ユーザー事例

Museum of Science, Boston Creates Interactive Activities That Teach Visitors About Their Own Biology

Challenge

Engage museum visitors with interactive activities based on hand-grip strength sensors and eye-tracking hardware

Solution

Work with MathWorks engineers to develop MATLAB applications that acquire data from the hardware, process the data, and present it in visually appealing formats

Results

  • Software enhancements rapidly implemented
  • Staff empowered to continue development with training and knowledge transfer
  • Software updates based on ongoing research enabled

“The MathWorks teams were both enthusiastic and responsive. They created MATLAB applications that provide a modern, interactive interface to cutting-edge research, as well as a mechanism to easily gather, interpret, and present the results of experiments in a way that engages and informs our visitors.”

Dr. Elizabeth Kong, Museum of Science, Boston
Museum of Science attendees participating in the hand-grip activity, led by MathWorks training engineer Louvere Walker.

The Hall of Human Life at the Museum of Science in Boston features dozens of interactive stations designed to help visitors understand their own biology using their own data. One popular activity measures visitors’ grip strength and lets them compare their performance with that of past visitors of the same gender, age, or handedness. A second activity uses an eye tracker to help visitors understand human eyesight.

The Museum of Science worked with MathWorks engineers from Application Engineering, Consulting, and Training Services to build reliable software for both activities using MATLAB®.

“We have about 1.4 million visitors annually, so the robustness of our interactive activities is key, as is the ability to make data visually accessible,” says Dr. Elizabeth Kong, director of the Hall of Human Life at Museum of Science, Boston. “The MathWorks teams helped us achieve both those objectives with MATLAB, and they collaborated with us to generate custom visualizations that our visitors find interesting and meaningful.”

Challenge

The original software for the hand-grip activity, built using another programming language, had several drawbacks. First, its interface was difficult to use for some museum volunteers, who range in age from 14 to 86. Second, its output was not always clear to a general audience. Third, the museum’s technical staff lacked the knowhow to make minor fixes and enhancements to the code. The museum wanted a solution that was intuitive, accessible, and maintainable yet still used the existing grip strength sensor hardware.

For the eye-tracking station, a new activity at the hall, the museum wanted to collaborate with Dr. Peter Bex and his colleagues at Schepens Eye Research Institute, Massachusetts Eye and Ear Infirmary. Dr. Bex’s team had already developed algorithms used in eye-tracking experiments. The museum sought to build upon these algorithms for new activities, including comparing eye and hand reaction times, identifying faces in a crowd, and tracking eye movement when reading.

Solution

The Museum of Science partnered with MathWorks engineers to re-engineer the hand-grip activity and develop the eye-tracking activity.

Working in MATLAB, the engineers generated customized plots and charts from sample data, and met with museum staff to select visualizations that would be accessible to a broad audience.

Using Data Acquisition Toolbox™, they connected to a Vernier hand-grip dynamometer and read its force measurements into MATLAB. They applied regression analysis and other statistical techniques with Statistics and Machine Learning Toolbox™ to process the data. They then used MATLAB to generate plots that compared the visitor’s grip strength with that of other visitors of the same age, gender, and handedness.

For the eye-tracker activity, the MathWorks teams met with Dr. Kong and with Dr. Bex and his team to plan how best to represent the lab’s research in the activity.

Using MATLAB, Data Acquisition Toolbox, and the Psychophysics Toolbox (a set of MATLAB functions for vision research available on the MATLAB Central File Exchange), the team imported data from SensoMotoric Instruments (SMI) eye-tracking hardware. They used Image Processing Toolbox™ to apply filters and outline objects to be tracked in the images presented to visitors.

The team developed a MATLAB application that integrated with Dr. Bex’s existing MATLAB algorithms to plot response times for the tracked eye movements.

The success of the eye tracker and hand-grip activities has prompted the museum to create other interactive activities that let visitors use their own data to explore their biology.

Results

  • Software enhancements rapidly implemented. “When we’ve needed to make changes or enhancements to the hand-grip activity, the MathWorks teams have been very responsive,” says Dr. Kong. “They typically have it implemented the next week, which has contributed to the overall usability, reliability, and robustness of the software.”

  • Staff empowered to continue development with training and knowledge transfer. “One of our goals was to be able carry on development work ourselves,” says Dr. Kong. “The MathWorks teams have already provided us with onsite, customized training, and they continue to work closely with us to make sure that we are self-sufficient.”

  • Software updates based on ongoing research enabled. “The MathWorks teams built the MATLAB applications so that they can be easily extended and expanded later on,” notes Dr. Kong. “I am excited about that potential, particularly for the eye tracker, because we will be able to compare visitor data with what ongoing research is showing.”