast cancer, extracted from the French Health Insurance database from January 2013 to December 2018. Data analysis was conducted from January 2021 to May 2022. Main Outcomes and Measures: The main outcome was the detection of episodes of nonpersistence and nonadherence 6 months before they happened. Adherence was defined as the ratio between the time covered by a drug purchase and the time between 2 purchases; patients were considered nonadherent if the ratio of their next 3 purchases was less than 80%. Disparities in persistence and adherence based on criteria such as age, treatment type, and income were identified. Results: A total of 229 695 female patients (median [IQR] age, 63 [52-72] years) with localized breast cancer were included. A deep learning model based on a gated-recurrent unit architecture was used to detect episodes of nonpersistence or nonadherence. This model demonstrated an area under the receiving operating curve of 0.71 for persistence and 0.73 for adherence. Analyzing the Shapley Additive Explanations values also gave insights into the contribution of the different features over the model's decision. Patients older than 70 years, with past nonadherence, taking more than 1 treatment in the previous 3 months, and with low income had greater risk of episodes of nonpersistence. Age and past nonadherence, including regularity of past adherence, were also important features in the nonadherence model. Conclusions and Relevance: This cohort study found associations of patient age and past adherence with nonpersistence or nonadherence. It also suggested that regular intervals in treatment purchases enhanced adherence, in contrast to irregular purchasing patterns. This research offers valuable tools for improving persistence of and adherence to oral anticancer treatment among patients with early breast cancer.

Human locomotion with reinforcement learning using bioinspired reward reshaping strategies.

Recent learning strategies such as reinforcement learning (RL) have favored the transition from applied artificial intelligence to general artificial intelligence. One of the current challenges of RL in healthcare relates to the development of a controller to teach a musculoskeletal model to perform dynamic movements. Several solutions have been proposed. However, there is still a lack of investigations exploring the muscle control problem from a biomechanical point of view. Moreover, no studies using biological knowledge to develop plausible motor control models for pathophysiological conditions make use of reward reshaping. Consequently, the objective of the present work was to design and evaluate specific bioinspired reward function strategies for human locomotion learning within an RL framework. The deep deterministic policy gradient (DDPG) method for a single-agent RL problem was applied. A 3D musculoskeletal model (8 DoF and 22 muscles) of a healthy adult was used. A virtual interactive environment was developed and simulated using opensim-rl library. Three reward functions were defined for walking, forward, and side falls. The training process was performed with Google Cloud Compute Engine. The obtained outcomes were compared to the NIPS 2017 challenge outcomes, experimental observations, and literature data. Regarding learning to walk, simulated musculoskeletal models were able to walk from 18 to 20.5 m for the best solutions. A compensation strategy of muscle activations was revealed. Soleus, tibia anterior, and vastii muscles are main actors of the simple forward fall. A higher intensity of muscle activations was also noted after the fall. All kinematics and muscle patterns were consistent with experimental observations and literature data. Regarding the side fall, an intensive level of muscle activation on the expected fall side to unbalance the body was noted. The obtained outcomes suggest that computational and human resources as well as biomechanical knowledge are needed together to develop and evaluate an efficient and robust RL solution. As perspectives, current solutions will be extended to a larger parameter space in 3D. Furthermore, a stochastic reinforcement learning model will be investigated in the future in scope with the uncertainties of the musculoskeletal model and associated environment to provide a general artificial intelligence solution for human locomotion learning. Graphical abstract.