Search

Historically, probabilistic models for decision support have focused on discrimination, e.g., minimizing the ranking error of predicted outcomes. Unfortunately, these models ignore another important aspect, calibration, which indicates the magnitude of correctness of model predictions. Using discrimination and calibration simultaneously can be helpful for many clinical decisions. We investigated tradeoffs between these goals, and developed a unified maximum-margin method to handle them jointly. Our approach called, Doubly Optimized Calibrated Support Vector Machine (DOC-SVM), concurrently optimizes two loss functions: the ridge regression loss and the hinge loss. Experiments using three breast cancer gene-expression datasets (i.e., GSE2034, GSE2990, and Chanrion's datasets) showed that our model generated more calibrated outputs when compared to other state-of-the-art models like Support Vector Machine (p=0.03, p=0.13, and p

Seawater electrolysis is valuable for hydrogen production, but there are significant challenges such as severe Cl– corrosion and competition reaction of the chlorine evolution reaction (CER) due to high Cl– concentrations. Here, a core–shell structure was developed on the nickel foam substrate, consisting of a sulfur-modified amorphous Fe-(OH)3 layer on top of a crossing NiSe nanowire (named S–Fe-(OH)3/NiSe/NF). The S–Fe-(OH)3/NiSe/NF electrode demonstrates outstanding catalytic performance for both the hydrogen evolution reaction (HER) and the oxygen evolution reaction (OER) in simulated and natural alkaline seawater electrolytes. The overpotentials at 100 mA/cm2 for the OER in simulated and natural alkaline seawater electrolytes are 234 and 232 mV, respectively. For HER, the values are 331 and 341 mV, respectively, at a current density of 100 mA/cm2. When S–Fe-(OH)3/NiSe/NF serves as both the anode and cathode, the electrolyzer demonstrates excellent performance with voltages of 1.85 and 1.87 V at 100 mA/cm2 in simulated and natural seawater electrolytes, respectively. This electrolyzer holds significant promise for practical seawater electrolysis. [ABSTRACT FROM AUTHOR]

Background: The progressive cognitive decline, an integral component of Alzheimer's disease (AD), unfolds in tandem with the natural aging process. Neuroimaging features have demonstrated the capacity to distinguish cognitive decline changes stemming from typical brain aging and AD between different chronological points. Objective: To disentangle the normal aging effect from the AD-related accelerated cognitive decline and unravel its genetic components using a neuroimaging-based deep learning approach. Methods: We developed a deep-learning framework based on a dual-loss Siamese ResNet network to extract fine-grained information from the longitudinal structural magnetic resonance imaging (MRI) data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. We then conducted genome-wide association studies (GWAS) and post-GWAS analyses to reveal the genetic basis of AD-related accelerated cognitive decline. Results: We used our model to process data from 1,313 individuals, training it on 414 cognitively normal people and predicting cognitive assessment for all participants. In our analysis of accelerated cognitive decline GWAS, we identified two genome-wide significant loci: APOE locus (chromosome 19 p13.32) and rs144614292 (chromosome 11 p15.1). Variant rs144614292 (G > T) has not been reported in previous AD GWA studies. It is within the intronic region of NELL1, which is expressed in neurons and plays a role in controlling cell growth and differentiation. The cell-type-specific enrichment analysis and functional enrichment of GWAS signals highlighted the microglia and immune-response pathways. Conclusions: Our deep learning model effectively extracted relevant neuroimaging features and predicted individual cognitive decline. We reported a novel variant (rs144614292) within the NELL1 gene. [ABSTRACT FROM AUTHOR]

Long-term excessive application of mineral fertilizer leads to phosphorus (P) accumulation, increasing the risk of P migration and loss from the soil profile. The colloids in the soil profile are important carriers for P migration due to their high P adsorption and transport capacity. It is not clearly understood how colloidal P (CP) is distributed in subsoils (<1.2 m) of a Vertisol, contributing to subsurface P loss. Understanding the depth sequence distribution and speciation of colloidal P in the soil profile is critical for a comprehensive assessment of P loss. In this study, water-extractable colloids (WECs) with the size of 0.35–2 µ m were obtained from a 0–120 cm soil profile by a sedimentation and centrifugation scheme. The dissolved reactive P (DRP) and dissolved total P (DTP) in soil supernatant with particle sizes <0.35 µ m were measured by molybdate blue colorimetry. Solution 31 P nuclear magnetic resonance (NMR) and P K-edge XANES (X-ray absorption near-edge structure) were used to characterize the species and distribution of CP in the soil profile of fertilized farmland. Total and available P in bulk soil and colloids decreased with soil depth. The organic P (OP) contained 97–344 mg kg -1 per bulk soil and 110–630 mg kg -1 per WEC. The OP in soil profile consists of orthophosphate mono-esters and diesters primarily according to NMR results. It suggested that OP in WECs from subsoils might be affected by the translocation of CP from surface soils, probably due to soil acidification and preferential flow caused by swelling–shrinkage clays, including montmorillonite and nontronite detected by X-ray powder diffractometer (XRD) results. Additionally, the more negative zeta potential of surface soil colloids suggests the high mobility of colloidal P towards the subsoils. The CP concentration for <2 µ m was about 38–93 mg P kg -1 per bulk soil, which is 6–37 times that of DRP, suggesting that CP plays a dominant role in P transport within the soil profile. The relatively small fraction of orthophosphate diesters suggests limited P assimilation by microorganisms for the accumulation of WECs containing organically bound P in subsoils. The P K-edge XANES results indicated that the proportions of Al-P, Fe-P, and inositol hexakisphosphate (IHP) of WECs decreased, but hydroxyapatite (HAP) increased with soil depth. This study showed that inorganic and organic P migrated from the surface to deeper layers along the soil profile, with soil colloids having a significant effect on P migration from both surface and subsurface layers. The findings have an important significance for soil P migration evaluation and agricultural non-point source pollution control in Vertisols. [ABSTRACT FROM AUTHOR]

Background Alzheimer's disease (AD) is a progressive neurological disorder with no specific curative medications. Sophisticated clinical skills are crucial to optimize treatment regimens given the multiple coexisting comorbidities in the patient population. Objective Here, we propose a study to leverage reinforcement learning (RL) to learn the clinicians' decisions for AD patients based on the longitude data from electronic health records. Methods In this study, we selected 1736 patients from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. We focused on the two most frequent concomitant diseases—depression, and hypertension, thus creating 5 data cohorts (ie, Whole Data, AD, AD-Hypertension, AD-Depression, and AD-Depression-Hypertension). We modeled the treatment learning into an RL problem by defining states, actions, and rewards. We built a regression model and decision tree to generate multiple states, used six combinations of medications (ie, cholinesterase inhibitors, memantine, memantine-cholinesterase inhibitors, hypertension drugs, supplements, or no drugs) as actions, and Mini-Mental State Exam (MMSE) scores as rewards. Results Given the proper dataset, the RL model can generate an optimal policy (regimen plan) that outperforms the clinician's treatment regimen. Optimal policies (ie, policy iteration and Q-learning) had lower rewards than the clinician's policy (mean −3.03 and −2.93 vs. −2.93, respectively) for smaller datasets but had higher rewards for larger datasets (mean −4.68 and −2.82 vs. −4.57, respectively). Conclusions Our results highlight the potential of using RL to generate the optimal treatment based on the patients' longitude records. Our work can lead the path towards developing RL-based decision support systems that could help manage AD with comorbidities. [ABSTRACT FROM AUTHOR]