Water resistant, biodegradable and flexible corn starch/carboxymethyl cellulose composite film for slow-release fertilizer coating materials.

The inherent limitations of Cornstarch (CS) and Carboxymethyl Cellulose (CMC) membranes, such as brittleness, fragility, and water solubility, limit their use in controlled-release fertilizers. This study reports on the synthesis of crosslinked CMC/CS-20-E composite membranes using the casting technique, with epichlorohydrin (ECH) as the crosslinking agent in an acidic environment to crosslink CS and CMC. The synthesized composite film demonstrates remarkable water resistance, as evidenced by the insignificant alteration in its morphology and structure post 72 h of water immersion. Its flexibility is reflected in its capacity to endure knotting and bending, with an elongation at break reaching 78.1 %. Moreover, the degradation rate surpasses 90 % within a span of seven days. The CMC/CS-20-E-x-urea controlled-release fertilizer was subsequently produced using a layer-by-layer self-assembly technique, where urea particles were incorporated into the crosslinked composite solution. This CMC/CS-20-E-x-urea controlled-release fertilizer displayed superior controlled-release performance over a duration of seven days when juxtaposed with pure urea. In particular, the CMC/CS-20-E-3 %-urea controlled-release fertilizer showed a cumulative release rate of 84 % by the seventh day. The controlled-release fertilizers developed in this study offer a promising strategy for creating eco-friendly options that are crucial for fertilizing crops with short growth cycles.

Hierarchical reinforcement learning for automatic disease diagnosis.

MOTIVATION: Disease diagnosis-oriented dialog system models the interactive consultation procedure as the Markov decision process, and reinforcement learning algorithms are used to solve the problem. Existing approaches usually employ a flat policy structure that treat all symptoms and diseases equally for action making. This strategy works well in a simple scenario when the action space is small; however, its efficiency will be challenged in the real environment. Inspired by the offline consultation process, we propose to integrate a hierarchical policy structure of two levels into the dialog system for policy learning. The high-level policy consists of a master model that is responsible for triggering a low-level model, the low-level policy consists of several symptom checkers and a disease classifier. The proposed policy structure is capable to deal with diagnosis problem including large number of diseases and symptoms. RESULTS: Experimental results on three real-world datasets and a synthetic dataset demonstrate that our hierarchical framework achieves higher accuracy and symptom recall in disease diagnosis compared with existing systems. We construct a benchmark including datasets and implementation of existing algorithms to encourage follow-up researches. AVAILABILITY AND IMPLEMENTATION: The code and data are available from https://github.com/FudanDISC/DISCOpen-MedBox-DialoDiagnosis. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.