Search for: All records

Abstract ObjectiveThis study aims to develop machine learning models that provide both accurate and equitable predictions of 2-year stroke risk for patients with atrial fibrillation across diverse racial groups. Materials and MethodsOur study utilized structured electronic health records (EHR) data from the All of Us Research Program. Machine learning models (LightGBM) were utilized to capture the relations between stroke risks and the predictors used by the widely recognized CHADS2 and CHA2DS2-VASc scores. We mitigated the racial disparity by creating a representative tuning set, customizing tuning criteria, and setting binary thresholds separately for subgroups. We constructed a hold-out test set that not only supports temporal validation but also includes a larger proportion of Black/African Americans for fairness validation. ResultsCompared to the original CHADS2 and CHA2DS2-VASc scores, significant improvements were achieved by modeling their predictors using machine learning models (Area Under the Receiver Operating Characteristic curve from near 0.70 to above 0.80). Furthermore, applying our disparity mitigation strategies can effectively enhance model fairness compared to the conventional cross-validation approach. DiscussionModeling CHADS2 and CHA2DS2-VASc risk factors with LightGBM and our disparity mitigation strategies achieved decent discriminative performance and excellent fairness performance. In addition, this approach can provide a complete interpretation of each predictor. These highlight its potential utility in clinical practice. ConclusionsOur research presents a practical example of addressing clinical challenges through the All of Us Research Program data. The disparity mitigation framework we proposed is adaptable across various models and data modalities, demonstrating broad potential in clinical informatics. 

Abstract Microbiome data from sequencing experiments contain the relative abundance of a large number of microbial taxa with their evolutionary relationships represented by a phylogenetic tree. The compositional and high-dimensional nature of the microbiome mediator challenges the validity of standard mediation analyses. We propose a phylogeny-based mediation analysis method called PhyloMed to address this challenge. Unlike existing methods that directly identify individual mediating taxa, PhyloMed discovers mediation signals by analyzing subcompositions defined on the phylogenic tree. PhyloMed produces well-calibrated mediation test p -values and yields substantially higher discovery power than existing methods. 

Abstract We propose an explainable approach for relation extraction that mitigates the tension between generalization and explainability by jointly training for the two goals. Our approach uses a multi-task learning architecture, which jointly trains a classifier for relation extraction, and a sequence model that labels words in the context of the relations that explain the decisions of the relation classifier. We also convert the model outputs to rules to bring global explanations to this approach. This sequence model is trained using a hybrid strategy: supervised, when supervision from pre-existing patterns is available, and semi-supervised otherwise. In the latter situation, we treat the sequence model’s labels as latent variables, and learn the best assignment that maximizes the performance of the relation classifier. We evaluate the proposed approach on the two datasets and show that the sequence model provides labels that serve as accurate explanations for the relation classifier’s decisions, and, importantly, that the joint training generally improves the performance of the relation classifier. We also evaluate the performance of the generated rules and show that the new rules are a great add-on to the manual rules and bring the rule-based system much closer to the neural models. 

We propose an explainable approach for relation extraction that mitigates the tension between generalization and explainability by jointly training for the two goals. Our approach uses a multi-task learning architecture, which jointly trains a classifier for relation extraction, and a sequence model that labels words in the context of the relation that explain the decisions of the relation classifier. We also convert the model outputs to rules to bring global explanations to this approach. This sequence model is trained using a hybrid strategy: supervised, when supervision from pre-existing patterns is available, and semi-supervised otherwise. In the latter situation, we treat the sequence model’s labels as latent variables, and learn the best assignment that maximizes the performance of the relation classifier. We evaluate the proposed approach on the two datasets and show that the sequence model provides labels that serve as accurate explanations for the relation classifier’s decisions, and, importantly, that the joint training generally improves the performance of the relation classifier. We also evaluate the performance of the generated rules and show that the new rules are great add-on to the manual rules and bring the rule-based system much closer to the neural models. 

Three fused-ring electron acceptors ( SIDIC , DIDIC and TIDIC ) were designed and synthesized using single bond, vinylene and acetylene units linked indaceno[3,2- b ]dithiophene dimers as electron-rich cores and 3-(1,1-dicyanomethylene)-5,6-difluoro-1-indanone as electron-deficient termini. These molecules exhibit strong absorption from 550 to 900 nm with large attenuation coefficients of 1.8–2.0 × 10 5 M −1 cm −1 and high electron mobilities of 2.2–4.9 × 10 −3 cm 2 V −1 s −1 . In combination with wide-bandgap polymer FTAZ as a donor, organic solar cells exhibit efficiencies of 9.3–13.1%. Effects of the linking units on optical, electronic, morphologic, and photovoltaic properties were revealed. Relative to SIDIC , vinylene-bridged DIDIC shows red-shifted absorption, while acetylene-bridged TIDIC shows blue-shifted absorption. Compared with SIDIC and DIDIC , TIDIC has a lower HOMO, higher electron mobility, and higher device efficiency.