he communication process. These include alternative language representation systems based on semantic (iconic), alphanumeric, or other codes; and prediction systems, which provide choices based on previously selected letters or words Some individuals can produce speech, but it is dysarthric and very difficult to understand. Yet the utterance does contain information. Can this limited information be used to figure out what the individual wanted to say, and then voice it by artificial means? Research labs are now employing neural network theory to determine which pauses in an utterance are due to content (ie, between a word or sentence)and which are due to unwanted halts in speech production 119.5 Appropriate Technology Rehabilitation engineering lies at the interface of a wide variety of technical, biological, and other concerns. a user might(and often does) put aside a technically sophisticated rehabilitation device in favor of a simpler device that is cheaper and easier to use and maintain. The cosmetic appearance of the device(or cosmesis)sometimes ecomes the overriding factor in acceptance or rejection of a device. A key design factor often the appropriate technology accomplish the task adequately, given the extent of the resources available to solve the problem and the residual capacity of the client Adequacy can be verified by determining that increasing the technical content of the solution results in disproportionately diminishing gains or escalating costs. Thus, a rehabilitation engineer must be able to distinguish applications where high technology is required from those where such technology results in an incremental gain in cost, durability, acceptance, and other factors. Further, appropriateness very much depends on location. What is appropriate to a client near a major medical center in a highly developed country might not be appropriate to one in a rural setting or in a developing country. This is not to say that rehabilitation engineers should shun advances in technology. In fact, a fair proportion of rehabilitation engineers work in a research setting where state-of-the-art technology is being applied to the needs of the disabled. However, it is often difficult to transfer complex technology from a laboratory to disabled consumers not directly associated with that laboratory. Such devices are often designed for use only in a structured environment, are difficult to repair properly in the field, and often require a high level of user interaction or sophistication Technology transfer in the rehabilitation arena is difficult, due to the limited and fragmented market. dvances in rehabilitation engineering are often piggybacked onto advances in commercial electronics. For example, the exciting developments in text-to-speech and speech-to-text devices mentioned above are being driven by the commercial marketplace, and not by the rehabilitation arena. But such developments will be welcomed by rehabilitation engineers no less. 119.6 The Future of engineering in rehabilitation The traditional engineering disciplines permeate many aspects of rehabilitation. Signal processing, control and information theory, materials design, and computers are all in widespread use from an electrical engineering perspective. Neural networks, microfabrication, fuzzy logic, virtual reality, image processing, and other emerg- ing electrical and computer engineering tools are increasingly being applied. Mechanical engineering principles are used in biomechanical studies, gait and motion analysis, prosthetic fitting, seat cushion and back support design, and the design of artificial joints. Materials and metallurgical engineers provide input on newer bio- compatible materials. Chemical engineers are developing implantable sensors. Industrial engineers are increa ingly studying rehabilitative ergonomics The challenge to rehabilitation engineers is to find advances in any field -engineering or otherwise-that will aid their clients who have a disability. c 2000 by CRC Press LLC© 2000 by CRC Press LLC the communication process. These include alternative language representation systems based on semantic (iconic), alphanumeric, or other codes; and prediction systems, which provide choices based on previously selected letters or words. Some individuals can produce speech, but it is dysarthric and very difficult to understand. Yet the utterance does contain information. Can this limited information be used to figure out what the individual wanted to say, and then voice it by artificial means? Research labs are now employing neural network theory to determine which pauses in an utterance are due to content (i.e., between a word or sentence) and which are due to unwanted halts in speech production. 119.5 Appropriate Technology Rehabilitation engineering lies at the interface of a wide variety of technical, biological, and other concerns. A user might (and often does) put aside a technically sophisticated rehabilitation device in favor of a simpler device that is cheaper and easier to use and maintain. The cosmetic appearance of the device (or cosmesis) sometimes becomes the overriding factor in acceptance or rejection of a device. A key design factor often lies in the use of the appropriate technology to accomplish the task adequately, given the extent of the resources available to solve the problem and the residual capacity of the client. Adequacy can be verified by determining that increasing the technical content of the solution results in disproportionately diminishing gains or escalating costs. Thus, a rehabilitation engineer must be able to distinguish applications where high technology is required from those where such technology results in an incremental gain in cost, durability, acceptance, and other factors. Further, appropriateness very much depends on location. What is appropriate to a client near a major medical center in a highly developed country might not be appropriate to one in a rural setting or in a developing country. This is not to say that rehabilitation engineers should shun advances in technology. In fact, a fair proportion of rehabilitation engineers work in a research setting where state-of-the-art technology is being applied to the needs of the disabled. However, it is often difficult to transfer complex technology from a laboratory to disabled consumers not directly associated with that laboratory. Such devices are often designed for use only in a structured environment, are difficult to repair properly in the field, and often require a high level of user interaction or sophistication. Technology transfer in the rehabilitation arena is difficult, due to the limited and fragmented market. Advances in rehabilitation engineering are often piggybacked onto advances in commercial electronics. For example, the exciting developments in text-to-speech and speech-to-text devices mentioned above are being driven by the commercial marketplace, and not by the rehabilitation arena. But such developments will be welcomed by rehabilitation engineers no less. 119.6 The Future of Engineering in Rehabilitation The traditional engineering disciplines permeate many aspects of rehabilitation. Signal processing, control and information theory, materials design, and computers are all in widespread use from an electrical engineering perspective. Neural networks, microfabrication, fuzzy logic, virtual reality, image processing, and other emerg￾ing electrical and computer engineering tools are increasingly being applied. Mechanical engineering principles are used in biomechanical studies, gait and motion analysis, prosthetic fitting, seat cushion and back support design, and the design of artificial joints. Materials and metallurgical engineers provide input on newer bio￾compatible materials. Chemical engineers are developing implantable sensors. Industrial engineers are increas￾ingly studying rehabilitative ergonomics. The challenge to rehabilitation engineers is to find advances in any field — engineering or otherwise — that will aid their clients who have a disability