Sensalert

Wearable | IoT | Health Tracking | First Responder
Sensalert is a wearable health tracking application that enables monitoring of a target population’s (first responders) health status in real-time.
Published at ACM SIGCHI 2018
https://dl.acm.org/doi/10.1145/3170427.3188675
Project Overview
The objective of this application is to provide early warning, support decision making, and increase situational awareness for first responders by displaying comprehensive information and providing alerts at the right time and place. The goal is to address how to best design a health tracking application to
1) provide the right information to increase situational awareness, and
2) assist with decision making for protecting people at high risk of health-related problems.
My Contributions
I am the only user experience designer and researcher working closely with a group of software engineers, data scientists and domain experts. I identified the user needs, design principles, and created the health tracking applications aimed at increasing situational awareness of individuals as well as groups of people. This is the first application designed to use a wearable to monitor vital signs of first responders in action.

Clickable Prototype

https://invis.io/KHP7536Y2MC#/263494303_Group_Copy_5
The target users for this application are the analysts who monitor the situation for a group of in-action first responders. The analyst's job is to protect the first responders' health by warning or pulling them out of the operation if necessary.

After interviewed with 6 analysts, here are the user requirements I gathered.
Requirement 1
: Depending on the tasks, different analysts have different opinions on what is an important feature for them. For instance, sometimes one variable is more important than another depending on the field operational task and/or risk environment.
Requirement 2: The data from the group and from individuals are both important to view. For instance, if the average data from the group is abnormal, it is more likely to be caused by something in the environment rather than an individual-based health-related problem.
Requirement 3 : How the alerts are triggered is very important. The analysts’ mission is to protect the workers in the field, so they don’t want to miss any potential risk. However, too many false positives could potentially confound real alerts. Therefore, giving the analysts a certain degree of control on the alert settings is the key.
Requirement 4 : As the data is updated in real time, the ability to rewind, mark, and revisit the historic data and alerts is required.

The Process

Skeches

One of the ideas. The user feedback was the visualization is complicated and hard to follow.
The idea that the user preferred. The main reason is the visualization is very straight forward and easy to follow.

High fidelity design

First iteration
The team thoughts it's too complicated to develop with the tight timeline we had.
The final design
Greatly simplified the visualization. Heavily using color to indicate the threat level. The analyst can know who is at danger at a glimpse.
View Other Projects
GridPIQ
Power Grid | Renewable Energy | Screening and Analysis Tool | Data Visualization
WatchOwl
AI | Data Visualization |  Social Media Analytics
EyeSea
Wearable | IoT | Health Tracking | First Responder