Loading Now

Summary of Use Of a Multiscale Vision Transformer to Predict Nursing Activities Score From Low Resolution Thermal Videos in An Intensive Care Unit, by Isaac Yl Lee et al.


Use of a Multiscale Vision Transformer to predict Nursing Activities Score from Low Resolution Thermal Videos in an Intensive Care Unit

by Isaac YL Lee, Thanh Nguyen-Duc, Ryo Ueno, Jesse Smith, Peter Y Chan

First submitted to arxiv on: 30 May 2024

Categories

  • Main: Computer Vision and Pattern Recognition (cs.CV)
  • Secondary: Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)

     Abstract of paper      PDF of paper


GrooveSquid.com Paper Summaries

GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!

Summary difficulty Written by Summary
High Paper authors High Difficulty Summary
Read the original abstract here
Medium GrooveSquid.com (original content) Medium Difficulty Summary
A novel approach to measuring caregiver workload in hospital intensive care units (ICUs) is proposed, leveraging Multiscale Vision Transformer (MViT) models trained on low-resolution thermal videos. By passively predicting the Nursing Activities Score (NAS), this method aims to improve patient care and reduce worker burnout. The MViTv2 model achieved impressive results, with an average 5-fold accuracy of 57.21%, area under the receiver operating characteristic curve (ROC AUC) of 0.865, F1 score of 0.570, and mean squared error (MSE) of 18.16.
Low GrooveSquid.com (original content) Low Difficulty Summary
In hospitals, nurses who take care of patients in intensive care units work very hard. They often feel stressed and tired because they have too much to do. One way to measure how busy these nurses are is by using a system called the Nursing Activities Score (NAS). However, this score is usually recorded manually and not often enough. Researchers are trying new ways to make measuring caregiver workload easier and more accurate. This letter proposes using special computer vision models that can learn from low-quality videos of nurses working with patients in ICUs. The model they used, called the Multiscale Vision Transformer (MViT), was trained on 458 videos taken in an ICU in Melbourne, Australia. The results show that this model is very good at predicting how busy nurses are based on what it sees in the video.

Keywords

» Artificial intelligence  » Auc  » F1 score  » Mse  » Vision transformer