Journal article

Generalizability of A Neural Network Model for Circadian Phase Prediction in Real-World Conditions

Julia E Stone, Andrew JK Phillips, Suzanne Ftouni, Michelle Magee, Mark Howard, Steven W Lockley, Tracey L Sletten, Clare Anderson, Shantha MW Rajaratnam, Svetlana Postnova

SCIENTIFIC REPORTS | NATURE PUBLISHING GROUP | Published : 2019

Abstract

A neural network model was previously developed to predict melatonin rhythms accurately from blue light and skin temperature recordings in individuals on a fixed sleep schedule. This study aimed to test the generalizability of the model to other sleep schedules, including rotating shift work. Ambulatory wrist blue light irradiance and skin temperature data were collected in 16 healthy individuals on fixed and habitual sleep schedules, and 28 rotating shift workers. Artificial neural network models were trained to predict the circadian rhythm of (i) salivary melatonin on a fixed sleep schedule; (ii) urinary aMT6s on both fixed and habitual sleep schedules, including shift workers on a diurnal..

View full abstract

Grants

Funding Acknowledgements

We thank the technicians, staff and students of the Monash University Sleep and Circadian Medicine Laboratory who aided data collection. For the shift work study we thank Saranea Ganesan, Megan Mulhall, Jessica Papaleo, Matthew McLaren, Aaron Johnson, Kaitlyn Crocker, Niamh McDonald, Emma Giliberto, and Trisha D'Lima for their data collection efforts, and Dr Glenn Eastwood, Helen Young, Dr Graeme Hart, and the staff of the Intensive Care Unit at Austin Health for their support. For the non-shift work study we thank Dr Leilah Grant, William McMahon, and Jinny Collet for research assistance; Caroline Beatty for recruitment efforts; study physicians Dr Simon Joosten and Dr Eric Kuo; and Sylvia Nguyen for psychiatric screening. We gratefully thank Mark Salkeld at the Adelaide Research Assay Facility for conducting the aMT6s and melatonin radioimmunoassays. This research was supported by the Cooperative Research Centre for Alertness, Safety and Productivity, Australia, and by an Australian Government Research Training Program (RTP) Scholarship.