Artificial intelligence driven software systems for cardiovascular disease detection using physiological signals: a systematic review

Abstract Cardiovascular disease exacts an extraordinary global toll, with annual mortality exceeding 17.9 million lives and representing nearly one-third of all deaths worldwide. Reducing diagnostic delays, particularly in resource-limited settings, requires innovative, scalable, and cost-effective solutions. Artificial intelligence integrated with physiological signals, such as electrocardiogram (ECG), photoplethysmogram (PPG) and phonocardiogram (PCG), offers promising pathways for scalable, real-time cardiovascular assessment. Despite rapid advances, the evidence base remains fragmented, impeding identification of validated approaches and clinical translation pathways. A PRISMA-guided search across five databases (Scopus, Web of Science, PubMed, Embase, IEEE Xplore) identified 35 eligible peer-reviewed studies. Quality and risk of bias were appraised using a QUADAS-2 adapted framework. Most systems utilised ECG (71.4%) with deep learning architectures, including convolutional and recurrent neural networks. Critical gaps were quantified: only 23% of studies conducted external validation, only 14% achieved clinical-scale deployment, no study integrated principled multi-signal fusion (ECG + PPG + PCG), and only one system (1/35, 2.9%) implemented a patient-facing conversational interface. Reported accuracies ranged from 70.5 to 99.9% (n = 21 reporting; unweighted mean 93.7%), but inconsistent metric reporting and limited external validation tempered translational confidence. A three-directional framework is proposed, addressing generalisability through multi-signal (ECG + PPG + PCG) models, interpretability through LLM-driven clinical narratives, and accessibility through edge-cloud deployment, to guide future research toward clinically deployable systems.

Authors

Institutions

Publication Details

Journal
Artificial Intelligence Review
Published
2026-11-14
DOI
https://doi.org/10.1007/s10462-026-11756-7
Primary Topic
ECG Monitoring and Analysis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Artificial intelligence driven software systems for cardiovascular disease detection using physiological signals: a systematic review

Abdullah Al-Mamun Bulbul, MD Tasnim Alam
2 links
Artificial Intelligence Review
ECG Monitoring and Analysis
article

Artificial intelligence driven software systems for cardiovascular disease detection using physiological signals: a systematic review

Abdullah Al-Mamun Bulbul, MD Tasnim Alam
article en

Abstract

Abstract Cardiovascular disease exacts an extraordinary global toll, with annual mortality exceeding 17.9 million lives and representing nearly one-third of all deaths worldwide. Reducing diagnostic delays, particularly in resource-limited settings, requires innovative, scalable, and cost-effective solutions. Artificial intelligence integrated with physiological signals, such as electrocardiogram (ECG), photoplethysmogram (PPG) and phonocardiogram (PCG), offers promising pathways for scalable, real-time cardiovascular assessment. Despite rapid advances, the evidence base remains fragmented, impeding identification of validated approaches and clinical translation pathways. A PRISMA-guided search across five databases (Scopus, Web of Science, PubMed, Embase, IEEE Xplore) identified 35 eligible peer-reviewed studies. Quality and risk of bias were appraised using a QUADAS-2 adapted framework. Most systems utilised ECG (71.4%) with deep learning architectures, including convolutional and recurrent neural networks. Critical gaps were quantified: only 23% of studies conducted external validation, only 14% achieved clinical-scale deployment, no study integrated principled multi-signal fusion (ECG + PPG + PCG), and only one system (1/35, 2.9%) implemented a patient-facing conversational interface. Reported accuracies ranged from 70.5 to 99.9% (n = 21 reporting; unweighted mean 93.7%), but inconsistent metric reporting and limited external validation tempered translational confidence. A three-directional framework is proposed, addressing generalisability through multi-signal (ECG + PPG + PCG) models, interpretability through LLM-driven clinical narratives, and accessibility through edge-cloud deployment, to guide future research toward clinically deployable systems.

Artificial Intelligence Review
Griffith University (AU), Khulna University (BD), The University of Queensland (AU), Imam Mohammad ibn Saud Islamic University (SA)
Good health and well-being
ECG Monitoring and Analysis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon