The Future of Medicine Is Here

A recent paper delivered at ISSCC back in February really intrigued me, especially since I used to work as a medical diagnostic equipment designer.

The design described is technically pretty slick. It's an improved method of delivering medication transdermally. This design also demonstrates how designers can take full advantage of integrated analog ICs: Read down further — then try doing this design with a bunch of discrete op amps.

The paper, published as part of the 2013 IEEE International Solid-State Circuits Conference (ISSCC), is “An 87mA⋅min Iontophoresis Controller IC with Dual-Mode Impedance Sensor for Patch-Type Transdermal Drug Delivery System.” The authors are Kiseok Song, Unsoo Ha, Jaehyuk Lee, Kyeongryeol Bong, and Hoi-Jun Yoo. The authors describe an improved method of transdermal drug delivery via iontophoresis.

Some definitions will help:

  • Transdermal: literally, across the skin; so a method of transporting a drug across (or through) the dermal tissue layers without the need for pointy objects (good for people who don't like needles);
  • Iontophoresis: literally a migration caused by ions (positively or negatively charged atoms or molecules); so, a charge is induced onto the drug molecules, which are then propelled into the tissue.

By way of summary, the authors write the following:

In this paper, we present an iontophoresis controller IC with real-time monitoring of total injected charge quantity and skin condition. The proposed IC contains 32-level programmable current stimulator, a temperature sensor, and a dual impedance sensor to monitor skin temperature, contact impedance, and tissue impedance. The measured temperature and impedances, related with the skin condition, are used for adaptive charge injection by modifying the current stimulation levels through the real-time feedback path. An implemented fabric patch type drug delivery system provides up to 87mA⋅min dosage, larger than the dosage range (80mA⋅min) of commercial iontophoresis patches.

The units of measure above are milliampere-minutes. That's the product of the electrode current and time. The currents used are actually quite low (microamperes) and the time to administer the drug is typically around a few hours.

I find this especially exciting design work (not to mention good karma) — the authors have improved upon previous methods that were simplistic open-loop techniques with a voltage source. Now, they've closed the loop and monitor the current. That's just the beginning.

The device they've designed also monitors a number of conditions and adjusts the excitation current as needed. More importantly, they have designed a low-power mixed signal IC to control the excitation current, thereby making this device sophisticated, compact, and battery operated. Here's a look at their prototype. We can see the physical portion responsible for the drug delivery and the details of the circuit board and programming technique.

Here's a block diagram that provides a bit more detail. The IC certainly has some digital content, but let's not hold that against them. There is a microcontroller to support operations. There is 4kB of memory to hold drug delivery parameters and to record the actual details of the delivery. But as can be seen, there is a very large amount of analog circuitry of many flavors. This device is the poster child of integrated analog.

The device is powered from a coin cell. It uses a boost switcher to generate 3.3V. It has a clock oscillator that uses a 3-bit network of switched capacitors for frequency selection. It uses a 5-bit DAC to set the excitation current level. It has what looks to be a simple inductive loop to allow for non-galvanic connection for programming and data retrieval. It has a temperature sensor. It monitors electrode current to detect an open electrode and sound an alarm. There are active filters (low- and high-pass) and threshold detectors as part of the skin impedance measurement circuitry.

Imagine trying to build this with individual op-amps, switched capacitor filters, power supply devices, and a μP.

(Source: All images from 'An 87mA⋅min Iontophoresis Controller IC with Dual-Mode Impedance Sensor for Patch-Type Transdermal Drug Delivery System,' ISSCC)

(Source: All images from “An 87mA⋅min Iontophoresis Controller IC with Dual-Mode Impedance Sensor for Patch-Type Transdermal Drug Delivery System,” ISSCC)

Devices like this represent the future of medical diagnostic and treatment delivery systems. Have you had an experience with these devices yet — either designing or using?

Related posts:

19 comments on “The Future of Medicine Is Here

  1. Davidled
    June 15, 2013

    Patch concept looks similar to Zio@ Patch, iRhythm, monitoring cardiac rhythm. Do device monitor the battery power before battery runs out? LED might check the status of battery. First, Circuit may be calibrated. There is no way to tuning unless making other chips.  Second, Customer might check clinic test report before buying it. Third, I am wondering how patient feel when electronic feed into skin. I am very concern for algorithm related to amount of drug being delivered by using   temperature and skin condition.  

