Results 91 to 100 of about 111,792 (225)

Multimodal Cross‐Attentive Graph‐Based Framework for Predicting In Vivo Endocrine Disruptors

Advanced Science, EarlyView.
A multimodal cross‐attentive graph neural network integrates molecular graphs with androgen and estrogen adverse outcome pathway (AOP)–anchored in vitro assay signals to predict in vivo endocrine disruption. By fusing information on Tier‐1 AOP logits with chemical structures, the framework achieves high accuracy and provides assay‐traceable ...
Eder Soares de Almeida Santos, Gustavo Felizardo Santos Sandes, Artur Christian Garcia da Silva, Holli‐Joi Martin, Eugene N. Muratov, Rodolpho de Campos Braga, Bruno Junior Neves   +6 more
wiley   +1 more source

Intrinsic PPG–ECG Coupling for Accurate and Low‐Power Blood Pressure Monitoring

Advanced Science, EarlyView.
A PPG–ECG coupling strategy for continuous blood pressure monitoring that intrinsically synchronizes signals within a single waveform is demonstrated, minimizing synchronization errors and hardware complexity. This approach halves power consumption while maintaining high accuracy, enabling compact, energy‐efficient wearable devices for personalized ...
Sitong Chen, Hua Luo, Zhentao Yao, Zhou Jiang, Xiaodong Wu, Hanmin Liu   +5 more
wiley   +1 more source

Multi‐Scale Mapping of Gene Expression from Whole‐slide Images for Identifying Phenotype‐Associated Subpopulations

Advanced Science, EarlyView.
BiSCALE: A pathology‐driven deep learning framework for multi‐scale gene expression prediction from whole‐slide images. It accurately infers bulk and near‐cellular spot‐level expression, links predictions to clinical phenotypes, identifies disease‐associated niches, and enables applications in risk stratification and cell‐identity annotation, providing
Hailong Zheng, Jiajing Xie, Luqi Wang, Hailong Xie, Tong Zhi, Yujia Guo, Bofeng Zhu, Dong Wang, Yu Chen   +8 more
wiley   +1 more source

A hybrid optimization and data-driven approach to understand the role of the risk-aversion profile parameter in portfolio optimization problems with shorting constraints

Operations Research Perspectives
This study contributes to the optimization literature with an approach that would help investors understand how the risk-aversion profile hyperparameter affects excess returns, risk, and Sharpe ratio curves in portfolio optimization problems with short ...
Mariano Carbonero-Ruz, Francisco Fernández-Navarro, Antonio M. Durán-Rosal, Javier Pérez-Rodríguez   +3 more
doaj   +1 more source

Rapid Proteome‐Wide Discovery of Protein–Protein Interactions With ppIRIS

Advanced Science, EarlyView.
ppIRIS is a lightweight deep learning framework for proteome‐wide protein–protein interaction prediction directly from sequence. By fusing evolutionary and structural embeddings with a regularized Siamese architecture, ppIRIS achieves state‐of‐the‐art accuracy across species, enables minute‐scale screening, and reveals biologically validated bacterial ...
Luiz Felipe Piochi, Di Tang, Johan Malmström, Yasaman Karami, Hamed Khakzad   +4 more
wiley   +1 more source

A Scalable Framework for Comprehensive Typing of Polymorphic Immune Genes from Long‐Read Data

Advanced Science, EarlyView.
SpecImmune introduces a unified computational framework optimized for long‐read sequencing to resolve over 400 highly polymorphic immune genes. This scalable approach achieves high‐resolution typing, enabling the discovery of cross‐family co‐evolutionary networks and population‐specific diversity.
Shuai Wang, Xuedong Wang, Mengyao Wang, Qian Zhou, Lusheng Wang, Shuai Cheng Li   +5 more
wiley   +1 more source

Leveraging Artificial Intelligence and Large Language Models for Cancer Immunotherapy

Advanced Science, EarlyView.
Cancer immunotherapy faces challenges in predicting treatment responses and understanding resistance mechanisms. Artificial intelligence (AI) and machine learning (ML) offer powerful solutions for cancer immunotherapy in patient stratification, biomarker discovery, treatment strategy optimization, and foundation model development.
Xinchao Wu, Mengtao Sun, Lusheng Li, Jieqiong Wang, Shibiao Wan   +4 more
wiley   +1 more source

Comparative Analysis of Hyperparameter Optimization Using Optuna and Hyperopt for Convolutional Neural Networks

Современные информационные технологии и IT-образование
The process of training a neural network model is controlled by selecting optimal hyperparameters, which have a significant impact on its quality and performance. This impact has been confirmed both theoretically and empirically by numerous studies.
Tatyana Samoilova
doaj   +1 more source

Deep Brain Stimulation Induces Antidepressant Effects by Restoring High‐Fidelity Communication in the BNST‐NAc Circuit

Advanced Science, EarlyView.
This cross‐species study reveals that pathological hyperactivity of BNST neurons in depressive states disrupts inhibitory period and isolated spikes in the BNST‐NAc circuit. DBS achieves its antidepressant effects by precisely restoring network inhibitory periods and high‐fidelity signal transmission.
Xin Lv, Yu Cao, Yuhan Wang, Yunhao Wu, Yingying Zhang, Kuanghao Ye, Xian Qiu, Qingfang Sun, Liuguan Bian, Halimureti Paerhati, Valerie Voon, Shikun Zhan, Bomin Sun   +12 more
wiley   +1 more source

The Implementation of Bayesian Optimization for Automatic Parameter Selection in Convolutional Neural Network for Lung Nodule Classification

Jurnal Nasional Pendidikan Teknik Informatika (JANAPATI)
Lung cancer's high mortality rate makes early detection crucial. Machine learning techniques, especially convolutional neural networks (CNN), play a very important role in lung nodule detection.
Kadek Eka Sapta Wijaya, Gede Angga Pradipta, Dadang Hermawan   +2 more
doaj   +1 more source