Diagnosis of Takotsubo Syndrome by Robust Feature Selection from the Complex Latent Space of DL-Based Segmentation Network

Researchers have shown significant correlations among segmented objects in various medical imaging modalities and disease related pathologies. Several studies showed that using hand crafted features for disease prediction neglects the immense possibility to use latent features from deep learning (DL) models which may reduce the overall accuracy of differential diagnosis. However, directly using classification or segmentation models on medical to learn latent features opt out robust feature selection and may lead to overfitting. To fill this gap, we propose a novel feature selection technique using the latent space of a segmentation model that can aid diagnosis. We evaluated our method in differentiating a rare cardiac disease: Takotsubo Syndrome (TTS) from the ST elevation myocardial infarction (STEMI) using echocardiogram videos (echo). TTS can mimic clinical features of STEMI in echo and extremely hard to distinguish. Our approach shows promising results in differential diagnosis of TTS with 82% diagnosis accuracy beating the previous state-of-the-art (SOTA) approach. Moreover, the robust feature selection technique using LASSO algorithm shows great potential in reducing the redundant features and creates a robust pipeline for short- and long-term disease prognoses in the downstream analysis.

1. INTRODUCTION

Quick diagnosis and accurate treatment by giving proper medication to patients is necessary for life threatening diseases such as acute myocardial infarction (AMI). But TTS can mimic clinical and electrocardiographic (ECG) features of AMI and hard to distinguish between them using just echo. Current guidelines advocate the use of coronary angiography to direct differential diagnosis and treatment [1], which is not only invasive, but also slow in process that may endanger the patients in the emergency room [2]. Recently deep learning models are studied for the classification of TTS and ST elevation myocardial infarction (STEMI) using echo [3], [4]. In our earlier study, we demonstrated that using deep learning model as a binary classifier between the two diseases can significantly improve the detection accuracy compared to the physicians and help them make the judgement calls in difficult cases [3]. Despite having good classification accuracy, DCNN classifiers do not guarantee robust feature selection, particularly in a noisy dataset such as echo. Moreover, artifacts and speckle noise in the echos can generate irrelevant and wrong features that may reduce the overall accuracy.