  2. Herby6262
    June 15, 2013

    I'm not sure what medical applications this is most being looked at, but the first thought that crossed my mind honestly was “what is the potential for abuse” – could someone be able to force or gain control of the amount of drug?  It sounded like it was regulates itself with threshold detectors and skin impedance.  But people get awfully creative when they have enough time.  Just curious if that's been discussed as even a concern, or are the applications of what it is used for take away that risk on their own?


    edit: spelling

  3. Davidled
    June 15, 2013

    In the NA (North America), I think that FDA should approve this product in order to release it in the market in terms of regulation. I am wondering how other country's government controls all product related to drug by law. Potentially, this product might be abused in terms of the aging-product and patient using this product.  The problem is that none in FDA understand all mechanism of this kind of product, unless they hire the analog engineer as to investigate it.  

  4. Netcrawl
    June 17, 2013

    @Daej you're right right, they're still in the process of learning how to regulate this product, I think we still need to train and educate our personnel about the product detail.

    June 17, 2013

    First thing that came to mind was the device that Dr McCoy used in one of the Star Trek movies (old cast).

    FDA will need some good training on how this could be used, monitored, and abuse prevention. I would also look for hacking prevention as well since that is a concern for some pace makers.

    Otherwise, this is a piece of art in terms of the analog integration – oh, and that digital stuff too.

  6. Brad Albing
    June 17, 2013

    @Herby – Cerytainly, if s/o chose to, they could abuse the device and self-administer more than the prescribed dose of a drug. Altho', the device would keep a record of that, so that would help mitigate against long-term misuse or abuse.

    At this point with this version of the device, that is not under consideration. Here (both my blog and the original ISSCC paper), the intent is simply to inform. The marketplace and the regulatory agencies will deal with the issues you've cited later.

  7. Brad Albing
    June 17, 2013

    @Derek – yes, moving ever closer to the medical deiagnostic and intervention methods as shown in Stra Trek. Sci-fi moves closer to sci-fact.

    Hacking is an issue, as always with any technology – even the simplest:

    So methods must be designed to safeguard the equipment.

  8. jkvasan
    June 18, 2013


    Nice post with interesting information.

    As technology evolves, the day is not far when some of the devices behave like a typical SWAT team – go to the infection site, kill the virus and get out without a trace ( Ref :Transient Electronics : Medical Devices Packaged to Die) and under-skin chips monitor and analyse blood to provide online and accurate results (Ref : Sub-Dermal Lab-On-Chip Takes Blood Analysis to a New Level).

    When such path-breaking technology is introduced, people express their anxiety and apprehensions. I believe regulators , Medical Device Manufacturers and other such stakeholders are on a constant quest to make these devices reliable and useful.


  9. amrutah
    June 19, 2013


       I most important application that comes to my mind is administering “insulin” for hyeperglycemic patients.  They can get rid of the daily dose of Insulin injection and get it using Iontophorosis.

      Actually there are major setups for invasive drug administration, but getting a whole system on a chip is commendable thing.

      Iontophorosis and reverse iontophorosis could help administer drugs and also extract waste (avoiding the painly Dialysis) invasively.

  10. amrutah
    June 19, 2013


         I agree, this is a very interesting post as well as the topic.  Also the reference you mentioned are useful.

        There is a lot of work going on where scientists are mimicking organs of chip. Brain-on-chip and lung-on-a-chip which are used for testing of drugs.  Many things are happening and soon there will catalogue of Transient Electronic products in market.

  11. Brad Albing
    June 19, 2013

    It would be interesting to see if the reverse method could remove enough waste products to make it useful.

  12. Brad Albing
    June 19, 2013

    >>First thing that came to mind was the device that Dr McCoy used in one of the Star Trek movies . “I'm a doctor, not an analog design engineer.”

  13. Gregst
    June 19, 2013

         This is an ancient idea of delivery of medications by electrophoresis. It originated in the beginning of 20-th century, was very popular in Germany in 1920-s – 1930-s. It migh be good for some medications, not good for others. Ions would not simply penetrate the blood vessels and create a sufficient concentration in the bloodstream. Nay, the process is much more complicated. The other idea, currently widely used, is topical application of a medication, where the medication is propagated by diffusion. Why use electrophoresis if any medication, even non-polar, hydrophobic, can simply propagate by diffusion ?

