Hospital Locator
Abstract
Patients within a hospital will often move around a lot. Between scans, room changes, and specialist consultations, finding out where a patient is located at any particular time can be quite difficult, as it relies on hospital staff accurately recording where the patient is meant to be at any time. Hospital Locator aims to cut out the human error in this system by automating the tracking of patients within the hospital. The key attributes for this system are availability, reliability, and scalability. Availability ensures that the user will be able to locate any patient at any time, while reliability will make sure that the reported location is as accurate as possible. Scalability is a necessity given the potentially thousands of patients per hospital that will need their locations being tracked, multiplied by however many hospitals are using the system.
Author
Name: Mitchell Hall
Student number: 46978581
Functionality
Hospital Locator will integrate with a piece of wearable technology such as a wristband that will report the wearer’s location. Hospital staff will be able to search a patient name and get their current location within the hospital, and also view their historical location data. Locations will be determined based on preset room information, where each room in the hospital will have its location stored and then compared with the location data of the patient. The location will also be displayed on a map of the hospital. The system will be able to account for altitude to determine the floor the patient is on and patient locations should be kept updated in every 10 minute interval, while also allowing staff to send a request to refresh a particular patient’s location. Hospital Locator will also integrate with existing hospital software so that when a patient is checked out, their data is cleared to allow for an incoming patient to use the tracker.
Scope
The MVP (minumum viable product) of Hospital Locator will include the following essential features to keep track of patient locations in a hospital:
- Patient Creation: When a patient is checked in, their name is associated with the tracker they must wear, and their location can start being recorded
- Hospital Registration: The locations and names of each area within a hospital including the floor and ward can be registered
- Tracking: The system should integrate with the tracker and periodically retrieve location information from it. This data should then be be interpreted using the hospital data to give a general location for the patient (e.g. what ward they are currently in)
- Patient Listing: Displays a list of all patients for the hospital staff to view. This should have capability to search for a individual patient and view their data in more depth
- Patient Deletion: Once a patient is checked out within the existing hospital software, the patient data within Hospital Locator (especially the association with the tracker) should be deleted
Quality Attributes
- Availability: The ability to retrieve patient locations should be accessible at all times in case of emergency. If a patient cannot be located and the system goes down, that patient’s life may be at risk. For this reason, it is crucial that the system be available as much as possible.
- Reliability: Reliability is the accuracy of the data returned by Hospital Locator. As there can be slight inaccuracies in the precise location returned by trackers, it may be unfeasible to always get the exact room the patient is in, but the general zone and floor should always be correct.
- Scalability: The system should be able to function with a large amount of data inbound for every hospital using the system. As this system may need to be used in emergency cases, load times should be kept as minimal as possible when in high stress periods, and be able to handle multiple patients for multiple hospitals. For this reason, it would be considered acceptable for the latest location data to be at most 10 minutes old, with functionality in place for immediate retrieval if necessary
Evaluation
- Availability: Location should be accurate to the last 10 minutes. The list of all patients can be queried with no more than a 5 second delay. The system must also be online at all times, which will be tested by using ping and stress testing.
- Reliability: The system should in 99% of cases correctly identify floor and ward. Errors in room number should only occur between two adjacent rooms, and only happen in less than 5% of reports. These tests will be conducted in both low-stress and high-stress scenarios to ensure the system stays reliable at all times.
- Scalability: The system can handle at least 900 incoming location data packets across 3 endpoints, and be able to register new patients and hospitals without interfering with existing functionality. This will be tested using the K6 stress-testing framework.