Paper Reading #4: Implanted user interfaces

Intro:
Title: Implanted User Interfaces

Reference Information: CHI '12 Proceedings of the 2012 ACM annual conference on Human Factors in Computing Systems, Pages 503-512

Author bios:

Christian Holz is a PhD student at the University of Potsdam, where he works on ultra-small, ultra-mobile touch devices.

Tovi Grossman is a Senior Principal Research Scientist at Autodesk; he focuses on Human-Computer Interaction, especially multi-touch projects and miniature projects. He received his PhD in Human-Computer Interaction from the University of Toronto.

George Fitzmaurice is the Head of User Interface Research at Autodesk and has contributed to over 30 patents in his 13 years there. He got his bachelor's at MIT, his master's at Brown, and his PhD at the University of Toronto.

Anne Agur is a faculty member in the Musculoskeletal Anatomy Laboratory at the University of Toronto; she received both her M.S. and her PhD there. She focuses on muscle form and function and clinically applied neuroanatomy.

 

One of the devices used in the experiment

Summary: Implanted user interface devices were tested in two scenarios. In the first, the devices were implanted just under the skin in a cadaver arm. The input devices (three types of touch sensors and two types of hoer sensors) were tested for responsiveness with dropping pistons and a research assistant's hovering hand. Additionally, the output devices (an LED light, an audio speaker, and a vibration motor) were tested both outside for reference and implanted to determine their effectiveness. User tests were then conducted by placing the devices under fake skin (silicone) for qualitative evaluation by the users. Input and output methods are the most difficult challenges of this project; basic input and output can be accomplished with the devices used in this experiment, but more detailed interaction would require more and different sensors. Additionally, the devices are capable of Bluetooth communication and wireless charging.

Related Works:

Ultrasonic transcutaneous energy transfer for powering implanted devices  - Ozeri, Schmilovitz
An investigation of ultrasonic transcutaneous energy transfer for recharging implanted devices. The implanted user interfaces use inductive charging technology, but both are wireless charging systems.

Body-based interfaces - Kim, Han, Yang, Cho
A study of 4 types of body-based interfaces: body-inspired metaphor, body-as-interaction-surface, mixed mode, and object mapping. Also an implementation that requires no other interface.

Wearable Computing - Roggen
An overview of three areas in which wearable computing may aid in robotics research. Wearable computing is a predecessor to implanted computers.

Multi-sensor Activity Context Detection for Wearable Computing - Kern, Schiele, Schmidt
A hardware platform for wearable sensors to detect activities is presented; this is a non-intrusive method for getting input.

Weakly Supervised Recognition of Daily Life Activities with Wearable Sensors - Stikic

A method for unobtrustively recording information about the use of wearable sensors is introduced. Could be useful for investigating everyday use of implanted computers.

Linux watch: Hardware platform for wearable computing research - Kamijoh, Inoue, Kishimoto, Tamagawa
A watch-like computer with Linux has been developed for information access; its ubiquity and multi-modal interface is similar to that of the implanted devices.

Embedded Human-Computer Interaction - Baber, Baumann
This paper explores the possibilities of ubiquitous computing and interaction with domestic devices. Ubiquitous computing is a goal of the implanted devices.

Wearable Sensor-Based Hand Gesture and Daily Activity Recognition for Robot-Assisted Living - Zhu
A system for natural CHI for elderly and disabled people in a robot-assisted living environment is developed; may contribute to user interfaces for implanted devices.

Overview of Bottom-up Nano Electronics Materials and Its Application - Singh, Sangita, Sharma
An overview of materials used in nanotechnology; nanotechnology could easily facilitate implanted devices.

Based on relevant works, this work is very novel and in its early stages.

Evaluation: The authors used a quantitative, objective evaluation of device effectiveness and accessibility. Initially, the use of the input devices, output devices, and Bluetooth communication were tested outside of any implantation or implantation simulation. Once they were implanted, their effectiveness was quantitatively measured. For example, the brightness of the LED was measured by a camera at various power levels and the force needed to activate the touch sensors was measured using dropped pistons. In all cases, device effectiveness was reduced by the skin (understandably), so the LED required more power to be visible, etc. With the implantation simulation, a qualitative, subjective evaluation was used to determine how much the users liked it.

Discussion: This technology is both gross and engrossing. The potential for implanted user interfaces is huge; phone calls could eventually be made without phones, Wikipedia searches could be done as quickly as asking a question, and more. Personally, I find it fascinating, especially because I have a pacemaker. These technologies could vastly improve current medical devices. This is a very new area of research and the possibilities are almost endless.