        Large ions do not penetrate the biological cells easily. A cell is surrounded with the membrane, phospholipid bilayer, that is inside very ionophobic. Generally, to get ions through the membrane, we need to surround them with hydrophobic capsules, micelles, or somehow get them through the ionophores and active transporters in the membrane. Some ions can act outside the cells by binding to the receptors and being signalling molecules. On those occasions, electrophoresis might work.I'm saying might, because even if an ion gets into a blood vessel, what prevents it from getting outside of it and travelling to the electrode of the opposite polarity ? Does it mean that electrophoresis should be pulsed, in sync with the blood flow speed ?

      The idea of electrophoresis for a broader spectrum of medications is very naive. Most of those German inventions in physiotherapy gradually disappeared from the Western medicine. Some of them were very barbaric by modern standards, like treating the throat infections with UV light and heating the deep tissues with microwaves: imagine what genetic mutations both methods caused. But, of course, they were developed long before people knew about  DNA and mutations. Electrophoresis belongs to the same generation. I'm amazed that it is now revived as some form of panacea.

  14. Brad Albing
    June 19, 2013

    @Gregst – Thanks – it's good to get info from someone who knows more about the biochemical and physiological aspects of this as opposed to just the electronics.

  15. Gregst
    June 19, 2013

    You don't want to remove ions from your body unjudiciously, even the excessive ones. Removal of ions is a very precise function of your kidneys, and electrophoresis would not help. The other waste in your body is detoxed by the liver. If you lose your liver and kidneys, you are as good as dead, even with dialysis. No electrophoresis would help.

    Actually, you might do some experiments. Do such electrophoresis that would remove from your body those pesky cations of sodium and potassium. I'd love to see you after that, not because I'm sadistic, but for scientific purposes only.

    See, I have a unique point of view: I'm simultaneously an Electrical Engineer and Biochemist/Molecular Biologist. Sometimes, the purely engineering approach to biology looks very funny.

  16. Brad Albing
    June 19, 2013

    Point taken. Now give me back my ions. In return, I'll resist the urge to post another corny engineering joke.

  17. eafpres
    June 19, 2013

    @Gregst–I think you missed the point. Diffusion through liquids is very slow–D < 10^-5 cm^2/s. iontophoresis uses the applied field to speed transport thus allowing controlled time vs dosage. Yes, some meds work well in simple transdermal application--pain meds, nicotine replacement, etc. But in cases where you need non-constant delivery this technology will be very useful. Also my understanding is people can tolerate and even benefit from current flow. So I would withhold judgement until we see applications. The enablement by the analog integration will lead to more ideas.

  18. Gregst
    June 19, 2013

    @eafpres: Indeed, I'm aware of the fact that the transdermal diffusion if very slow. Yet take a look at the chemical formula of nicotine that is frequently delivered by topical means: two well stabilized rings – hexagonal and pentagonal, the single polar group theoretically able to form hydrogen bond is N-H, and even it is not very active because it is stabilized by resonance within the ring. Nicotine pKAs are 6 and 11, pH of blood is 7.4, so the pentagonal N would become cationic, but it would be stabilized by resonance on the ring. Nicotine could be ionic only in salts, upon the salt dissociation it is a resonant structure, well stabilized. So, the molecula is essentially non-polar, will be subjected to the hydrophobic force, and will go through the phospholipid bilayer easily. Yet since dissociated nicotine is not ionic or polar under any physiological pH, it is not a subject to electrophoresis. The bottom line is that only small ions, such as moleculae of water or sodium would be able to get into the system circulation as a result of electrophoresis. Even potassium would have hard time penetrating it (potassium gets in through an active transporter). Hence, here is the question: what use does electrophoresis have for the larger ionic moleculae that need to get into the systemic circulation ? Apparently, none.

    Another problem is that constant electrophoresis will remove the very same cations that by some chance got into the blood. In the very same way they got into the blood, they would be extracted from it by the negative electrode.

    Yet another problem with both electrophoresis and diffusion is: we want the drugs in the systemic circulation, not in the local tissues. Imagine a significant release of certain bioactive moleculae into the local tissues: you might get necrosis there, cells would die, burst, release enzymes harmful to other cells…you know the rest. And you would get a huge concentration of these ionic bioactive substances in the local tissues prior to any of them would reach the blood vessels.

    There is a good reason for why electrophoresis is not used in any known to me systemic drug delivery in the Western World.

  19. jkvasan
    June 20, 2013


    Transient electronics is soon to gain market acceptance in the coming decade. Tasks/treatment which would be otherwise unimaginable could be possibly done with such implantable/decomposable electronics. The intricate details of their impact on the immune system needs to be studied for seamless implementation.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.