UW prof focuses research on sensor technology

WATERLOO, Ont. -- Studying the potential uses of sensors, including in the medical field, gained a new focus at the University of Waterloo last fall with the launch of a chair in "sensor technology" held by a distinguished researcher, UW Prof. Arokia Nathan, electrical and computer engineering.

Sensors can convert an optical signal (or anything that is radiant, e.g., visible light) into an electrical signal that can be recognized, stored in a computer and later, transmitted. They take information from the analog world and make it computer compatible. Thus sensors can serve as key components of information technology.

The technology includes electronic devices that can tell us things we simply can't see with our eyes; for example, they can tell us where or in which position an object is in the dark (using infrared that can't be seen by human eyes); they can also follow, and record in great detail, things that move too quickly for the human eye.

Much of Nathan's research involves developing new "imaging" technologies; for example, optical imaging that can be used with document scanners, digital copiers, fax machines, position and motion detectors, or X-ray imaging that can be used in space telescopes, for digital image capture in biomedical applications, as well as materials and structures testing . . . including the non-destructive testing of materials such as an aircraft wing, for signs of fatigue and corrosion.

Over the past 10 years, Nathan has been particularly concerned with "amorphous silicon technology" used in liquid crystal displays such as in laptop computers. "There could be promising additional applications for amorphous silicon technology besides large area displays, including high-definition television," he said.

Amorphous silicon involves an arrangement of silicon atoms that are random rather than periodic (as in a crystalline silicon chip) and it can be put on top of glass. An "amorphous" crystal is actually not crystalline. Amorphous silicon technology promises significant commercial advantages including lower manufacturing costs, higher reliability, and the ability to create very large scale electronics on glass, as opposed to crystalline silicon, substrates.

Nathan said one important feature of glass chips involves the possibility of putting far more transistors on a sheet of glass than can be done on a pentium (silicon) chip. A glass "chip" can be many times larger than a pentium chip . . . it can be virtually as large as needed (for example, more than a foot square).

Amorphous silicon chips, he added, have a further advantage over individual silicon chips; because the latter are so small, and so many of them have to be brought together, "packaging" becomes a problem thus limiting its range of applications in large area imaging.

"The silicon chip designer has to deal with a myriad of connections required when thousands of tiny devices are brought together," he noted, "whereas with amorphous semiconductors you can have a continuous array over the whole area of the chip, so you eliminate one major packaging issue."

The success of this research could help place Ontario at the leading edge of an emerging industry. Nathan and his research group, along with the Waterloo company DALSA Corp., are currently doing basic research that could contribute to the development of an electronic camera based on amorphous silicon technology for use in digital radiography.

Amorphous silicon-based radiography offers several advantages over conventional X-ray technology. Images can be displayed almost immediately on a computer screen and improved through software, enabling quick and effective diagnosis; images can be shared electronically with radiologists who are not on-site; and images can be stored electronically, eliminating the need for massive X-ray photographic archives.

Nathan said ultra-high speed, which can be tremendously important for some chips, is not important for the applications for which his amorphous silicon sensors are intended. In any event, the speed of the display from such a detector is much faster than the speed of the human eye . . . so amorphous silicon sensors are as fast as one could possibly hope for in a wide range of applications.

As for the application of amorphous silicon technologies to radiography, he and his associates feel it should be possible to use amorphous semiconductors to do virtually everything conventional X-ray film now does -- show what the inside of the body looks like.

But semiconductor technology promises to do it better, do it more cheaply, and perhaps also do it more safely . . . without a need to subject patients to the high energy fields (radiation) required to generate an X-ray picture. This is because an amorphous silicon detector could be far more sensitive than film, and thus would make it possible to use lower levels of radiation, reducing the risk of radiation damage to the patient.

As well, the picture that an amorphous silicon device would produce would not involve a photographic plate; the information would be stored digitally in a computer . . . and could be electronically transmitted through communication networks.

Inevitably, telediagnosis is going to become more widely used in the future; someone with a medical problem in James Bay, for example, would not have to travel to Toronto to have an X-ray picture taken so a highly-skilled radiologist could study it. Digital data could be transferred over a telephone line from a computer in the remote location to one in a radiologist's office in Toronto, or anywhere else in the world. Thus, a doctor in Australia could have virtually instant access to the "X-rays" of a Canadian visitor who had refractured a bone during the visit to that country.

One of the main challenges in digital X-ray imaging is in mammography, for the detection of breast cancer. Mammography is important for early detection, and it is done at regular intervals.

Yet some have concerns over this continuous application of conventional X-ray technology since high energy rays are involved. As well, conventional X-ray mammography may involve discomfort for some women. Breasts must be flattened, sometimes painfully, in order to achieve a reliable picture; that is, the tissue being X-rayed has to be of uniform depth.

An amorphous silicon system promises to make mammography much less unpleasant, Nathan said, though he added: "We still have a lot of optimization work to do in terms of pixel resolution and array performance. But if we can arrange the electronics so they will curve over a curved surface -- if we can develop an 'array' for a curved object -- then our technology could prove less painful or inconvenient than X-rays. We are only now starting to work on this challenge. It involves lots of technological issues. The area of curved or 'warped' electronics is a whole new field. What we, and others, are trying to do is deposit amorphous silicon on plastic substrates, rather than glass."

This could be helpful in achieving warped electronics (detecting over curved surfaces) because the substrates themselves could be curved. Because of the curvature, the waves entering the body travel different distances, and one must be able to understand the geometrical implications of this, to correct for the curvature effect. Thus plastic substrates could make it easier to do radiography to detect breast cancer. Nathan noted also that the same basic technology could also be used to detect corrosion and fatigue in the curved wing of an airplane, or in many other applications.

"A lot of work still needs to be done," he cautioned, "if we want the findings from this technology to be sufficiently accurate, if we want to reduce the energy or radiation risk to a minimum, and as well, if there is to be a reduction of the parasitics (the input of energy without return in terms of a radiograph). If we are to achieve these goals, lots of engineering issues must be solved."

Nathan is working closely with X-ray researchers at the University of Western Ontario's medical school. The school is renowned for the excellence of its group of radiologists, whose expertise includes ultrasound and MRI (magnetic resonance imaging) as well as X-ray technology. He said that there has been a tremendous increase in interest in amorphous silicon sensing technologies, not only among Western's radiologists, but all over the world, within recent months.

"There are many potential 'niche' areas for this technology, which could permit small and modest-sized companies to compete very well with the very large ones," Nathan said, adding that researchers have only started looking into this technology in terms of its medical applications.

"The amorphous silicon research involves many facets. We have to find out how to make the detecting device, and then we have to find out how to make an electronic switch that can read the detector. In other words, every component has to be developed from scratch," he said.

Government support for the new chair comes from the federal government's Natural Sciences and Engineering Research Council and from Communications and Information Technology Ontario, one of the Ontario Government's centres of excellence.

The work is also receiving support from UW. Nathan said that the reason the federal and provincial governments are supporting the chair is because the research has the potential to spell tremendous benefits to small and medium-sized Canadian high tech industries.

"In addition," Nathan pointed out, "it will provide trained, multi-disciplined graduate and post-graduate manpower, and that will strengthen Canada's engineering resource pool and increase its global competitiveness."

The UW team's current studies, he noted, are built on work done at UW by Prof. Savvas Chamberlain years earlier. A few years ago, Chamberlain, now a retired professor, set up a highly successful Waterloo manufacturing company, DALSA, to put into use the basic CCD (charged couple device) technology he had earlier developed. "What we are doing now is using a lot of the infrastructure laid in place by Dr. Chamberlain," Nathan said.

He added that other research teams interested in the use of amorphous semiconductors in digital X-ray imaging, include a group at the Xerox Corporation in Palo Alto, Calif., working on a phosphor-based approach in which the phosphor converts X-rays to visible light, which is then detected by amorphous silicon sensors.

"Their system has been optimized and has shown good success only at high X-ray energies, whereas we are more interested in low-energy problems, including new detection schemes," Nathan said.

Another approach is being followed by a group at the University of Toronto, based on amorphous selenium technology and the use of very high voltages (eight- to 10-thousand volts).

"At Waterloo we are looking at very low voltages (as low as five volts)," Nathan said. "That is all we need to get a sufficient signal so our sensors can detect it. Even at that however, we continue to be interested in ways to improve the sensitivity of our amorphous semiconductors, and also in various ways of designing detectors."

He said a major concern is over the stability of amorphous semiconductors; their soft, disordered crystal structures and the presence of hydrogen in the material can cause them to "drift" with time.

"The idea is to find ways to live with this instability and in fact, we have been coming up with ways to do so. We find of course that our low-voltage environment is very helpful in terms of keeping instabilities to a minimum."

As the first occupant of UW's new sensor technology chair, Nathan is heading a sizable research group that includes three post-doctoral fellows, five PhD students and three master's students, as well as an engineer. The group has substantial research resources including laboratory facilities for chip making that came into existence on the UW campus several years ago, partly in support of the research by Chamberlain.

Total funding for the chair will be in the amount of $1.725 million over the next five years including the considerable support from DALSA.

The company employs about 200 persons, a large percentage of whom have science and engineering backgrounds, and manufactures and markets worldwide an array of sensor imaging products, including high speed cameras. It is a modest-sized company and a venture into the area of radiographic equipment would involve serious competition from the X-ray equipment manufacturers now dominating the field.

Still, at DALSA's annual meeting in May, Chamberlain, company president, referred to the creation of the new UW chair and to Nathan's research, and indicated that its success could contribute significantly to his company's further progress over the next few years. He said DALSA currently has annual sales in the $30-million area, and expects to hit the $100-million mark by 2002.

Contact:

Prof. Arokia Nathan, (519) 888 4803; E-mail: anathan@venus.uwaterloo.ca

Written by Bob Whitton for the UW News Bureau, (519) 888-4444

UW experts/releases: http://www.adm.uwaterloo.ca/infonews/

Release no. 113 -- July 8, 1998