Download PDFOpen PDF in browser

ECG Signal Extraction from Intensive Care Unit Monitor Videos

EasyChair Preprint no. 9578, version 2

Versions: 12history
9 pagesDate: May 30, 2023


Computer Vision (CV) application benefits the health area, notably in its applications for assistive technologies and objective and in-depth analysis of biomedical images. However, there are currently no CV resources that innovate by collecting patients' vital signs directly from the ICU (Intensive Care Unit) medical equipment panel. Thus, the present work goal was to extract the electrocardiogram (ECG) signal from ICU monitors. The approach consisted of transforming ECG monitor signals into one-dimensional digital signals by segmenting them into frames, then extracting the segmentation's upper contour. We used nine heart monitor screen recordings (videos) available on YouTube as a database. The segmentation results validate using the Dice coefficient. Two frames of every recording were validated, generating 18 validations and an average Dice of 90.02 ± 5.74. We concluded that the approach proposed can extract ECG images from videos of Intensive Care Unit monitors and transform them into a signal in the time domain. It can help future ECG assessments, via computation vision, regarding the changes in heart rhythm (arrhythmias). It can also help circumvent limitations to access the ECG in Intensive Care Units by using, for example, a simple video camera, such as those of cell phones, close to the monitor. Such an innovative approach, in turn, would allow obtaining and transmitting the signals to the computer that will be responsible for its analysis.

Keyphrases: ECG signal, image processing, Intensive Care Unit monitor, Segmentation

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
  author = {Isabela Garcia Mello E Silva and Regina Célia Coelho and Irineu Antônio Zibordi Júnior and Sophia Silvestre Camargo and Carlos Marcelo Gurjão de Godoy},
  title = {ECG Signal Extraction from Intensive Care Unit Monitor Videos},
  howpublished = {EasyChair Preprint no. 9578},

  year = {EasyChair, 2023}}
Download PDFOpen PDF in browser