Explainable machine learning for neoplasms diagnosis via electrocardiograms: an externally validated study

Abstract Background Neoplasms are a major cause of mortality globally, where early diagnosis is essential for improving outcomes. Current diagnostic methods are often invasive, expensive, and inaccessible in resource-limited settings. This study explores the potential of electrocardiogram (ECG) data...

Full description

Saved in:
Bibliographic Details
Main Authors: Juan Miguel Lopez Alcaraz, Wilhelm Haverkamp, Nils Strodthoff
Format: Article
Language:English
Published: BMC 2025-07-01
Series:Cardio-Oncology
Subjects:
Online Access:https://doi.org/10.1186/s40959-025-00370-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849389130812751872
author Juan Miguel Lopez Alcaraz
Wilhelm Haverkamp
Nils Strodthoff
author_facet Juan Miguel Lopez Alcaraz
Wilhelm Haverkamp
Nils Strodthoff
author_sort Juan Miguel Lopez Alcaraz
collection DOAJ
description Abstract Background Neoplasms are a major cause of mortality globally, where early diagnosis is essential for improving outcomes. Current diagnostic methods are often invasive, expensive, and inaccessible in resource-limited settings. This study explores the potential of electrocardiogram (ECG) data, a widely available and non-invasive tool for diagnosing neoplasms through cardiovascular changes linked to neoplastic presence. Methods A diagnostic pipeline combining tree-based machine learning models with Shapley value analysis for explainability was developed. The model was trained and internally validated on a large dataset and externally validated on an independent cohort to ensure robustness and generalizability. Key ECG features contributing to predictions were identified and analyzed. Results The model achieved high diagnostic accuracy in both internal testing and external validation cohorts. Shapley value analysis highlighted significant ECG features, including novel predictors. The approach is cost-effective, scalable, and suitable for resource-limited settings, offering insights into cardiovascular changes associated with neoplasms and their therapies. Conclusions This study demonstrates the feasibility of using ECG signals and machine learning for non-invasive neoplasm diagnosis. By providing interpretable insights into cardio-neoplasm interactions, this method addresses gaps in diagnostics and supports integration into broader diagnostic and therapeutic frameworks.
format Article
id doaj-art-8f2ffad84e8d45f788a1c03d0a23b3fb
institution Kabale University
issn 2057-3804
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series Cardio-Oncology
spelling doaj-art-8f2ffad84e8d45f788a1c03d0a23b3fb2025-08-20T03:42:03ZengBMCCardio-Oncology2057-38042025-07-0111111910.1186/s40959-025-00370-1Explainable machine learning for neoplasms diagnosis via electrocardiograms: an externally validated studyJuan Miguel Lopez Alcaraz0Wilhelm Haverkamp1Nils Strodthoff2AI4Health Division, Carl von Ossietzky Universität OldenburgDepartment of Cardiology, Angiology and Intensive Care Medicine, Charité Campus Mitte, German Heart Center of the Charité-University MedicineAI4Health Division, Carl von Ossietzky Universität OldenburgAbstract Background Neoplasms are a major cause of mortality globally, where early diagnosis is essential for improving outcomes. Current diagnostic methods are often invasive, expensive, and inaccessible in resource-limited settings. This study explores the potential of electrocardiogram (ECG) data, a widely available and non-invasive tool for diagnosing neoplasms through cardiovascular changes linked to neoplastic presence. Methods A diagnostic pipeline combining tree-based machine learning models with Shapley value analysis for explainability was developed. The model was trained and internally validated on a large dataset and externally validated on an independent cohort to ensure robustness and generalizability. Key ECG features contributing to predictions were identified and analyzed. Results The model achieved high diagnostic accuracy in both internal testing and external validation cohorts. Shapley value analysis highlighted significant ECG features, including novel predictors. The approach is cost-effective, scalable, and suitable for resource-limited settings, offering insights into cardiovascular changes associated with neoplasms and their therapies. Conclusions This study demonstrates the feasibility of using ECG signals and machine learning for non-invasive neoplasm diagnosis. By providing interpretable insights into cardio-neoplasm interactions, this method addresses gaps in diagnostics and supports integration into broader diagnostic and therapeutic frameworks.https://doi.org/10.1186/s40959-025-00370-1Neoplasm diagnosisElectrocardiogram (ECG)Explainable Artificial Intelligence (XAI)Machine Learning
spellingShingle Juan Miguel Lopez Alcaraz
Wilhelm Haverkamp
Nils Strodthoff
Explainable machine learning for neoplasms diagnosis via electrocardiograms: an externally validated study
Cardio-Oncology
Neoplasm diagnosis
Electrocardiogram (ECG)
Explainable Artificial Intelligence (XAI)
Machine Learning
title Explainable machine learning for neoplasms diagnosis via electrocardiograms: an externally validated study
title_full Explainable machine learning for neoplasms diagnosis via electrocardiograms: an externally validated study
title_fullStr Explainable machine learning for neoplasms diagnosis via electrocardiograms: an externally validated study
title_full_unstemmed Explainable machine learning for neoplasms diagnosis via electrocardiograms: an externally validated study
title_short Explainable machine learning for neoplasms diagnosis via electrocardiograms: an externally validated study
title_sort explainable machine learning for neoplasms diagnosis via electrocardiograms an externally validated study
topic Neoplasm diagnosis
Electrocardiogram (ECG)
Explainable Artificial Intelligence (XAI)
Machine Learning
url https://doi.org/10.1186/s40959-025-00370-1
work_keys_str_mv AT juanmiguellopezalcaraz explainablemachinelearningforneoplasmsdiagnosisviaelectrocardiogramsanexternallyvalidatedstudy
AT wilhelmhaverkamp explainablemachinelearningforneoplasmsdiagnosisviaelectrocardiogramsanexternallyvalidatedstudy
AT nilsstrodthoff explainablemachinelearningforneoplasmsdiagnosisviaelectrocardiogramsanexternallyvalidatedstudy