BIBM 2023

228 papers
The 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) featured 230 papers focusing on the integration of decentralized machine learning and multimodal data analysis. A major research direction involved federated learning, specifically addressing challenges in data heterogeneity and personalized healthcare AI. Another prominent theme was the analysis of brain connectomes through spatio-temporal graph learning and dynamic functional connectivity, exemplified by methods like Chat2Brain for mapping semantic queries to brain activation maps. Research also emphasized multimodal integration, such as grounding clinical language in medical imaging and combining text and images for depression detection. Methodological advancements included joint CNN-GNN frameworks for Alzheimer’s diagnosis and deep-learning optimization for molecular property prediction. Prolific contributors to this year's proceedings included May D. Wang, Wenqi Shi, and Tianming Liu. The collective work reflected a trend toward robust medical imaging across clinical sites and the application of knowledge graphs for clinical reasoning, alongside specialized studies in spatial transcriptomics and longitudinal reasoning.

Top Research Directions

Research directions most represented by BIBM 2023 papers.

The Scalability Gap in Knowledge Graph Reasoning 1
Local Prediction Stability for Uncertainty Estimation 1
Guiding Neural Networks with Structured Domain Knowledge 1
Asymmetric Quantization for Skewed Activation Distributions 1
Semantic Collision and False Negatives in Contrastive Learning 1
Learning Representations Directly from Raw Physiological Waveforms 1
Grounding Clinical Models with Structured Knowledge Graphs 1
Aligning Predictive Confidence with Empirical Accuracy 1
Feature Disentanglement in Parallel Attention Subspaces 1
Scaling Training Data via Distant Supervision 1
Learning Similarity Metrics with Twin Networks 1
Masked Patch Reconstruction for Vision Transformers 1
Data Augmentation in the Latent Space 1
Overcoming Calibration Data Sensitivity in Model Compression 1
Multi Domain Feature Engineering for Time Series Analysis 1

Top Authors

AuthorPapers
Wenqi Shi7
Jin Liu6
Lei Deng5
Xuequn Shang5
Jianqiang Li4
Lei Xu3
Li Shen3
Shikui Tu3
Tianming Liu3
Alexander G. Ioannidis2
Bin Xiao2
Chenhao Lin2
Chong Ma2
Daniel Mas Montserrat2
Fei Qi2
Feng Tan2
Hongmin Cai2
Houliang Zhou2
Jiahui Pan2
Jiang Bian2

Impactful Papers

Papers from BIBM 2023 that shaped research clusters.

Influential

Papers

Title
MoTIF: a Method for Trustworthy Dynamic Multimodal Learning on Omics
MSCAP: DNA Methylation Age Predictor based on Multiscale Convolutional Neural Network
MTMap: A Long-Read Alignment Tool based on Multi-Core DSPs
Multi-granularity Information Flow Enhanced Skeleton-based Infant Seizure Detection
Multi-Kernel Tensor Fusion on Grassmann Manifold for Genomic Data Clustering
Multi-level Knowledge Integration with Graph Convolutional Network for Cancer Molecular Subtype Classification
Multimodal-AIR-BERT: A Multimodal Pre-trained Model for Antigen Specificity Prediction in Adaptive Immune Receptors
Multi-Modal Deep Feature Integration for Alzheimer's Disease Staging
Multi-modal Learning with Missing Modality in Predicting Axillary Lymph Node Metastasis
Multi-omics Sampling-based Graph Transformer for Synthetic Lethality Prediction
Multi-scale convolutional neural networks for brain disease diagnosis
Multi-source unsupervised domain-adaptation for automatic sleep staging
Named entity recognition of Traditional Chinese Medicine cases based on RoBERTa-BiLSTM-CRF
Non-rigid Medical Image Registration Based on Unsupervised Self-driven Prior Fusion
PACS: Prediction and analysis of cancer subtypes from multi-omics data based on a multi-head attention mechanism model
Parallel Computation of Conformational Stability for CD4+ T-cell Epitope Prediction
Parsing Clinical Trial Eligibility Criteria for Cohort Query by a Multi-Input Multi-Output Sequence Labeling Model
Pathogenesis of the salivation and prediction of traditional Chinese medicine prescription:A study based on bioinformatics and molecular docking
Patient-Centered Data Sharing and Revision Framework Based on Redactable Blockchain
Personalized COVID-19 Early Detection Using Wearable Data Based on Self-Reported Records
Personalized federated adaptive regularization for heterogeneous medical image classification: COVID-19 CT results
PESI: Paratope-Epitope Set Interaction for SARS-CoV-2 Neutralization Prediction
PEST: A General-Purpose Protein Embedding Model for Homology Search
Plug-and-Play Feature Generation for Few-Shot Medical Image Classification
PMDF: Preference-based Multimodal Deep Forest for Overall Survival Prediction in Gastric Cancer
PnP-AE: A Plug-and-Play Module for Volumetric Medical Image Segmentation
Polyp2Former: Boundary Guided Network Based on Transformer for Polyp Segmentation
Predicting Associations between circRNAs and Drug Sensitivity using Heterogeneous Graphs and Graph Attention Networks
Predicting Tertiary Lymphoid Structures in Head and Neck Contrast-Enhanced CT Images using a 3D CNN
Predicting Tumor Mutation Burden of Lung Cancer Based on Residual Network
Privacy-Preserving Publishing of Individual-Level Medical Data for Cloud Services
Protein Function Prediction with Primary-Tertiary Hierarchical Learning
PTDA-SWGCL: Predicting tRNA-Disease Associations using Supplementarily Weighted Graph Contrastive Learning
R2L-Net: Rapid Medical Image Segmentation Network Regularized by Self-Supervised Relative Localization Task
Radiology Report Generation via Structured Knowledge-Enhanced Multi-modal Attention and Contrastive Learning
Replay with Stochastic Neural Transformation for Online Continual EEG Classification
Rethinking Low-Dose CT Synthesis: Degrading Normal-Dose CT from Origin for Pairwise Training of CT Denoiser
SE(3) Equivalent Graph Attention Network as an Energy-Based Model for Protein Side Chain Conformation
SEDGCN: Sentiment Enhanced Dual Graph Convolutional Networks for Detecting Adverse Drug Reactions
Segment Anything Model (SAM) for Medical Image Segmentation: A Preliminary Review
Selecting Heterogeneous Features Based on Unified Density-Guided Neighborhood Relation for Complex Biomedical Data Analysis
Self-distillation Augmented Masked Autoencoders for Histopathological Image Understanding
Self-supervised learning for atrial fibrillation detection with ECG using CNNTransformer
Semantic-guided Unknown-aware Rare Disease Diagnosis System
Semantic-Similarity Collaborative Knowledge Distillation Framework for Whole Slide Image Classification
Semi-MedSeq: Semi-supervised Semantic Segmentation for Medical Image Sequences
Semi-supervised Medical Image Segmentation via Feature-perturbed Consistency
Sequential Representation of Sparse Heterogeneous Data for Diabetes Risk Prediction
Simultaneous Super-Resolution and Denoising on MRI via Conditional Stochastic Normalizing Flow
Spatial-Temporal Fusion Pseudo-Labeling Based Informative Frame Classification for Confocal Laser Endomicroscopy Video

« prev Page 4/5 next »