Results 121 to 130 of about 115,093 (312)

Contrastive Learning with Synthetic Positives

Contrastive learning with the nearest neighbor has proved to be one of the most efficient self-supervised learning (SSL) techniques by utilizing the similarity of multiple instances within the same class. However, its efficacy is constrained as the nearest neighbor algorithm primarily identifies "easy" positive pairs, where the representations are ...
Zeng, Dewen, Wu, Yawen, Hu, Xinrong, Xu, Xiaowei, Shi, Yiyu   +4 more
openaire   +2 more sources

Remote Monitoring in Myasthenia Gravis: Exploring Symptom Variability

Annals of Clinical and Translational Neurology, EarlyView.
ABSTRACT Background Myasthenia gravis (MG) is a rare, autoimmune disorder characterized by fluctuating muscle weakness and potential life‐threatening crises. While continuous specialized care is essential, access barriers often delay timely interventions. To address this, we developed MyaLink, a telemedical platform for MG patients.
Maike Stein, Maximilian Mönch, Meret Herdick, Frauke Stascheit, Sarah Hoffmann, Hannah Preßler, Philipp Mergenthaler, Carla Dusemund, Paolo Doksani, Haoqi Sun, Pushpa Narayanaswami, Andreas Meisel, Lea Gerischer, Sophie Lehnerer   +13 more
wiley   +1 more source

Applying an Ethical Lens to the Treatment of People With Multiple Sclerosis

Annals of Clinical and Translational Neurology, EarlyView.
ABSTRACT The practice of neurology requires an understanding of clinical ethics for decision‐making. In multiple sclerosis (MS) care, there are a wide range of ethical considerations that may arise. These involve shared decision‐making around selection of a disease‐modifying therapy (DMT), risks and benefits of well‐studied medications in comparison to
Methma Udawatta, Farrah J. Mateen
wiley   +1 more source

Self-Supervised ECG Anomaly Detection Based on Time-Frequency Specific Waveform Mask Feature Fusion

IEEE Access
The imbalance of ECG signal data and the complexity of labeling pose significant challenges for deep learning-based anomaly detection. Traditional contrastive learning approaches for ECG anomaly detection often rely on reconstruction or generation ...
Chongrui Tian, Fengbin Zhang
doaj   +1 more source

CLF-AIAD: A Contrastive Learning Framework for Acoustic Industrial Anomaly Detection [PDF]

, 2023
Zhaoyi Liu, Yuanbo Hou, Haoyu Tang, Álvaro López-Chilet, Sam Michiels, Dick Botteldooren, Jon Ander Gómez, Danny Hughes   +7 more
openalex   +1 more source

Visualizing and Understanding Contrastive Learning

IEEE Transactions on Image Processing
Accepted to IEEE Transactions on Image ...
Fawaz Sammani, Boris Joukovsky, Nikos Deligiannis   +2 more
openaire   +4 more sources

Development of a Prediction Model for Progression Risk in High‐Grade Gliomas Based on Habitat Radiomics and Pathomics

Annals of Clinical and Translational Neurology, EarlyView.
ABSTRACT Objective To investigate the value of constructing models based on habitat radiomics and pathomics for predicting the risk of progression in high‐grade gliomas. Methods This study conducted a retrospective analysis of preoperative magnetic resonance (MR) images and pathological sections from 72 patients diagnosed with high‐grade gliomas (52 ...
Yuchen Zhu, Yuxi Gong, Weilin Xu, Xingjian Sun, Gefei Jiang, Lei Qiu, Kexin Shi, Mengxing Wu, Yinjiao Fei, Jinling Yuan, Jinyan Luo, Yurong Li, Yuandong Cao, Minhong Pan, Shu Zhou   +14 more
wiley   +1 more source

Contrastive explanations for machine learning predictions in chemistry

Journal of Cheminformatics
The concept of contrastive explanations originating from human reasoning is used in explainable artificial intelligence. In machine learning, contrastive explanations relate alternative prediction outcomes to each other involving the identification of ...
Alec Lamens, Jürgen Bajorath
doaj   +1 more source

Remote Assessment of Ataxia Severity in SCA3 Across Multiple Centers and Time Points

Annals of Clinical and Translational Neurology, EarlyView.
ABSTRACT Objective Spinocerebellar ataxia type 3 (SCA3) is a genetically defined ataxia. The Scale for Assessment and Rating of Ataxia (SARA) is a clinician‐reported outcome that measures ataxia severity at a single time point. In its standard application, SARA fails to capture short‐term fluctuations, limiting its sensitivity in trials.
Marcus Grobe‐Einsler, Vivian Maas, Arian Taheri Amin, Jennifer Faber, Tamara Schaprian, Katharina Hill, Matthis Synofzik, Lisa H. Graf, Heike Jacobi, Manuela Lima, Ana F. Ferreira, Bart P. van de Warrenburg, Ilse H. J. Willemse, Dagmar Timmann, Andreas Thieme, Paola Giunti, Hector Garcia‐Moreno, Magda M. Santana, Jeannette Hübener‐Schmid, Elin H. Davies, Thomas Klockgether   +20 more
wiley   +1 more source

Rethinking spatial-temporal contrastive learning for Urban traffic flow forecasting: multi-level augmentation framework

Complex & Intelligent Systems
Graph neural networks integrating contrastive learning have attracted growing attention in urban traffic flow forecasting. However, most existing graph contrastive learning methods do not perform well in capturing local–global spatial dependencies or ...
Lin Pan, Qianqian Ren, Zilong Li, Xingfeng Lv   +3 more
doaj   +1 more source