Skip to main content
Home
  • Main Menu
  • Topics
    • Video
    • Provider
    • Payer
    • Pharma
    • Consumer
    • Investor

Regions

  • ANZ
  • ASIA
  • EMEA
  • Global Edition
ASIA
Predictive analytics

Korean AI predicts mood episodes using sleep-wake data

The AI models only utilise wearable sleep and circadian rhythm data to analyse patterns driving mood disorders.
By Adam Ang
November 26, 2024
08:20 pm
Share

Photo: Ketut Subiyanto/Pexels

Researchers from South Korea have built machine learning-based models that can predict mood episodes using only sleep and circadian rhythm data from wearable devices. 

The team comprised researchers from the Institute for Basic Sciences (IBS), Korea Advanced Institute of Science and Technology (KAIST), and Korea University College of Medicine.

FINDINGS

In a study, which findings were published in Nature's npj Digital Medicine journal, the research team first collected and analysed 429 days' worth of sleep-wake data generated from Fitbit of 168 Korean patients with mood disorders, including major depression and bipolar disorders. It is said that mood disorders are closely associated with irregular sleep and circadian rhythms.

From this dataset, they extracted 36 sleep and circadian rhythm features, which were then applied to train models based on the machine learning library XGBoost to predict mood episodes.

Based on findings, the predictive models achieved 80%, 98%, and 95% accuracy in predicting depressive, manic, and hypomanic episodes, respectively. 

Findings also suggested that daily changes in circadian rhythm may be a key predictor of mood episodes; delayed circadian rhythms can potentially lead to depressive episodes while advanced circadian rhythms can raise the possibility of manic episodes. 

WHY IT MATTERS

While there already exist AI models for predicting the mood states of people with mood disorders, these typically require various data including sleep, heart rate, light exposure, phone usage, and GPS data, which may be costly to collect. It may also pose security and privacy risks and storage challenges, the research team emphasised. 

"By developing a model that predicts mood episodes based solely on sleep-wake pattern data, we have reduced the cost of data collection and significantly improved clinical applicability," claimed Kim Jae Kyoung, KAIST associate professor and IBS chief investigator. 

The research team also point to the potential of complementing the prediction of mood states with the use of digital therapeutics to evaluate the daily risk of mood episode relapse and promote healthy sleep-wake cycles and circadian rhythms by providing reminders. 

MARKET SNAPSHOT

A similar research in Singapore also utilised sleep patterns and circadian rhythms to predict depression risk. The AI-based Ycogni model by a research team from Nanyang Technology University demonstrated 80% accuracy in identifying those at high risk of depression using also sleep-wake data collected from Fitbit. 

Early this year, research from Japan developed what could be the world's first AI model that can predict Alzheimer's disease using wearables data. Eisai and Oita University collected and analysed a range of biological and lifestyle data to develop a predictive model that can be used as a pre-screening tool for people suspected of having the disease.

Tags: 
KAIST, South Korea, wearables, mood disorders, depression, bipolar disorder, predictive AI, Fitbit

More regional news

Three surgeons operating on a patient

Thyroid cancer AI assistant developed in Hong Kong

By
Adam Ang
April 29, 2025
A cancer patient consulting with a doctor

AI-driven cervical cancer screening comes to Ho Chi Minh City

By
Adam Ang
April 23, 2025
Photo of an eye

AI-driven ADHD screening using eye images developed in Korea

By
Adam Ang
April 22, 2025
Share

Top Story

Three surgeons operating on a patient
Thyroid cancer AI assistant developed in Hong Kong

Editor's Pick

23andMe files for bankruptcy, CEO Wojcicki resigns
Q&A: Oura appoints its first chief medical officer
First blood pressure management system integrated with DeepSeek in Hong Kong
Emirates will offer inflight telemedicine aboard 300 aircraft
Nanowear announces partnership with Dexcom
Navina raises $55M to expand AI for value-based care

More Stories

Pharmacist interacting online
Serve You Rx, Amazon Pharmacy partner for home delivery
Two people shaking hands with six others watch
DarioHealth closes up to $50M debt refinancing and more funding news
Nurse consulting with patient
Teton unveils 3D hospital digital twin using Denmark's supercomputer
Young person communicating with a cell phone
Talkspace, Bark Technologies partner to monitor children's online activities
Pharmacist giving a person an injection
Plenful scores $50M and more funding news
Person interacting with a healthcare professional via tablet
Teladoc acquires UpLift for $30M, releases first quarter results
Person in a pharmacy setting
Hims & Hers, Novo Nordisk partner to offer weight-loss drug Wegovy
Healthcare professional looking at a PC screen
Commure, HealthTap partner for expanded after-hours care
Home

More News

  • Healthcare IT News
  • Healthcare IT News Australia
  • Healthcare Finance News
  • HIMSS TV

MobiHealthNews

 

Connect

Add MobiHealthNews to your network.
HIMSS25 European Health Conference & Exhibition
HIMSS25 European Health Conference & Exhibition
Get ready for knowledge-sharing, all the latest innovations, and in-depth demos with Europe's most influential healthcare community.
10 - 12 June, 2025 | Paris
Learn More
AI in Healthcare Forum
AI in Healthcare Forum
The HIMSS AI in Healthcare Forum cuts through the hype to showcase real-world examples illustrating the transformative potential, and realistic challenges of AI application across the care continuum.
10 - 11 July 2025 | New York
Learn More

Footer Menu

  • About
  • Advertise
  • Reprints
  • Contact
  • Privacy Policy

© 2025 MobiHealthNews is a publication of HIMSS Media

X

Topics

  • Video
  • Provider
  • Payer
  • Pharma
  • Consumer
  • Investor

Career

  • Events
  • Jobs
  • Learning Center

More

  • About
  • Advertise
  • Contact
  • Privacy Policy

Regions

  • ANZ
  • ASIA
  • EMEA
  • Global Edition

MobiHealthNews:

The latest news in digital health delivered daily to your inbox.

Search form

Top Stories

Smart phone and smart watch
Oura and Google Fitbit partner with academia and others to battle opioid disorder
Three surgeons operating on a patient
Thyroid cancer AI assistant developed in Hong Kong
A doctor discussing findings with an older patient
Digital dementia support project gets $2M federal funding and more briefs