Segmentation of Pulmonary Embolism Using Deep Learning

Abstract

Pulmonary Embolism (PE) is a condition that necessitates immediate medical attention. A doctor's examination is usually used to determine the severity of PE (Pulmonary Embolism), which takes time and can result in death. A deep learning-based methodology for detecting pulmonary embolism in CT scans is suggested in this study. Deep learning algorithms are widely employed in medical imaging for improved picture interpretation because instead of requiring a set of pre-programmed instructions, computers may autonomously learn representations from massive amounts of data [1]. They can assist doctors in making rapid diagnoses, saving time and effort in the process. Deep learning algorithms use a predetermined logical structure to analyze data and come to similar conclusions as humans. Deep learning achieves this through the use of neural networks, which are multi-layered algorithms. Some of the data pre-processing that is customary in machine learning is eliminated with deep learning.