Fully Automated Multi-Chamber Strain Profiling in Consecutive Patients Undergoing Transthoracic Echocardiography: Feasibility, Functional Clustering, and Prognostic Implications

Q: What are the key findings of this study?

AI-identified pan-chamber strain dysfunction was associated with higher risk of all-cause death or HF hospitalization vs preserved function (HR 3.44, 95% CI 2.43-4.88; P<0.001).

Q: What is the clinical evidence from this study?

Study design: Cohort. Population: Consecutive adults undergoing routine transthoracic echocardiography (n=8350). Intervention: Pan-chamber dysfunction (Cluster 3) on fully automated AI-derived multi-chamber strain profiling (LVGLS, LASr, RVFWS) vs. Preserved multi-chamber strain function (Cluster 1). Primary outcome: Composite of all-cause death and heart failure hospitalization (HR 3.44, 95% CI 2.43-4.88, p=<0.001).

Q: What does this research mean for the field?

AI-derived multi-chamber strain profiling identifies a pan-chamber dysfunction phenotype that is associated with a more than three-fold higher risk of all-cause mortality or heart failure hospitalization compared to preserved multi-chamber function. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Q: What question did this study set out to answer?

To evaluate the feasibility and clinical prognostic value of fully automated multi-chamber myocardial strain assessment in consecutive patients undergoing routine transthoracic echocardiography.

Su A Noh; Jiesuck Park; Yeonyee Elizabeth Yoon; Jaeik Jeon; Yeonggul Jang; Seung-Ah Lee; Minjung Bak; Hong-Mi Choi; In‐Chang Hwang; Goo‐Yeong Cho

doi:10.64898/2026.10.20.26365880

October 21, 2026Open Access

Fully Automated Multi-Chamber Strain Profiling in Consecutive Patients Undergoing Transthoracic Echocardiography: Feasibility, Functional Clustering, and Prognostic Implications

View Full Paper

Ask AI

Bookmark

Share

Key result

AI-identified pan-chamber strain dysfunction is linked to ~3.4-fold higher risk of death or HF hospitalization.

HR 3.44
95% CI 2.43-4.88
P<0.001
n=8,350

Why the study?

Clinical application of myocardial strain has largely focused on individual chambers or selected patient populations, leaving the feasibility and clinical relevance of fully automated multi-chamber strain assessment in routine practice uncertain.

Does fully automated AI-based multi-chamber myocardial strain profiling and clustering provide feasible functional characterization and incremental prognostic value for all-cause death or heart failure hospitalization in consecutive adults undergoing routine transthoracic echocardiography?

Key Points

To evaluate the feasibility and clinical prognostic value of fully automated multi-chamber myocardial strain assessment in consecutive patients undergoing routine transthoracic echocardiography.
Retrospective study of 8,350 consecutive adults undergoing routine transthoracic echocardiography evaluated with a hospital-wide artificial intelligence system.
Automatically measured left ventricular global longitudinal strain (LVGLS), left atrial reservoir strain (LASr), and right ventricular free-wall strain (RVFWS), followed by Gaussian mixture modeling to identify functional clusters.
Complete multi-chamber strain assessment was successfully achieved in 90.3% (7,539 of 8,350) of patients, with three-view LVGLS available in 94.8%, LASr in 94.1%, and RVFWS in 94.5%.
Gaussian mixture modeling identified three distinct functional clusters: preserved multi-chamber function, right ventricular-dominant dysfunction, and pan-chamber dysfunction.
Compared with preserved function, pan-chamber dysfunction had the highest risk of all-cause death or heart failure hospitalization (adjusted HR, 3.44; 95% CI, 2.43–4.88; P<0.001), remaining evident in patients with left ventricular ejection fraction ≥50%.

Study Design

Type

Cohort (n=8,350)

Multicenter

No

Structured PICO

Does fully automated AI-based multi-chamber myocardial strain profiling and clustering provide feasible functional characterization and incremental prognostic value for all-cause death or heart failure hospitalization in consecutive adults undergoing routine transthoracic echocardiography?

P

Population

8,350 consecutive adults undergoing routine transthoracic echocardiography at a tertiary hospital in South Korea (7,539 with complete multi-chamber strain profiling), followed for a median of 6.7 months.

E

Exposure

Hospital-wide artificial intelligence (AI)-based echocardiographic analysis system (Sonix Health, Ontact Health Co., Ltd.) for fully automated measurement of left ventricular global longitudinal strain (LVGLS), left atrial reservoir strain (LASr), and right ventricular free-wall strain (RVFWS), with Gaussian mixture modeling used to identify three multi-chamber strain clusters (Cluster 1: preserved multi-chamber strain; Cluster 2: RV-predominant dysfunction; Cluster 3: pan-chamber dysfunction).

