Well Being Digital Limited (WBD) develops highly accurate physiological sensing technology including ActivHeartsTM (dynamic heart rate sensing technology) and MogoTM (posture/balance/gait sensing technology). The devices provide consumers with awareness of their pre-medical conditions.
Digital Insurer’s Asian Startup Insurtech Award
Well Being Digital Limited
1. Introduction
Well Being Digital Limited (WBD) develops highly accurate physiological sensing technology including ActivHearts™ (dynamic heart rate sensing technology) and Mogo™ (posture/balance/gait sensing technology).
Backed by worldwide patents, we have won multiple international and local awards including Mobile World Congress, 43rd Geneva Inventions Convention and the Grand Prize of the Hong Kong ICT 2016 (appendix)
The technologies’ accuracies have been tested by 3rd party customers (Appendix) and has reached a stage of accuracy and maturity that is applicable for Pre-Medical purposes, these refers to highly accurate devices that when compared to medical devices, have up to 0.99 correlation. They provide consumers with awareness of their pre-medical conditions.
Industry acclaimed technology building-blocks to develop unique wearables;
Has more than 35+ patents to protect partners;
Besides licensing to world-wide brands like Parrot and Muzik LLC (http://www.cnbc.com/2016/01/04/twitter-invests-in-muzik-a-high-end-headphone- startup.html), WBD supports startups like Actywell and Grit to build stylish Pre-Medical grade wearables;
Work with established industrial experts to implement their scientific research, such as the 3 Minutes VO2max Test, that is self-administrable;
Collaborating with a Major Insurer to help build their Healthcare 2.0 plan;
Collaborating with a Top Hong Kong uni to develop preventive healthcare devices for their upcoming Medical Center.
2. Abstract
With increasing medical costs and ageing population, current stakeholders turn to technology to reduce high medical cost that provides insights of people’s physiological conditions to reduce deteriorations.
With building-block technology, WBD enabled multiple Joint Venture companies (JVs) to create it ensures:
Devices are stylish enough to be worn continuously;
These bio-parameters are properly authenticated and highly accurate;
Insurers can get information towards de-risking their underwriting;
A multi-tiered “walled garden” of devices that collect and attribute importance to different levels of data integrities.
3. WBD’s Vision of Healthcare 2.0
3.1. Weakness of current hardware-agnostic approach
Currently wearables developers and brands range from high quality ones (Apple Watch, Fitbit) to lower qualities (e.g. Taobao) which provide limited value towards customers behavior understanding.
While the appeal of being hardware-agnostic sounds sexy to those Insurers who want to attract the biggest possible pool of wearables users or who do not want to manage inventory, data from questionable accuracies pollutes the database resulting in garbage- in-garbage-out.
Hardware-agnostic implies no common standard of data accuracy and integrity;
Hardware-agnostic provides no value-add for the Insurers building Healthcare 2.0. Devices like Apple Watch neither customize its features/functions for Insurers nor
provide additional granularity, e.g. while it reports the steps per day, it does not provide the time stamp association, leading to different recommendations and insights.
3.2. Strength of WBD’s Multi-Tiered Hardware Approach
WBD’s approach is a semi-walled garden where reputable data are placed at higher levels, while data from questionable hardware are for reference only to maintain consumer touch points.
WBD works with sports science laboratory like with the Sports Science Lab of CUHK and implement their scientific 3 minute VO2max test that is a good indicator of fitness and longevity.
Raw data is useful as it includes time-stamp, and granularity such as distinguishing between a walk and a run step; when correlating with that moment’s heart rate helps Insurers understand if the customer is over-exerting already and needs to slow down
(e.g. the customer is known to be fit, but exceeding normal running heart rate, he could be ill) or that the customer should be encouraged to take things slowly (e.g. an over- weight customer that doesn’t exercise frequently exceeded heart rate norm even with a little walk).
4. Our Building-Blocks
4.1 Pre-Medical Grade Devices
4.2 Measurable Physiological Data
5. Our Big Data Base
Health Data & Device Ecosystem (HDDE) is our Architecture to data generation, processing and analysis. Any device connected to the HDDE is expected to perform a certain health related function, e.g.
measuring some status of the consumers. These, either in raw form or in algorithmic filtered and extracted form, will be communicated via a secure channel to an intermediary data store, usually a mobile App and can be secured wirelessly (like Bluetooth) or wired (e.g. Lightning cable).
The intermediary data store enhance the data with additional information such as date and time, GPS location etc. Furthermore, it is also used to analyze long term observations by filtering for outliers or recognizing trends.
From this intermediary, there is another secured channel to transport the enriched & extracted data into a cloud-based Backend system that acts as the final integrator, across all users. This Backend is responsible to allow user specific retrieval of data, for personal use and analysis. But it also enables aggregated trend analysis and comparison across various (anonymized) users and user segments within the HDDE.
Key feature of our HDDE is that it supports the tiered approach described above. In the WBD-IoT devices we want to support user identification and authentication as much as possible, to detect when the device is worn by another person than the owner itself. This is convenient for the user if a device is e.g. borrowed to a friend for demonstration or testing, as the gained data for the tester is not mingled with the actual measurements of the user. This also drastically enhances the level of credibility if user data is being used in an insurance context, where certain health attributes will influence commercial decisions (premium discount, benefits).
Per the diagram, HDDE will be able to be ‘downward-compatible’ and able to take in data from highly accurate devices that don’t support the ID/Authentication features (Tier 2) and data from 3rd party devices without requiring any attestation of accuracy or authentication (Tier 3).
Advanced analysis, trend recognition and health attestation features of HDDE will be available only for the Tier 1&2 levels, while Tier 3s will be used for reference only.
6. Conclusion
To develop Preventive Healthcare level eco-systems for InsurTech, Asia needs to have its own building- block technology like we have demonstrated, we are also building rule-based HDDE with Dr. J Rahmel (see appendix) to complete the eco-systems while most companies are doing only one or the other.
COMMENT