BIBM 2024
369 papers
The 2024 edition of BIBM featured 357 papers focusing on the integration of deep learning with biomedical informatics. Key research themes included the development of hybrid CNN-Transformer architectures for global-local vision and strategies for bridging the domain gap in medical foundation models. Methodological trends emphasized spatially adaptive parameter maps for image restoration, weakly supervised segmentation via label enrichment, and geometric strategies for balancing rare class distributions. Researchers also addressed technical challenges such as mitigating error propagation in iterative self-training and enhancing target segmentation for spatially sparse data.
Specific applications presented this year ranged from the TourSynbio multi-modal framework for protein engineering to the localization of capsule robots in ultrasound images. Contributions included PretrainedBA for compound-protein binding affinity prediction and Image2SignalNet for extracting neuronal signals from calcium imaging. Other studies explored secure e-health record sharing and speech envelope reconstruction from EEG data. Prolific authors contributing to these advancements included Yu Guang Wang, Yiqing Shen, Wei-Long Zheng, and Yiqiang Chen. This year's proceedings reflect a continued shift toward large-scale pre-training and multi-modal modeling in computational biology and healthcare technology.
Top Research Directions
Research directions most represented by BIBM 2024 papers.
Top Authors
Impactful Papers
Papers from BIBM 2024 that shaped research clusters.
Influential
FedDP: Privacy-preserving method based on federated learning for histopathology image segmentation
influential
Collaborative Learning Across Isolated D
PROTSOLM: Protein Solubility Prediction with Multi-modal Features
influential
Integrating Evolutionary Sequences with
Molecular Graph Representation Learning Integrating Large Language Models with Domain-specific Small
influential
Grounding Language Models with External
PM2: A New Prompting Multi-modal Model Paradigm for Few-shot Medical Image Classification
influential
Downstream Adaptation via Frozen Feature
BiSeg-SAM: Weakly-Supervised Post-Processing Framework for Boosting Binary Segmentation in Segment A
influential
Restoring Local Inductive Bias to Vision
Speed-enhanced Subdomain Alignment for Long-term Stable Neural Decoding in Brain-computer Interfaces
influential
Low-Latency Architectures for Real-Time
FunduSAM: A Specialized Deep Learning Model for Enhanced Optic Disc and Cup Segmentation in Fundus I
influential
Delineating Ambiguous Boundaries in Low-
SCOP: A Sequence-Structure Contrast-Aware Framework for Protein Function Prediction
influential
Integrating Evolutionary Sequences with
Decoupling Feature Representations of Ego and Other Modalities for Incomplete Multi-modal Brain Tumo
influential
Disentangling Shared and Specific Featur
ViDTA: Enhanced Drug-Target Affinity Prediction via Virtual Graph Nodes and Attention-based Feature
influential
Fusing Graphs and Sequences for Molecula
Papers
« prev Page 8/8