The first step was to fix the proof of concept (POC) to create a new demo to get the necessary funding for the project. However, even after the initial fixes, BioLink realized they needed to start over. The POC had to be rebuilt including the software, like web pages and portals, but also the hardware and firmware. Our solution was to architect a full-blown IoT solution from scratch.
With the initial POC, all of the data was on-premise. But with the new architecture, we moved everything to the cloud. We were able to ensure each patient had a unique identifier and that all data was encrypted, safe, and HIPAA compliant. In addition to re-creating their initial adult brief wearable, we created a wearable device to monitor patient vitals, including body temperature, oxygen levels, and heart rate.
Accurate data in real-time
Once we had a solution for collecting patient data by wearable devices, we needed to ensure that it was transmitted to the right people at the right time.
To do so, we created smart hubs that were spread around the facility to upload data to the cloud in addition to web portals and a mobile application for caregiving staff. The web portals allow caregiving staff to check-in at nursing stations and in offices, and the mobile application is something they can carry around to check on a patient immediately and provide real-time feedback when they are alerted to a patient concern.
These executions allow data to be transferred from the cloud to the portals and mobile apps where caregiving staff are notified with an alert of any patient that requires follow-up care.
Improving patient care & outcomes
Each patient device is outfitted with an NFC chip so once the nursing staff is alerted, they cannot dismiss the alert until they scan the chip on the wearable device and complete the appropriate assessment. If the alert is not dismissed in a timely manner, as identified by the nursing home administration, the alert is escalated to a new caregiver. This ensures that no alert is missed because a caregiver ise attending to another patient or busy, and interventions are completed in a timely manner.
Personalized monitoring & care powered by machine learning
Now that we have this data from patients, we can use machine learning to gain insights about a patient’s care and needs. Initially, the patient care staff will set thresholds for appropriate vital levels, but as more data is collected, the system will learn and be able to figure out what the normal levels are for that patient specifically and alert accordingly.
The POC device was acting as a Bluetooth beacon, so as we evolved and proposed a new solution, security was at the center of the entire project.
We needed to ensure that the data transmission was completely secure and HIPAA compliant, but also that the wearable device had significant battery life so that data was accurate and consistent. To ensure a long battery life, we created a wearable and smart hub that doesn’t have any screens and only communicates via Bluetooth. You simply have to put the wearable device into pairing mode and it exchanges the encryption key through the cloud. Then, the encryption key is sent to the smart hub, and the data is all encrypted through the device and decrypted once it is in the cloud.
In addition, each wearable device has its own dedicated encryption key so if one wearable is compromised, BioLink can simply address the issue that individual device without other wearables and data being compromised.