Electronic Integration Options for Implantable Medical Device Platforms



The health benefits in our future might just rest on how hale, hardy (and cost-effective) developers, including start-ups, can make their implantable medical device proof-of-concepts—and that warrants a close look at ASSP chips for help with size, power and development time hurdles.

A large number of smaller medical device companies are entering the rapidly growing field of implantable medical devices. Applications, including neurostimulation, for these devices encompass everything from alleviating back pain to the treatment of sleep apnea and autoimmune disease. While the potential uses of this technology are exciting to contemplate, the electronics design challenges are daunting.

Fig 1 implantable medical designs leveraged to reduce size-power 2
Figure 1. As we progress from Commercial Off the Shelf to Application Specific Standard Product (ASSP) to Application Specific Integrated Circuit (ASIC) closer targeting of implantable medical device size and power requirements becomes possible.

Competing Challenges

The two major challenges, beyond the government approval process, are system power and size. Many companies, particularly start-ups, must develop their proof-of-concepts under resource and time pressures. The challenge to complete the proof-of-concept in a timely fashion competes with that of optimizing the devices size and power requirements. Developers turn to commercially available off-the-shelf components to expedite proof-of-concept, but this approach often results in a proof-of-concept vehicle that is too big and too power consuming to be of practical use. Meanwhile, the task of reducing power and size while maintaining performance and features lingers until the next generation. Ultimately, the need to reduce size and power in the device becomes a critical path requirement.

A power requirement that is low makes it possible for battery capacities to be smaller, thereby reducing overall system volume. Achieving power requirements that are as low as possible even helps in the case of rechargeable batteries, with the lower power enabling smaller device size and reducing the frequency of recharging for the patient. Smaller-sized implants offer several advantages to the surgeon and patient. As implants become smaller (and the industry is heading in the direction of implantable devices), implantation at the point of therapy becomes possible. This approach lessens the need to implant a device in the chest or abdomen and then run long leads to the point of therapy. Smaller implantable devices can also reduce surgical complexity, thus helping to lower risk and saving both time and money.

One option for meeting implantable medical device size and power challenges is the use of an Application Specific Standard Product (ASSP) chip. It allows targeted customization for the specific medical implantable application being developed (Figure 1).

Critical Parameter Optimization

Typical commercial components are often overdesigned to meet the needs of several high-volume applications. ASSP devices, by contrast, target one or two applications. Minimizing the number of targeted applications reduces the number of design requirements. It also allows for optimization of the most critical parameters, in this case, size and low power.

Many off-the-shelf components require working temperature ranges of –40° C to +85° C. An ASSP for an implantable device need only work between 10° C to 50° C pre-implant and 35° C to 40° C post-implant. As well, performance at relatively high frequencies, certainly tens if not hundreds of megahertz, is expected for the bulk of off-the-shelf components. Implantable devices, on the other hand, work in a frequency space most typically at 1 KHz or less. Also, data converter precision demands are relatively modest for implantable devices, typically less than 12 bits. Along with moderate precision of the converters, the sampling rates of these converters can be quite low. All of these specific requirements allow the ASSP to be designed with lower power and smaller die size.

Cutting down on power requirements by choosing an ASSP device means battery capacity can be smaller. And as battery sizes, whether rechargeable or primary cells, are proportional to battery capacity (milliampere-hour or mA-h), this choice leads to a reduction in overall system size.

However this desirable size reduction has to be balanced against a drawback—the ASSP might not contain all the desired functionality. Thus, the level of integration may not be optimal. It’s important to keep in mind, though, that the ASSP device already exists. So it’s possible to get a “somewhat” optimized device for your application without the long cycle time and development costs of a full custom device. Taking a route that avoids a lengthy and costly development cycle has particular appeal to smaller start-ups out to prove their therapies with limited time and money.

Leveraging a Custom ASIC

An alternative to using an ASSP chip would be to develop a custom Application Specific Integrated Circuit (ASIC). Besides encompassing the ASSP advantages discussed earlier, the ASIC brings additional benefits that the ASSP likely will not offer. Working with a custom device lets the developer define the exact functionality the therapy requires. The system developer can also define the desired level of integration, choosing to integrate (or not) a number of features in order to optimize functionality, size and power. Features in play include power management, non-volatile memory, custom digital processing and/or a microcontroller—all on a single piece of silicon. With an ASSP, though, the level of integration has already been defined.

Of course all this flexibility with an ASIC comes with some significant tradeoffs. A custom ASIC will take time to design. In general, from design concept to first silicon can require 9-12 months. In addition, there can be substantial non-recurring engineering costs associated with ASIC development. Both of these concerns could make a custom ASIC challenging, particularly for a smaller medical start-up. To alleviate a portion of these concerns, it may be possible to modify an ASSP platform device. This might allow a timelier and more cost effective approach, thus creating a “quasi” custom ASIC (Figure 2).

Fig 2 Path of Implantable Medical Device Development
Figure 2. There are three primary electronics design paths in implantable medical device development.

Summing It Up

Implantable medical device system developers must make tradeoffs between critical device requirements and business realities. At the heart of this is considering development time and costs. Commercial off-the-shelf components have been perceived as initially offering the shortest development time and lowest costs. Today’s smallest implantable devices, however, could require a customized approach to meeting power and size requirements.

Although ASICs have significant upfront costs and long development times, the custom ASIC may be the best way to meet device requirements. Specifically due to the long cycle times, holding off on ASIC development is a bad idea. It can lead to the ASIC becoming the critical design path instead of just an optional one. It can also result in holding up the start of clinical trials.

Leveraging an ASSP during proof-of-concept development offers an intermediate step. By definition the ASSP is immediately available with no associated non-recurring engineering charges. Unit pricing is often higher than commercial commodity-type products. However, the cost is still relatively inexpensive compared to undertaking ASIC development. An ASSP can move the design closer to meeting final device requirements and can even offer an acceptable solution for production.

One example of an ASSP chip available today is Cactus Semiconductor’s CSI021. The device is a 4- channel programmable pulse generator targeting neurostimulation applications. The individually programmable parameters of the current pulses provide great flexibility to the system developer for implementing unique stimulation therapies. The ASSP can also be had with the company’s EB1CSI021 evaluation board. The evaluation board is provided with two Cactus Semiconductor CSI021 ASSPs, which can be synchronized.


James_McDonaldJames McDonald is the co-Founder and president of Cactus Semiconductor Inc., formerly known as Cactus Custom Analog Design. James has over 30 years of semiconductor and electronics industry experience with companies such as Medtronic, Intel, Motorola and BBN. Prior to co-founding Cactus Semiconductor, James was the design manager of Medtronic’s Semiconductor Group. He has an M.S.E.E. degree from Boston University and a B.S.E.E. degree from Arizona State University.

Share and Enjoy:
  • Digg
  • Sphinn
  • del.icio.us
  • Facebook
  • Mixx
  • Google
  • TwitThis

Tags: ,

Extension Media websites place cookies on your device to give you the best user experience. By using our websites, you agree to placement of these cookies and to our Privacy Policy. Please click here to accept.