{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/zero-shot-ecg-classification-with-multimodal","title":"Zero-Shot ECG Classification with Multimodal Learning and Test-time Clinical Knowledge Enhancement","arxiv_id":"2403.06659","date":"2024-03-11","proceeding":null,"authors":["Che Liu","Zhongwei Wan","Cheng Ouyang","Anand Shah","Wenjia Bai","Rossella Arcucci"],"abstract":"Electrocardiograms (ECGs) are non-invasive diagnostic tools crucial for detecting cardiac arrhythmic diseases in clinical practice. While ECG Self-supervised Learning (eSSL) methods show promise in representation learning from unannotated ECG data, they often overlook the clinical knowledge that can be found in reports. This oversight and the requirement for annotated samples for downstream tasks limit eSSL's versatility. In this work, we address these issues with the Multimodal ECG Representation Learning (MERL}) framework. Through multimodal learning on ECG records and associated reports, MERL is capable of performing zero-shot ECG classification with text prompts, eliminating the need for training data in downstream tasks. At test time, we propose the Clinical Knowledge Enhanced Prompt Engineering (CKEPE) approach, which uses Large Language Models (LLMs) to exploit external expert-verified clinical knowledge databases, generating more descriptive prompts and reducing hallucinations in LLM-generated content to boost zero-shot classification. Based on MERL, we perform the first benchmark across six public ECG datasets, showing the superior performance of MERL compared against eSSL methods. Notably, MERL achieves an average AUC score of 75.2% in zero-shot classification (without training data), 3.2% higher than linear probed eSSL methods with 10\\% annotated training data, averaged across all six datasets. Code and models are available at https://github.com/cheliu-computation/MERL","url_abs":"https://arxiv.org/abs/2403.06659v3","url_pdf":"https://arxiv.org/pdf/2403.06659v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"zero-shot-ecg-classification-with-multimodal","repo_url":"https://github.com/cheliu-computation/merl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"zero-shot-ecg-classification-with-multimodal","repo_url":"https://github.com/cheliu-computation/MERL-ICML2024","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"clinical-knowledge","task_name":"Clinical Knowledge"},{"task_slug":"descriptive","task_name":"Descriptive"},{"task_slug":"diagnostic","task_name":"Diagnostic"},{"task_slug":"ecg-classification","task_name":"ECG Classification"},{"task_slug":"prompt-engineering","task_name":"Prompt Engineering"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"self-supervised-learning","task_name":"Self-Supervised Learning"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"},{"task_slug":null,"task_name":"zero-shot-classification"}],"methods":[{"method_slug":"maml","method_name":"MAML"},{"method_slug":"merl","method_name":"MeRL"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2403.06659","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.06659"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/cheliu-computation/MERL-ICML2024","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/cheliu-computation/merl","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":8,"unverified":4},"by_repo_kind":{"official":{"samples":12,"ran":8,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"6d247f814f9a7559","entry":"ResNet18","repo":"cheliu-computation/merl","repo_kind":"official","path":"finetune/models/resnet1d.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/finetune/models/resnet1d.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6d247f814f9a7559"}},{"code_sha256_prefix":"4d972ff40178697e","entry":"ResNet34","repo":"cheliu-computation/merl","repo_kind":"official","path":"finetune/models/resnet1d.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/finetune/models/resnet1d.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4d972ff40178697e"}},{"code_sha256_prefix":"73b458636b1bb0bd","entry":"ResNet50","repo":"cheliu-computation/merl","repo_kind":"official","path":"finetune/models/resnet1d.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/finetune/models/resnet1d.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"73b458636b1bb0bd"}},{"code_sha256_prefix":"be1231d1e98a796f","entry":"compute_AUCs","repo":"cheliu-computation/merl","repo_kind":"official","path":"utils/zeroshot_val.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/utils/zeroshot_val.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"be1231d1e98a796f"}},{"code_sha256_prefix":"5f4556f64c743109","entry":"get_class_emd","repo":"cheliu-computation/merl","repo_kind":"official","path":"utils/zeroshot_val.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/utils/zeroshot_val.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5f4556f64c743109"}},{"code_sha256_prefix":"7f500d29d602d34e","entry":"get_ecg_emd","repo":"cheliu-computation/merl","repo_kind":"official","path":"utils/zeroshot_val.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/utils/zeroshot_val.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7f500d29d602d34e"}},{"code_sha256_prefix":"9eaf55ea5bb1d3ba","entry":"infer","repo":"cheliu-computation/merl","repo_kind":"official","path":"finetune/main_single.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/finetune/main_single.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9eaf55ea5bb1d3ba"}},{"code_sha256_prefix":"e5487ec054cd281e","entry":"precision_at_k","repo":"cheliu-computation/merl","repo_kind":"official","path":"utils/utils_loss.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/utils/utils_loss.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e5487ec054cd281e"}},{"code_sha256_prefix":"19c632192cb40d5c","entry":"clip_loss","repo":"cheliu-computation/merl","repo_kind":"official","path":"utils/utils_loss.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/utils/utils_loss.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"19c632192cb40d5c"}},{"code_sha256_prefix":"42fddfcd0b5f7db0","entry":"vit_middle","repo":"cheliu-computation/merl","repo_kind":"official","path":"finetune/models/vit1d.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/finetune/models/vit1d.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"42fddfcd0b5f7db0"}},{"code_sha256_prefix":"b86b9766958f2e32","entry":"vit_small","repo":"cheliu-computation/merl","repo_kind":"official","path":"finetune/models/vit1d.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/finetune/models/vit1d.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b86b9766958f2e32"}},{"code_sha256_prefix":"5bc84c9b81db6507","entry":"vit_tiny","repo":"cheliu-computation/merl","repo_kind":"official","path":"finetune/models/vit1d.py","file_url":"https://github.com/cheliu-computation/merl/blob/HEAD/finetune/models/vit1d.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5bc84c9b81db6507"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}