Machine learning for emerging infectious disease field responses.

Emerging infectious diseases (EIDs), including the latest COVID-19 pandemic, have emerged and raised global public health crises in recent decades. Without existing protective immunity, an EID may spread rapidly and cause mass casualties in a very short time. Therefore, it is imperative to identify cases with risk of disease progression for the optimized allocation of medical resources in case medical facilities are overwhelmed with a flood of patients. This study has aimed to cope with this challenge from the aspect of preventive medicine by exploiting machine learning technologies. The study has been based on 83,227 hospital admissions with influenza-like illness and we analysed the risk effects of 19 comorbidities along with age and gender for severe illness or mortality risk. The experimental results revealed that the decision rules derived from the machine learning based prediction models can provide valuable guidelines for the healthcare policy makers to develop an effective vaccination strategy. Furthermore, in case the healthcare facilities are overwhelmed by patients with EID, which frequently occurred in the recent COVID-19 pandemic, the frontline physicians can incorporate the proposed prediction models to triage patients suffering minor symptoms without laboratory tests, which may become scarce during an EID disaster. In conclusion, our study has demonstrated an effective approach to exploit machine learning technologies to cope with the challenges faced during the outbreak of an EID.

National retrospective cohort study to identify age-specific fatality risks of comorbidities among hospitalised patients with influenza-like illness in Taiwan.

OBJECTIVES: This study aimed to examine comprehensively the prognostic impact of underlying comorbidities among hospitalised patients with influenza-like illness (ILI) in different age groups and provide recommendations targeting the vulnerable patients. SETTING AND PARTICIPANTS: A retrospective cohort of 83 227 hospitalised cases with ILI were identified from Taiwan's National Health Insurance Research Database from January 2005 to December 2010. Cases were stratified into three different age groups: paediatric (0-17 years), adult (18-64 years) and elderly (≧65 years), and their age, sex, comorbidity and past healthcare utilisation were analysed for ILI-associated fatality. MAIN OUTCOME MEASURES: ORs for ILI-related fatality in different age groups were performed using multivariable analyses with generalised estimating equation models and adjusted by age, sex and underlying comorbidities. RESULTS: Hospitalised ILI-related fatality significantly increased with comorbidities of cancer with metastasis (adjusted OR (aOR)=3.49, 95% CI: 3.16 to 3.86), haematological malignancy (aOR=3.02, 95% CI: 2.43 to 3.74), cancer without metastasis (aOR=1.72, 95% CI: 1.54 to 1.91), cerebrovascular (aOR=1.24, 95% CI: 1.15 to 1.33) and heart diseases (aOR=1.19, 95% CI: 1.11 to 1.27) for all age groups. Adult patients with AIDS; adult and elderly patients with chronic kidney disease, tuberculosis and diabetes were significantly associated with elevated risk of death. Severe liver diseases and hypothyroidism among elderly, and dementia/epilepsy among elderly and paediatrics were distinctively associated with likelihood of ILI-related fatality. CONCLUSIONS: Different age-specific comorbidities were associated with increasing risk of death among hospitalised ILI patients. These findings may help update guidelines for influenza vaccination and other prevention strategies in high-risk groups for minimising worldwide ILI-related deaths.

Symmetric rearrangement of groundwater-fed streams.

Streams shape landscapes through headward growth and lateral migration. When these streams are primarily fed by groundwater, recent work suggests that their tips advance to maximize the symmetry of the local Laplacian field associated with groundwater flow. We explore the extent to which such forcing is responsible for the lateral migration of streams by studying two features of groundwater-fed streams in Bristol, Florida: their confluence angle near junctions and their curvature. First, we find that, while streams asymptotically form a 72° angle near their tips, they simultaneously exhibit a wide 120° confluence angle within approximately 10 m of their junctions. We show that this wide angle maximizes the symmetry of the groundwater field near the junction. Second, we argue that streams migrate laterally within valleys and present a new spectral analysis method to relate planform curvature to the surrounding groundwater field. Our results suggest that streams migrate laterally in response to fluxes from the surrounding groundwater table, providing evidence of a new mechanism that complements Laplacian growth at their tips.

A free-boundary model of diffusive valley growth: theory and observation.

Valleys that form around a stream head often develop characteristic finger-like elevation contours. We study the processes involved in the formation of these valleys and introduce a theoretical model that indicates how shape may inform the underlying processes. We consider valley growth as the advance of a moving boundary travelling forward purely through linearly diffusive erosion, and we obtain a solution for the valley shape in three dimensions. Our solution compares well to the shape of slowly growing groundwater-fed valleys found in Bristol, Florida. Our results identify a new feature in the formation of groundwater-fed valleys: a spatially variable diffusivity that can be modelled by a fixed-height moving boundary.

Path selection in the growth of rivers.

River networks exhibit a complex ramified structure that has inspired decades of studies. However, an understanding of the propagation of a single stream remains elusive. Here we invoke a criterion for path selection from fracture mechanics and apply it to the growth of streams in a diffusion field. We show that, as it cuts through the landscape, a stream maintains a symmetric groundwater flow around its tip. The local flow conditions therefore determine the growth of the drainage network. We use this principle to reconstruct the history of a network and to find a growth law associated with it. Our results show that the deterministic growth of a single channel based on its local environment can be used to characterize the structure of river networks.