C

Comparator

Conventional chamber-specific echocardiographic parameters (LVEF, LAVI, RV FAC) and comparison across AI-derived functional clusters (using Cluster 1 with preserved multi-chamber strain as the reference group).

O

Outcome

Composite of all-cause death and heart failure (HF) hospitalization at median follow-up of 6.7 months (IQR 5.6-8.1), ascertained from electronic medical records and death certificates.composite

Main Result

Hazard Ratio: 3.44 (95% CI 2.43–4.88)

p-value: p=<0.001

Fully automated AI-derived multi-chamber strain assessment is feasible in over 90% of routine echocardiograms and identifies distinct functional phenotypes, with pan-chamber dysfunction independently predicting a more than three-fold higher risk of death or heart failure hospitalization beyond conventional echocardiographic measures.

Limitations

Retrospective single-center observational design
9.7% of patients were excluded from outcome and clustering analyses due to incomplete multi-chamber strain measurements, with higher rates of heart failure and atrial fibrillation in those with unsuccessful analysis
Relatively short follow-up duration (median 6.7 months)
Retrospective, single-center study design which may limit generalizability to other institutions, populations, imaging protocols, and ultrasound systems
Short follow-up duration resulting in a limited number of events for some outcomes, particularly within the RV-dominant cluster
Multi-chamber strain analysis was not feasible in all examinations (9.7% unsuccessful), with unsuccessful analysis more common in patients with heart failure and atrial fibrillation

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

May support AI multi-chamber strain for risk stratification on routine echo; hypothesis-generating pending prospective validation.

Abstract

Aims Myocardial strain enables sensitive assessment of cardiac function, but its clinical application has largely focused on individual chambers or selected patient populations. We evaluated the feasibility and clinical relevance of fully automated multi-chamber strain assessment in consecutive patients undergoing routine transthoracic echocardiography (TTE). Methods and results This retrospective study included 8,350 consecutive adults undergoing TTE. A hospital-wide artificial intelligence (AI)–based system automatically measured left ventricular (LV) global longitudinal strain (LVGLS), left atrial (LA) reservoir strain (LASr), and right ventricular (RV) free-wall strain (RVFWS). AI-derived LVGLS was available from at least one apical view in 97.1% of patients and from all three views in 94.8%; LASr and RVFWS were available in 94.1% and 94.5%, respectively, resulting in complete multi-chamber strain assessment in 90.3%. Strain measures showed modest correlations with conventional chamber-specific echocardiographic parameters and differential associations with clinical outcomes. Gaussian mixture modeling identified three distinct clusters characterized by preserved multi-chamber function, RV-dominant dysfunction, and pan-chamber dysfunction. Compared with the preserved cluster, pan-chamber dysfunction was associated with the highest risk of all-cause death or heart failure hospitalization (adjusted hazard ratio, 3.44; 95% CI, 2.43–4.88; P<0.001). Prognostic associations remained evident in patients with LV ejection fraction (LVEF) ≥50%, and cluster assignment provided incremental prognostic information beyond clinical factors, LVEF, and RV fractional area change. Conclusions Fully automated multi-chamber strain assessment was feasible in a large consecutive TTE cohort. Integrating strain across cardiac chambers may provide complementary functional characterization and identify clinically relevant clusters with prognostic information beyond conventional assessment.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Noh et al. (2026) conducted a cohort in Consecutive adults undergoing routine transthoracic echocardiography (n=8,350). Pan-chamber dysfunction (Cluster 3) on fully automated AI-derived multi-chamber strain profiling (LVGLS, LASr, RVFWS) vs. Preserved multi-chamber strain function (Cluster 1) was evaluated on Composite of all-cause death and heart failure hospitalization (HR 3.44, 95% CI 2.43-4.88, p=<0.001). AI-identified pan-chamber strain dysfunction was associated with higher risk of all-cause death or HF hospitalization vs preserved function (HR 3.44, 95% CI 2.43-4.88; P<0.001).

synapsesocial.com/papers/6ae28e94cbadf70daf6c435c https://doi.org/10.64898/2026.10.20.26365880

View Full Paper

Ask AI

Bookmark

Share

Fully Automated Multi-Chamber Strain Profiling in Consecutive Patients Undergoing Transthoracic Echocardiography: Feasibility, Functional Clustering, and Prognostic Implications

Key Points

Study Design

Structured PICO

Main Result

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Fully Automated Multi-Chamber Strain Profiling in Consecutive Patients Undergoing Transthoracic Echocardiography: Feasibility, Functional Clustering, and Prognostic Implications

Key Points

Study Design

Structured PICO

Main Result

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading