Exploiting the microelectronics toolbox for healthcare applications

For decades now, the semiconductor industry has created ever more powerful chips for computation and communication. Maximal scaling, complex integration, and extreme multiplexing have enabled this evolution. It has set the trend toward ever more powerful computers and enabled mobile…

For decades now, the semiconductor industry has created ever more powerful chips for computation and communication. Maximal scaling, complex integration, and extreme multiplexing have enabled this evolution. It has set the trend toward ever more powerful computers and enabled mobile devices with ever more functionalities. In the same way that chips have revolutionized the electronics industry, chips are now starting to change the face of healthcare.

The capabilities of semiconductor technologies for healthcare applications are enormous. Semiconductor technologies are one of the few technologies that can provide tools that operate at the same scale as biology. They allow designers to design microsized tools that interface with biology.,They can integrate these small tools into heterogeneous applications, analyzing orders of magnitude faster and more intelligently than the tools currently used. Instead of having one bulky tool for every test, engineers can integrate several functionalities on a single chip. Ultimately, these functionalities will be executed using nothing more than a smartphone, or designers can exploit the expertise built into ultralow-power circuits and ultralow-power sensors to create wearable devices that allow for continuous and long-term monitoring of physiological and biological signals. With such instrumentation, the medical device industry can initiate a shift toward a more personalized, preventive, predictive, and participative healthcare system.

Semiconductors have the potential to reduce the cost of healthcare in the same way they reduced the costs of computing and communication. Take the case of the rapidly advancing field of genomics. The cost of sequencing a human genome has dropped precipitously over the past decade. While the initial phases of this decrease in cost were achieved through conventional technologies (typically fluorescence imaging), the newest generation of genome-sequencing tools employ state-of-the-art complementary metal-oxide-semiconductor (CMOS) circuits and specialty devices.

Life sciences and healthcare companies are increasingly interested in customizing CMOS technologies to turn them into useful devices for improved healthcare. This article presents several examples to illustrate the enormous potential of customized chip-based technologies in a variety of healthcare and life sciences applications.

Customizing Chips for Use in Life Sciences


Customizing Chips at the Molecular Level. Many of the tools used in life sciences, such as DNA sequencers, are expensive, bulky, and complex to operate. In many cases, testing involves a number of these tools at different locations. Technology has played a key role in driving down the cost of sequencing a genome from hundreds of millions of dollars to a few thousand dollars. Technology will continue to play a key role in this progress as we aim to microsize DNA analysis tools by building chips with millions of sensors that can analyze strands of DNA in parallel. The same could be done for protein and metabolite analysis, and all these tests are becoming increasingly important in diagnosis, patient monitoring, and basic medical research. One of the key challenges is to integrate a variety of sensing technologies into a small footprint while also routing fluids in a controlled way.

There is growing interest from life sciences companies to jointly develop customized chips for DNA analysis. For example, in collaboration with Panasonic, imec has developed a fully integrated device to perform quantitative molecular diagnostic tests. The device is a chip about half the size of a credit card that performs fast, simple, and sensitive detection of genetic markers, specifically single-nucleotide polymorphisms (SNPs). Detecting these SNPs makes it possible to check for hereditary diseases or investigate a person’s response to treatments. The device itself provides a versatile platform that automates all stages of a polymerase chain reaction (PCR) experiment, from sample preparation through DNA detection and quantification. The platform houses microfluidic conduits, filters, and mixers, enabling seamless integration of sample preparation modules with miniaturized PCR cyclers and sensing elements onto a single disposable device. This design is possible because all elements are processed onto a silicon substrate, which further enables low-cost, mass-manufacturable point-of-care devices. The technology is expected to be used for personalized therapy, and could spur widespread adoption.

Customizing Chips at the Cellular Level. Today, the production of cells from stem cells for therapeutic use is still in its infancy, but promising work indicates that this type of cell therapy will play an important role in the curing of many diseases that are today not curable, such as diabetes and heart disease. To give the cell therapy community new tools to improve the cell production processes, a new microscope implements the nonscalable parts of the microscope (i.e., the optics) in software, in what is the digital equivalent of a microscope. These microscopes are based on ultrasmall image sensors, lasers, and the necessary circuits and software to create and analyze the images. The resulting system, called a lens-free microscope, lends itself to high throughput inspection of cell production. In addition, imec is developing advanced biosensor platforms based on various transduction principles, such as optical and electrical, that can be integrated into bioreactors for online or offline sensing of cellular products or growth-medium components.

Customizing Chips at the Neural Level. There is a growing interest among neuroscientists to use chips that directly interface with the brain. These microsystems or neural probes measure the activity of many neurons simultaneously at high spatial and temporal resolution across multiple-length scales. They allow researchers to build an understanding of the brain circuitry at increasingly larger scales. Ultimately, they will allow them to develop better therapeutic approaches to treat neurodegenerative diseases, such as spinal cord injuries or Parkinson’s. Researchers are developing new generations of such probes that will allow researchers to take leaps in understanding the brain. Imec’s most recent neural probe has as many as 455 electrodes distributed along a 10-mm silicon shank, with amplification under each electrode. The probe also includes electronics for signal conditioning (amplification, filtering, multiplexing) and digitization. These are directly integrated in the probe. Hence, it removes the need for heavy and expensive bench-top equipment for data acquisition and extensive cabling.

Exploiting Micro- and Nanoelectronics for Wearable Health Products

The expertise with ECG readout was further exploited to build a low-power intracardiac signal processing chip for the detection of intracardiac ventricular fibrillation. The new chip is an step toward next-generation cardiac resynchronization therapy options, which rely on a robust and accurate heart-rate monitoring of the right and left ventricles and right atrium. The chip delivers signal processing functionalities and consumes only 20 µW when all channels are active. This extreme low-power consumption is required to further reduce the size of cardiac implants and improve the patient’s quality of life.

Wearable Products for Brain Monitoring. A similar evolution is ongoing in the neurological space, where wearable electroencephalography monitoring systems are taking EEG outside the hospital environment. The team at imec recently developed a prototype wireless EEG system in collaboration with Panasonic. At the heart of the system is a low-power eight-channel EEG readout with active electrodes and built-in continuous impedance monitoring. In other words, the system can continuously record eight-channel EEG signals while concurrently recording electrode-tissue contact impedance. The read-out chip has been integrated into a miniaturized system with processing and wireless communication functionality. The result is a wearable headset with dry electrodes, which enables EEG recordings outside the hospital with minimal set-up time. The autonomy of the system ranges from 22 hours (8 channels of EEG with electrode-tissue contact impedance) to 70 hours (1 channel of EEG only). This demonstrator can be used for research and development of new EEG applications that would benefit from a wireless system with dry electrodes, quick and easy set-up, and long-term monitoring.

Wearable Products for Stress and Weight Management. The wireless EEG system can lead to improved and earlier diagnostics, but that alone will not be sufficient to solve the healthcare challenges we face. In the future, healthcare will have to focus increasingly on prediction and prevention of various health conditions in order to lower costs. Wearable tools can empower people to take control of their own health, but wireless medical devices can help prevent diseases only if patients change their behavior as a result of using them.

Managing fitness and stress are two key health challenges that must be tackled, and the program on BANs run by Holst Centre and imec seeks to address them. A good starting point for managing fitness is to provide people with tools that allow accurate quantification of their habitual physical activity in an ambulatory day-to-day setting. In a pilot study, researchers have shown that a wearable multiparameter patch can accurately measure energy expenditure. A comparison of indirect calorimetry with a single-lead ECG patch with built-in multiaxis accelerometer developed through the program has given further insight in the accuracy of the device. Wearable sensors are also used to measure emotional response. Imec previously developed a wireless sensor monitoring respiration, skin temperature, heart rate variability, and skin conductance as a means to assess personal stress levels in daily conditions such as work. Recently, researchers there built a second-generation wrist-watch-style wireless sensor for improved stress monitoring. The sensor monitors galvanic response, relative humidity, and temperature as well as activity.

Conclusion

This article has discussed several examples demonstrating the enormous potential of chip-based technologies in healthcare and life sciences applications. In life sciences, the deployment of customized-chip technology for DNA sequencing, cell production, and neural probing can revolutionize diagnostics, target treatment, and enhance fundamental understanding of diseases. In BANs, technology can improve longitudinal diagnostics in the cardiac and neurological space. Tools for assessing stress levels and energy expenditure allow people to manage their health and, as such, contribute to a more preventive and predictive (and lower cost) healthcare model.

As semiconductor technology continues to advance, there will be even more opportunity to use these advances in the healthcare sector. Technologies for low power devices, miniaturized sensors, high speed detectors and many others have utility across multiple industry sectors, healthcare being one of them. However, we are still in a period where medical applications of semiconductor technologies often follow behind technology advances. As more applications make use of chip-based technologies, we envision a future in which the needs of the healthcare industry will become the driver of chip technologies. This will only happen through close collaboration between technology developers and application scientists but we see this link growing stronger everyday.