EUVIMEDObjevte EUVIMED
Open access

Enhancing COVID-19 Screening Models With Epidemiological and Mobility Features: Machine-Learning Model Study

Verbesserung von COVID-19-Screeningmodellen mit epidemiologischen und Mobilitätsmerkmalen: Machine-Learning-Modellstudie

Hyunwoo Choo, Dohyung Lee, Soo-Yong Shin, Jiwoo Lee, Duhun Lee, Eonji Kim, Namsoo Oh, Christina Kim, Myeongchan Kim, Hyo Jung Kim

JMIR AI · 2026-03-05 · 5 · e54956-e54956

DOI: 10.2196/54956 ↗

Auf einen Blick

Bei 48.798 Personen verbesserte die Einbeziehung von Mobilitäts- und Epidemiedaten die Leistung von 5 ML-Screeningmodellen für COVID-19. Der höchste AUC-Wert stieg von 0,8712 auf 0,9104.

Fragestellung

Verbessern Mobilitäts- und Epidemiedaten zusätzlich zu Symptomen die Leistung von ML-Modellen zum COVID-19-Screening?

Methode

48.798 Personen meldeten per Smartphone-App täglich Symptome, Standortdaten und Testergebnisse; ergänzt um Daten von Our World in Data und nationale Regierungsdaten. Trainiert wurden 5 ML-Modelle.

Ergebnisse

Die Ergänzung von Mobilitäts- und Epidemiedaten verbesserte alle 5 Modelle. Der höchste AUC-Wert stieg von 0,8712 ohne diese Daten auf 0,9104 mit ihnen.

Schlussfolgerung der Autor:innen

Die Studie zeigte das Potenzial von Mobilitäts- und Epidemiedaten zusammen mit Symptomdaten, um die Genauigkeit von ML-Modellen zur COVID-19-Diagnose zu verbessern.

Kernaussagen

  • 48.798 Personen mit App-Daten zu Symptomen, Standort und Testresultaten
  • Mobilitäts- und Epidemiedaten verbesserten alle 5 ML-Modelle
  • Höchster AUC-Wert: 0,8712 ohne vs. 0,9104 mit Zusatzdaten

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

At a glance

Journal article in JMIR AI, published online on 2026-03-05. Written by Hyunwoo Choo and 9 co-authors. Institution: Department of Digital Health, Samsung Advanced Institute for Health Sciences and Technology (SAIHST), Sungkyunkwan University, Seoul, Republic of Kor…. Cited 1 times so far (Crossref); 45 references. The full text is freely available in PubMed Central.

Auf einen BlickAbstraktSubjectsBibliografické údajeCiteRelated studies

Abstract (original language)

Background

Despite the significant post–COVID-19 pandemic surge in research using symptom data and machine learning (ML) for patient screening, data on patient trajectories and epidemiological conditions, although crucial, have remained underused.

Objective

This study aimed to enhance the performance of ML models for COVID-19 screening by incorporating mobility and epidemic information in addition to patient symptom data.

Methods

Data, including daily self-reported symptoms, location information, and test results, were collected from 48,798 individuals using a smartphone app. These data were then combined with Our World in Data and national government epidemic information to train 5 ML-based screening models to classify patient infection status. The models were logistic regression, extreme gradient boosting, light gradient boosting machine, tabular data network, and Google AutoML.

Results

The addition of mobility and epidemic data significantly improved the performance of all 5 models. The highest area under the receiver operating characteristic curve score increased from 0.8712 without mobility and epidemic data to 0.9104 with mobility and epidemic data. This highlights the considerable impact of external information on enhancing the performance of ML models.

Conclusions

This study demonstrated the potential of using mobility and epidemic data, such as location information and epidemic data, in combination with patient symptom data to improve the accuracy of ML models for diagnosing COVID-19. Considering additional contextual information can enhance the ability to screen for COVID-19.

Subjects

Author keywords

COVID-19deep learningepidemiologymachine learningmass screeningmobility data

Authors and institutions

  • Department of Digital Health, Samsung Advanced Institute for Health Sciences and Technology (SAIHST), Sungkyunkwan University, Seoul, Republic of Korea.
  • Mobile Doctor Co., Ltd, Seoul, Republic of Korea.
  • Department of Biomedical Software Engineering, The Catholic University of Korea, 43 Jibong-ro, Wonmi-gu, Bucheon-si, Gyeonggi-do, 14662, Republic of Korea, +82-10-3093-1790.

Cite

APA

Choo, H., Lee, D., Shin, S. Y., Lee, J., Lee, D., Kim, E., Oh, N., Kim, C., Kim, M., & Kim, H. J. (2026). Enhancing COVID-19 Screening Models With Epidemiological and Mobility Features: Machine-Learning Model Study. JMIR AI, 5, e54956-e54956. https://doi.org/10.2196/54956

Vancouver

Choo H, Lee D, Shin SY, Lee J, Lee D, Kim E, et al. Enhancing COVID-19 Screening Models With Epidemiological and Mobility Features: Machine-Learning Model Study. JMIR AI. 2026;5:e54956-e54956. doi:10.2196/54956

BibTeX
@article{Choo2026,
  author = {Choo, Hyunwoo and Lee, Dohyung and Shin, Soo-Yong and Lee, Jiwoo and Lee, Duhun and Kim, Eonji and Oh, Namsoo and Kim, Christina and Kim, Myeongchan and Kim, Hyo Jung},
  title = {Enhancing COVID-19 Screening Models With Epidemiological and Mobility Features: Machine-Learning Model Study},
  journal = {JMIR AI},
  year = {2026},
  volume = {5},
  pages = {e54956--e54956},
  doi = {10.2196/54956}
}
Start for free →