A deep learning approach for the automatic identification of the left atrium within CT scans

Alex Deakyne, Erik Gaasedelen, Paul A. Iaizzo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Recent advancements in deep learning have led to the possibility of increased performance in computer vision tools. A major development has been the usage of Convolutional Neural Networks (CNN) for automatically detecting features within a given image. Architectures such as YOLO1 have obtained incredibly high performances for the real-time detection of every-day objects within images. However to date, there have been few reports of deep learning applied to detect anatomical features within CT scans; especially those within the cardiovascular space. We propose here an automatic anatomical feature detection pipeline for identifying the features of the left atrium using a CNN. Slices of CT scans were fed into a single neural network which predicted the four bounding box coordinates that encapsulate the left atrium. The network can be optimized end-to-end and generate predictions at great speed, achieving a validation smooth L1 loss of 11.95 when predicting the left atrial bounding boxes.

Original languageEnglish (US)
Title of host publicationFrontiers in Biomedical Devices, BIOMED - 2019 Design of Medical Devices Conference, DMD 2019
PublisherAmerican Society of Mechanical Engineers (ASME)
ISBN (Electronic)9780791841037
DOIs
StatePublished - Jan 1 2019
Event2019 Design of Medical Devices Conference, DMD 2019 - Minneapolis, United States
Duration: Apr 15 2019Apr 18 2019

Publication series

NameFrontiers in Biomedical Devices, BIOMED - 2019 Design of Medical Devices Conference, DMD 2019

Conference

Conference2019 Design of Medical Devices Conference, DMD 2019
CountryUnited States
CityMinneapolis
Period4/15/194/18/19

    Fingerprint

Keywords

  • CNN
  • Computer Vision
  • Deep Learning
  • Object Detection

Cite this

Deakyne, A., Gaasedelen, E., & Iaizzo, P. A. (2019). A deep learning approach for the automatic identification of the left atrium within CT scans. In Frontiers in Biomedical Devices, BIOMED - 2019 Design of Medical Devices Conference, DMD 2019 (Frontiers in Biomedical Devices, BIOMED - 2019 Design of Medical Devices Conference, DMD 2019). American Society of Mechanical Engineers (ASME). https://doi.org/10.1115/DMD2019-3282