While the COVID pandemic was at its peak, Nagpur was one city that was experiencing a high number of COVID-19 cases. This led to hospitals struggling to manage COVID-infected patients due to the lack of healthcare staff and the risk of infection associated with it. Indira Gandhi Government Medical College & Hospital (IGGMC ), Nagpur was one such hospital that was overwhelmed with COVID patients. IGMC relied on artificial intelligence and remote monitoring to reduce the risk of infection amongst their staff and adhere to the highest patient safety standards.
One of the biggest challenges faced by the hospital was time optimisation of the healthcare staff. According to the hospital, each nurse had to go to the wards to manually record the patients’ vitals and log it into their system, multiple times a day. Eventually, this became an arduous process for nurses, especially during the COVID pandemic, when the hospital has seen 100% admittance.
Sign up for your weekly dose of what's up in emerging technology.
With such a high number of COVID patients being admitted to the hospital, it became imperative for the hospital to safeguard the nurses and doctors by reducing the time they spent in the wards physically interacting with COVID-infected patients. And that is why, IGGMC decided to partner with Dozee, an advanced health intelligence tracker that acts as a contactless monitoring device for the hospital.
Using Dozee, IGGMC automated the entire vitals monitoring process using the AI-based contactless and remote model. Thus, nurses were no longer needed to spend hours taking patient’s readings, and doctors could remotely monitor the data through the dashboard. To understand how Dozee works and how the system has been developed, we got in touch with Mudit Dandwate, CEO and Co-Founder of Dozee.
How Does It Work?
With four years of rigorous research by collaborating with the top doctors and hospitals in India, the company created this AI-powered device. “Indira Gandhi Government Medical College & Hospital, Nagpur was introduced to Dozee, when the hospital was already dealing with many of Nagpur’s COVID patients,” said Mudit. “The company offered a hospital-specific solution which comes with the Dozee device and a dashboard where all the patient data is available for the doctors.”
The solution tracks key vitals of the patient’s body such as heart and respiration rate, oxygen saturation, sleep stages, stress-recovery with a medical-grade 98.4% accuracy. Alongside, the contactless sensors placed under the mattress, capture real-time body vitals without using any external wires touching the patient’s body. The sensor captures micro-vibrations produced by the body every time the heart pumps blood, during inhalation, exhalation, muscle twitches, tremors and body movements.
However, these vibrations are too noisy to make any sense out of them. And that’s where Dozee’s proprietary advanced AI algorithms step in to make it one of the most precise tools.
“We use artificial intelligence to remove any noise in the vibrations data so that the system can only collect the actual vitals that are being monitored. This is then converted into health parameters such as heart rate, respiration rate, oxygen saturation levels etc. which is what the doctors view on the dashboard,” explained Mudit.
Additionally, the solution comes with an AI-powered, Early Warning System then converts these signals into biomarkers and transmits the data to the cloud where it is processed through the patient monitoring dashboard — Sens Health. This helps doctors to take immediate actions for patients who are at a critical stage. “According to the data, this early AI detection system has saved more than 70 lives till date,” added Mudit.
While Dozee’s frontend has been built using Angular, React, Android and iOS, its backend was built using Python and Go — Python was used for machine learning modules, and the Go language has been used for the infrastructure. Adding to this, Mudit said that Dozee’s infrastructure was built primarily on Amazon Web Services. The company leveraged decision trees, classifications, and deep learning models such as convolutional neural network, recurrent neural networks for machine learning.
As of now, Dozee has managed to install 150 devices at IGGMC and a total of 250 devices in Nagpur city. Using Dozee made it easier for IGGMC healthcare staff to maintain physical distance from the patients and avoid the risk of getting infected and optimised their time with no manual processes for checking vitals in every ward.
Another huge advantage is that Dozee alerts the doctors about anomalies in the readings, for critical patients to be transferred to the ICU on time. “Our device has also replaced most of the vitals collection devices that need to be carried around the wards,” added Mudit.
“We are planning to expand to other hospitals in Nagpur soon and also working on enhancing the capabilities of the current device for all our partnering hospitals and future partners,” concluded Mudit.