All Competitors

Every biological foundation model, evaluated and ranked by the bio.rodeo team

Showing 16 of 6 filtered models

  • MELP

    291617
    The University of Hong KongJune 27, 2025contrastive_learningecg_classificationelectrocardiogram+4

    Multi-scale ECG-language pretraining model that aligns 12-lead ECG signals with clinical text at token, beat, and rhythm levels for zero-shot cardiac diagnosis.

    BiosignalsLanguage model
    68Openness
  • HeartLang

    5642
    Peking UniversityFebruary 15, 2025arrhythmia_detectionecg_classificationelectrocardiogram+5

    Self-supervised ECG foundation model that treats heartbeats as words and rhythms as sentences, using a QRS-Tokenizer and dual-level pretraining on MIMIC-IV-ECG.

    Biosignals
    78Openness
  • ECGFounder

    12432128
    Peking University +2 othersOctober 5, 2024arrhythmia_detectioncardiologycnn+6

    A 1D convolutional foundation model for electrocardiogram analysis, trained on over 10 million expert-annotated recordings across 150 diagnostic categories.

    Biosignals
    75Openness
  • D-BETA

    331323
    Singapore Management University +1 otherOctober 3, 2024autoencodercontrastive_learningecg_classification+6

    Contrastive masked ECG-text auto-encoder pretrained on paired electrocardiograms and clinical reports, enabling label-efficient and zero-shot cardiac diagnosis.

    BiosignalsLanguage model
    27Openness
  • ECG-JEPA

    1510
    Zuse Institute Berlin +1 otherOctober 2, 2024ecg_classificationelectrocardiogramfoundation_model+4

    A Vision-Transformer Joint-Embedding Predictive Architecture self-supervised on 1M+ ECG records, improving 12-lead ECG classification on PTB-XL.

    Biosignals
    62Openness
  • HeartBEiT

    25111
    Icahn School of Medicine at Mount SinaiJune 6, 2023cardiologydisease_diagnosisecg_classification+4

    A BEiT vision transformer pretrained on 8.5M 12-lead ECG images via masked image modeling, excelling at low-data-regime cardiac diagnosis.

    Biosignals
    30Openness