In 2017–2018, I undertook an engineering capstone project of designing dolls for a residential children’s hospital. I worked in a team of three students, and our clients, two nurses at the hospital, wanted the dolls to use in play therapy with residents and to explain complex medical diagnoses to families. Some use cases they mentioned:

• Almost all residents are non-verbal. Dolls that had their assistive devices would help residents more accurately communicate discomfort or other feelings, without putting their real assistive devices at risk.
• Some residents with complex needs come to stay in this hospital soon after birth. Siblings of a resident might meet the doll before they meet the child; the doll could be a more gradual way to explain their medical needs to family.

I acted as visual design lead throughout the project. I also served as project manager in Spring 2018. Recurring duties included researching material and manufacturing choices, writing reports and procedural documentation, and presenting our work to our classmates and clients.

Our clients provided us with the ten most common diagnoses of residents. I drew sketches incorporating all ten, as well as common assistive devices like tracheostomy tubes and gastrointestinal feeding tubes (g-tubes). These sketches were used to select designs to develop further.

For speed, we decided to only create new heads and modify existing doll bodies. Part of that process involved finding pigment ratios (via trial and error!) to color polyurethane plastic and match the bodies we chose.

I modified 3D scans of doll heads in Meshmixer. These represent microcephaly, hydrocephaly, and Down Syndrome. We checked in with our clients at every stage of the process to make sure our representation was accurate and sensitive.

These models were 3D printed. The team used silicone to make block molds from the prints, then cast heads in polyurethane (see pigment matching tests, above). The dolls had to be made in non-porous materials for purposes of easy sterilization.

We also 3D-modeled assistive devices (g-tube centers), combined these with surplus medical equipment, and tested secure attachment mechanisms. One of the most vivid stories we were told during the ideation process involved the terror nurses could feel at finding a loose g-tube on the floor—had it come from one of their prototype dolls or from a child? We developed verification and validation protocols to make sure that couldn’t happen with our prototypes.

Finished, usable prototypes. Our clients were delighted with the work we did over the course of the semester.