Key result
The TimeSformer AI model identified HFpEF with 100% sensitivity and 61.9% specificity (87.69% accuracy), and successfully detected incipient heart failure in 24 of 50 asymptomatic controls.
Why the study?
The authors aimed to develop a deep learning model capable of detecting early signs of heart failure in asymptomatic patients with a normal ejection fraction.
Does a TimeSformer AI model analyzing single A4C echocardiography videos accurately detect HFpEF and incipient heart failure in asymptomatic patients compared to standard clinical diagnosis?
Case-Control (n=130)
No
Does a TimeSformer AI model analyzing single A4C echocardiography videos accurately detect HFpEF and incipient heart failure in asymptomatic patients compared to standard clinical diagnosis?
Effect estimate: Accuracy 87.69% (95% CI 80.78% to 92.80%)
An AI video transformer model analyzing a single echocardiography clip can accurately detect HFpEF and identify asymptomatic individuals at risk of developing incident heart failure.
May enable earlier HF detection in asymptomatic patients with preserved EF; hypothesis-generating and requires prospective validation.
Background Recently deep learning models have helped differentiate echocardiography images of patients with heart failure with preserved ejection fraction (HFpEF) from normal controls. Our aim was to develop a model capable of detecting early signs of heart failure in asymptomatic patients with a normal ejection fraction. Methods We employed the TimeSformer, a video transformer model that classifies video data using a novel attentionbased mechanism. This self-attention mechanism that diverges from traditional convolutional neural networks (CNNs). It focuses on relevant parts of the video across both space and time, split into spatial attention, which processes each frame individually, and temporal attention, which integrates information across different frames. The training and validation of the TimeSformer model were conducted on the same dataset of echocardiography videos from 50 normal controls and 80 patients diagnosed with HFpEF, employing 5-fold cross-validation to ensure robust performance evaluation. Results The TimeSformer model effectively identified HFpEF in patients, as all diagnosed with HFpEF were flagged as abnormal. The echocardiography assessment, along with NT pro BNP levels, supported the diagnoses, with patients showing NT pro BNP levels of 1016±32 pg/ml. Conversely, 26 out of 50 normal controls were correctly identified. While 24 normal controls were identified as abnormal. Their NT pro BNP levels were 52±12 pg/ml. At 2 years 8 (33.3%) controls identified as abnormal developed symptoms of heart failure with NT pro BNP levels of 800±41 pg/ml. Conclusions The TimeSformer model demonstrated capability in identifying subtle deviations indicative of incipient heart failure in videos of normal controls, despite normal NT pro BNP levels. A significant number of these controls developed heart failure with elevated NT pro BNP levels at 2 years.
No takes yet. Share an insight, caveat, or question.
Adhyapak et al. (2024) conducted a case-control in Heart failure with preserved ejection fraction (HFpEF) and incipient heart failure (n=130). TimeSformer AI model for echocardiography video classification vs. Clinical diagnosis (2D Echocardiography and NT pro-BNP levels) was evaluated on Diagnostic accuracy for HFpEF (Accuracy 87.69%, 95% CI 80.78% to 92.80%). The TimeSformer AI model identified HFpEF with 100% sensitivity and 61.9% specificity (87.69% accuracy), and successfully detected incipient heart failure in 24 of 50 asymptomatic controls.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: