July 24, 2014 12:00 am JST

Alzheimer's screening might be in the (semiconductor) chips

TAKASHI KUROKAWA, Nikkei staff writer

TOKYO -- Kazuaki Sawada has a dream: One day, a quick test will be able to detect the early signs of Alzheimer's disease from a single drop of blood.

     Sawada is a professor at the Toyohashi University of Technology, where the focus is electrical engineering, not medicine. He is part of a movement trying to harness the technologies of the semiconductor industry for medical applications.

     Scientists are researching chip-based methods for detecting molecular markers related to disease by assaying samples and measuring for tiny changes in electrical signals. And they are fleshing out a concept for continuous health monitoring using chips attached to the body.

     Detecting the warning signs of Alzheimer's disease from a drop of blood is still a long way off. But Sawada is moving closer to the goal.

     Collaborating with Osamu Takikawa, head of the Center for Development of Advanced Medicine for Dementia at the National Center for Geriatrics and Gerontology, Sawada has developed a way to detect a peptide closely related to Alzheimer's in blood.

     The peptide is amyloid-beta, which accumulates in the brains of Alzheimer's patients and is thought to be associated with the death of brain cells and the emergence of memory loss and the other symptoms of dementia.

     Since trace amounts of amyloid-beta seep into the blood, early detection of the peptide could give doctors a chance to begin early treatment and retard the advance of the symptoms.

     Sawada, Takikawa and their colleagues have developed a semiconductor sensor to detect amyloid-beta. The detector is made from an orderly array of 16,000 elements fixed to a substrate measuring 3.3 sq. mm and covered by a layer of antibody molecules that selectively bind to amyloid-beta. When a blood sample is trickled onto this device, the antibodies grab hold of any amyloid-beta in the sample, and the device measures a slight change in voltage.

     Other, more conventional technologies to detect this peptide in the blood are under development. But detection takes five hours at minimum, and the testing requires a sample of several milliliters of blood. That means the patient would need to go to a hospital or clinic. What is more, expert knowledge is required to operate the measuring equipment.

     What Sawada and Takikawa have in mind is a test that people can do themselves at home in 10 minutes. A potential sufferer could use a simple tool to draw a drop of blood, put the drop on a sensor and insert that into a measuring device to get results.

     Blood contains all sorts of factors associated with various diseases, so by changing out the antibodies the same technology could be used to detect a variety of molecules and uncover early signs of various diseases, including diabetes and different cancers.

     "Unlike routine medical exams, this is a way to accurately grasp alterations in the human body on an individual basis," Takikawa said. "What we have here is the trump card for individualized, tailor-made medical therapy."

Diagnosing the inside story

Over at Waseda University, a team led by professor Tetsuya Osaka is also researching the use of semiconductor sensors for disease diagnosis. Its basic principle is the same as the Sawada group's, but rather than measuring a change in voltage, their devices measure a faint change in electric current.

     The Waseda team has developed a device to detect amyloid-beta, and in collaboration with Hokkaido University has confirmed that the same technology can detect the influenza virus.

     It has also developed a semiconductor sensor to measure pH, which is an indicator of acidity and alkalinity.

     How might that be used? When the interior of the mouth is too acidic, tooth enamel can dissolve away and increase the risk of tooth decay. So the Waseda group plans to attach tiny sensors to people's teeth and monitor pH changes throughout the day. The collected data should provide clues for research into treatments for dental caries.

     Also under development is a semiconductor sensor that can measure the pH of the skin with a simple touch. Readings could then help users buy the cosmetics that are best for them.

     These medical-use semiconductor sensors owe their existence to the technologies and knowledge accumulated by the electronics industry.

     By integrating elements in high density and amplifying electric signals, even trace amounts of substances are detectable. And thanks to the mass-production technologies of the semiconductor industry, the sensors would be inexpensive to make. That is important since these devices would be used once and thrown away to prevent the spread of infectious diseases.

     Tiny chip sensors would also make continual health monitoring possible.

     Attached to the skin and connected to a wireless transmitter, chips could take blood pressure, heart rate, blood sugar and other measurements, then send the data to a smartphone, which in turn could transmit the information via the Internet to a hospital or clinic. Doctors could then look at this information to provide guidance, or prompt a user to come in for treatment if there are signs of disease.

     The medical world is excited about the possibilities that semiconductor sensors present, but there are still problems that need overcoming and improvements that need to be made.

     One challenge is to develop ways to make the blood components flow easier. Another is the need for antibodies that bind more effectively to the various disease factors of interest.

     And then there is the lack of enthusiasm in the semiconductor industry.

     At least in Japan, semiconductor makers remain dubious of the business prospects. Sawada said that when he visits companies and explains his vision, the question his hosts always ask is if these medical sensors will lead to big sales.