Continuous diagnosis and prognosis by controlling the update process of deep neural networks

Chenxi Sun; Hongyan Li; Moxian Song; Derun Cai; Baofeng Zhang; Shenda Hong

doi:10.1016/j.patter.2023.100687

Continuous diagnosis and prognosis by controlling the update process of deep neural networks

Patterns (N Y). 2023 Feb 3;4(2):100687. doi: 10.1016/j.patter.2023.100687. eCollection 2023 Feb 10.

Authors

Chenxi Sun^{1

2}, Hongyan Li^{1

2}, Moxian Song^{1

2}, Derun Cai^{1

2}, Baofeng Zhang^{1

2}, Shenda Hong^{3

4}

Affiliations

¹ Key Laboratory of Machine Perception (Ministry of Education), Peking University, Beijing 100871, China.
² School of Intelligence Science and Technology, Peking University, Beijing 100871, China.
³ National Institute of Health Data Science, Peking University, Beijing 100191, China.
⁴ Institute of Medical Technology, Health Science Center of Peking University, Beijing 100191, China.

Abstract

Continuous diagnosis and prognosis are essential for critical patients. They can provide more opportunities for timely treatment and rational allocation. Although deep-learning techniques have demonstrated superiority in many medical tasks, they frequently forget, overfit, and produce results too late when performing continuous diagnosis and prognosis. In this work, we summarize the four requirements; propose a concept, continuous classification of time series (CCTS); and design a training method for deep learning, restricted update strategy (RU). The RU outperforms all baselines and achieves average accuracies of 90%, 97%, and 85% on continuous sepsis prognosis, COVID-19 mortality prediction, and eight disease classifications, respectively. The RU can also endow deep learning with interpretability, exploring disease mechanisms through staging and biomarker discovery. We find four sepsis stages, three COVID-19 stages, and their respective biomarkers. Further, our approach is data and model agnostic. It can be applied to other diseases and even in other fields.

Keywords: COVID-19; biomarker; continuous classification; deep learning; diagnosis; disease staging; prognosis; sepsis; time series.