Scholarlink

<p>Sleep plays a very important role in maintaining human health, but many people neglect it, which can cause many types of sleep disorders. So, doctors use something called sleep analysis techniques to figure out these problems. It is like checking for interruptions in sleep, extreme tiredness during the day, or snoring. However, this process can be unreliable because experts might see things differently. We can use an artificial intelligence-based Program Diagnostic Tool (PDT) and knowledge systems to overcome this. AI technologies with deep learning can help doctors analyze sleep data more accurately and build real-time information diagnostic systems. In this paper, we have looked at 36 studies done between 2013 and 2020 that used deep learning to study sleep data. We found that more than half of these studies look at brain activity recordings. We also found that using only EEG data is insufficient for perfect results. We should include other PSG recordings from sleep recordings to improve the diagnosis, like EEG and EMG signals. So, we need to use a mixture of different data sources to use deep learning for diagnosing sleep problems. This paper also discusses some methods from the last ten years that show how deep learning can work with these different data types to classify sleep stages.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep Learning-Based Sleep Stage Classification: A Systematic Review

  • 期刊论文

  • 发表日期：

    2025年9月17日

查看全文
  • Santosh Kumar Satapathy,
  • Tapan Patel,
  • Madhuram Modi,
  • Harsh Davra,
  • Biswajit Brahma,
  • Santosh Kumar Tripathy,
  • Akash Kumar Bhoi

摘要

Sleep plays a very important role in maintaining human health, but many people neglect it, which can cause many types of sleep disorders. So, doctors use something called sleep analysis techniques to figure out these problems. It is like checking for interruptions in sleep, extreme tiredness during the day, or snoring. However, this process can be unreliable because experts might see things differently. We can use an artificial intelligence-based Program Diagnostic Tool (PDT) and knowledge systems to overcome this. AI technologies with deep learning can help doctors analyze sleep data more accurately and build real-time information diagnostic systems. In this paper, we have looked at 36 studies done between 2013 and 2020 that used deep learning to study sleep data. We found that more than half of these studies look at brain activity recordings. We also found that using only EEG data is insufficient for perfect results. We should include other PSG recordings from sleep recordings to improve the diagnosis, like EEG and EMG signals. So, we need to use a mixture of different data sources to use deep learning for diagnosing sleep problems. This paper also discusses some methods from the last ten years that show how deep learning can work with these different data types to classify sleep stages.