The role of healthcare providers and caregivers in monitoring critically ill children: a qualitative study in a tertiary hospital, southern Malawi.

BACKGROUND: Critically ill children require close monitoring to facilitate timely interventions throughout their hospitalisation. In low- and middle-income countries with a high disease burden, scarce paediatric critical care resources complicates effective monitoring. This study describes the monitoring practices for critically ill children in a paediatric high-dependency unit (HDU) in Malawi and examines factors affecting this vital process. METHODS: A formative qualitative study based on 21 in-depth interviews of healthcare providers (n = 12) and caregivers of critically ill children (n = 9) in the HDU along with structured observations of the monitoring process. Interviews were transcribed and translated for thematic content analysis. RESULTS: The monitoring of critically ill children admitted to the HDU was intermittent, using devices and through clinical observations. Healthcare providers prioritised the most critically ill children for more frequent monitoring. The ward layout, power outages, lack of human resources and limited familiarity with available monitoring devices, affected monitoring. Caregivers, who were present throughout admission, were involved informally in monitoring and flagging possible deterioration of their child to the healthcare staff. CONCLUSION: Barriers to the monitoring of critically ill children in the HDU were related to ward layout and infrastructure, availability of accurate monitoring devices and limited human resources. Potential interventions include training healthcare providers to prioritise the most critically ill children, allocate and effectively employ available devices, and supporting caregivers to play a more formal role in escalation.

A U - Net Deep Learning Model for Infant Heart Rate Estimation from Ballistography.

Ballistography(BSG) is a non-intrusive and low- cost alternative to electrocardiography (ECG) for heart rate (HR) monitoring in infants. Due to the inter-patient variance and susceptibility to noise, heartbeat detection in the BSG waveform remains a challenge. The aim of this study was to estimate HR from a bed-based pressure mat BSG signal using a deep learning approach. We trained a U-Net deep neural network through supervised learning by deriving ground truth as the location of the heartbeats from simultaneously recorded ECG signals after peak matching. For improved generalization, we modified an existing U - Net to include an IC-layer. A predictive performance of 80% was achieved using the U-Net without the IC-layer. The inclusion of the IC-layer, while improving the generalization ability of the model to detect heartbeats, did not improve the HR estimation performance.