Surgical care in the developing world is poor. Patients in the developing world account for 80% of deaths from surgically treated conditions, but only 26% of surgical procedures carried out worldwide. Health facilities in rural areas have poor infrastructure and lack essential equipment needed for surgical care. Developing countries face a chronic health worker shortage and less-trained health workers, under sub-standard conditions, are less likely to be successful and further contribute to adverse surgical outcomes.
Limited human resources is associated with increased in-patient mortality. In a study conducted on nine European countries, an increase in a nurses' workload by one patient increased the likelihood of an inpatient dying within 30 days of admission by 7%.
Initiatives to increase staffing and resources have shown limited efficacy. Health surveillance assistants are demotivated due to lack of access to continuing education, lack of professional development and financial incentive, stress, workplace violence, bullying, harassment, and lack of feeling valued. More clinical resources (tools) may not immediately better care because of limited training for (unfamiliarity with) those resources and issues with adoption and compliance.
There is a need to maximize the utility of existing human resources.
What it does
Our device is an inpatient bedside monitor for low-resource settings to allow caretakers better allocate their care across all their patients. The monitor continuously measures the metrics most associated with post-operational convalescence: blood oxygen saturation, temperature, heart rate, and respiratory rate. Each of these metrics are compared to a threshold, which when crossed, sends an alert to the caregiver's device notifying them of a potential issue in real time. These thresholds can to stratified to signal different types of alerts depending on the severity, thereby allowing the caretaker to better manage their distribution of care.
How we built it
We built a hardware simulation of our conceived idea using an arduino uno and associated sensor components such as a temperature sensor. We also used different LEDs to simulate the different alerts associated with the metrics that we plan to monitor, such as oxygen saturation, temperature, heart rate, and respiratory rate. For the hackathon, we hard-coded the different thresholds for each metric, but eventually we hope to integrate an adaptive system that accounts for individual patient characteristics and use machine learning to determine optimal thresholds based on those characteristics.
Challenges we ran into
Limited access to hardware components.
Accomplishments that we're proud of
We built a simple device with an arduino uno to demonstrate our proof-of-concept.
What we learned
We learned that there are a lot of barriers to improving post-operational care in rural settings and third world countries. These issues cannot be solved as easily introducing more resources to these regions (which is already difficult by itself), but must start with fundamental change in how patient care is practiced in those settings.
What's next for Continuous Monitoring Code Calling Center
We hope to continue developing our hardware prototype and eventually integrate our bed-side device to an iOS device via blue-tooth. We also hope to develop backend algorithms for better classification of patient characteristics to determine better thresholds for each individual.