Artificial Intelligence in healthcare has the potential to grow to $6.16 billion at an annual rate of 68.55% between 2018 and 2022. However, the Indian healthcare industry is very reactive in the present scenario. This means if a person develops some disorder, starts to get certain symptoms then gets the tests done, intervention starts only after that.
Now, Bengaluru-based Dozee is one such startup that has taken initiatives to make healthcare proactive instead of reactive in order to prevent and manage the disease better. Founded by Mudit Dandwate and Gaurav Parchani, this young startup is using artificial intelligence to change this sector.
Analytics India Magazine caught up with Parchani to gain insights on how the company is leveraging AI in the health sector. Parchani stated that Dozee’s advanced health intelligence technology identifies the early signs of health deterioration and early intervention saves a great deal of pain as well as the cost to the user.
What They Do
Dozee is a contact-free health monitor which works best when it is placed under the mattress. It consists of a thin sheet of MEMS-based vibro-acoustic sensors, data accumulation, and a communication unit. The sensor sheet when placed under the mattress under the chest of the user, capturing micro-vibrations produced by the body every time the heart pumps blood, inhalation, exhalation, muscle twitches, tremors or body movements. These signals are then precisely converted into bio-markers by AI-based analytics engine.
According to Parchani, Dozee is very simple to use as well as provides high fidelity data at the same time. Almost 15% of the Dozee user base is above 80 years of age and is built specifically for the low resource (hence the price tag of ₹7,200) and home settings which requires no technical expertise to set-up. The product provides effective remote monitoring service that enables high-skill medical staff to take decisions remotely.
Tech Stack Behind
Dozee works on a technology called ballistocardiography (BCG). Dozee's advanced algorithms have been trained over billions of data points to achieve 98.4% accuracy in measuring heartbeat and respiration. In fact, for many cases, the AI model has successfully predicted early health deterioration in conditions ranging from fever to heart failure, even before they could see the symptoms.
From calculating movement instances to detecting every heartbeat to finding risk score for every individual, the startup uses AI/ML algorithms. Their engineering team of 15 also uses traditional ML algorithms like clustering, SVMs to algorithms such as CNNs and LSTMs to develop the model.
According to Parchani, the company faces a lot of problems in getting the right talent. There is a skill gap which makes candidates unsuitable even when though they have a lot of enthusiasm and exuberance to do the work.
Talking about its competitors, Parchani says, “Our closest competitor is an Israeli company called Early Sense. Secondly, wearables can also be considered as our competitors. The edge we have over them is the accuracy with the medical-grade devices and ease of use.”
Now, going forward, they definitely want to create better health deterioration detection algorithms, which he thinks will happen as they get more and more data that will further sharpen and fine-tune their predictive analysis capability. Currently, there are more than 1,000 users of Dozee and the company plans to reach 2 lakh users in the next two years globally. Future projects involve nocturnal seizure detection and Parkinson’s management.
Register for our upcoming events:
- Meetup: NVIDIA RAPIDS GPU-Accelerated Data Analytics & Machine Learning Workshop, 18th Oct, Bangalore
- Join the Grand Finale of Intel Python HackFury2: 21st Oct, Bangalore
- Machine Learning Developers Summit 2020: 22-23rd Jan, Bangalore | 30-31st Jan, Hyderabad
Enjoyed this story? Join our Telegram group. And be part of an engaging community.
What's Your Reaction?
A Technical Journalist who loves writing about Machine Learning and Artificial Intelligence. A lover of music, writing and learning something out of the box. Contact: firstname.lastname@example.org