Robert Clarke, the Managing Director and owner of RADLogic, has had a great amount of experience in integrated circuit design, working quite closely on the technical design of RADLogic’s offerings. As another fantastic speaker at the upcoming YOW! Connected conference this year, I asked him all about the IoT and using it for life-changing cancer treatment.
Robert Clarke and his team’s Ablation Controller
“Most people’s view of the IoT is a lot of smart ‘things’ talking to other smart ‘things’. That is true, but I also believe that the smart things will also need to communicate with objects that are not so smart.” – Robert Clarke
Robert is a talented guy who delves deeply into the technical side of things — developing hardware architectures, writing hardware description language (HDL) models, simulating in both digital and analog domains, and dealing with low-level design including the generation and verification of the underlying geometries that make up the layers etched into the silicon. It’s a job for those passionate about the area, and Robert definitely has that passion. They’ve even developed some of their own chip-design tools!
The IoT isn’t as new a concept as many think
When discussing RADLogic’s history, Robert points out that the concept of the Internet of Things has been around since before the term itself came about. They’ve been designing chips for early-IoT devices using Radio-Frequency Identification (RFID) for over 20 years. He sees IoT devices being split into two categories that need to be able to talk to each other — smart things and dumb things:
“RFID is a passive technology (no battery, usually powered by scavenged RF energy) that allows a smart object (such as a phone or piece of equipment) to communicate to a dumb object (like a sensor). In the IoT, not everything will be smart. I think there will be many more dumb things than there are smart things for the foreseeable future. As long as smart things can tell what the dumb things are, the smart things can determine what to do with the dumb things. To address this ‘dumb’ end of the spectrum, and after numerous enquiries from the US, we decided to do an investigation to determine how small and cheap we could make an RFID chip whose sole purpose in life is to provide its identity. The result is what we have facetiously called the IDIoT™ (ID for the Internet of Things), which can cost less than 0.1c to manufacture and is actually too small for currently available tag assembly methods (although this is likely to change soon).”
Treating cancer
It isn’t every day that a chip-design house gets involved with something as massive as treating cancer, so I was pretty eager to find out more about what was involved. How did it come about?
It turns out, it all began with a South Australian state government initiative called the SBIR challenge. The challenge came about as a result of a group of researchers discovering that they could “improve the efficiency of ablation [basically — surgical removal] of cancer cells by applying a DC current before and/or during the Radio Frequency treatment”. The challenge focused on requesting a feasibility study to look into how to control the ablation process in a more effective way as current techniques were “somewhat primitive”.
The rest of the story is so well told by Robert, that I’ll let him explain:
“After reading the request for proposals and seeing the constrained timeframe and budget for the project, I could see that technically, what was needed to be done (control and measure the ablation process) was not overly difficult, but that developing an alternate system from scratch to provide the requested features would cost too much and take too long for the feasibility study. Rather than just offering up a paper study (which would be boring), we decided to investigate alternatives so that we could deliver a working prototype by the end of the study.
We purchased a second-hand commercial Radio Frequency Ablation system (Boston Scientific RF3000) and interfaced it to a control port to allow software to take over. Controlling the DC power supplies that we had designed and measuring currents and voltages was the easy part (there are already chips available that could do this), but developing the command set, user interface, data logging, etc from scratch might take a lot of time and money. We also wanted to minimise cabling, because in a clinical setting things need to be kept as neat as possible.”
“if you get the right people together in the same room, things that seem difficult to one discipline often turn out to be straight-forward for another.” – Robert Clarke
RADLogic’s web app interface — The grey lines show the “specified” excitation and the coloured lines show the measured values.
As it turns out, Robert is a fan of something quite dear to my heart as well — web-based design tools and GUI development with JavaScript! These provide fast, easy to change interfaces and RADLogic had been using the concept internally prior to this. The main concern remaining — connecting the user and hardware without wires.
“As it happens I had seen advertised a feature-rich product manufactured by Texas Instruments called a ‘SensorTag’, which supports Low Energy Bluetooth (BLE) as well as just about every sensor known to man. It also included an interface port allowing it to be connected easily to extra hardware. I had already purchased one of these and had it sitting on my desk staring at me, waiting to be used for something other than an interesting curiosity.
Designing this into our circuit board provided the solution. Pre-existing JavaScript plugins for interfacing to the device’s Bluetooth stack completed the link, giving us direct control of the power supply and the ability to make the measurements of current, voltage and impedance that are needed to implement a control system.”
All of the hardware required to build the Ablation controller
With BLE sorted as their communication mechanism and a JavaScript-powered web interface as their front-end, they could then use the phone or tablet device running the interface to perform all of the complex computation — rather than the chips themselves.
“The processors in the phone/tablet provide all of the computation that is required, allowing complex ‘cooking profiles’ to be specified and dynamically controlled during the ablation process.” – Robert Clarke
“Data logging to allow post-processing of the ablation results is also implemented by the phone/tablet, and links to Dropbox or a local server allow all data and ablation sequences to be saved and recalled. We used Apache Cordova/Adobe PhoneGap to convert our web based code into a self-contained App that runs on any iOS or Android device.”
My favourite quote from the interview has to be this one:
“The butchers did look at me a bit strange when I kept turning up to ask for ‘very fresh’ livers to test the Ablation system.” – Robert Clarke
Points of note
Hardware costs were quite minimal — less than $100 (not including the RF3000) to provide equipment with complex capabilities.
Wide scope of potential use — Robert points out that “the technique is not limited to ablation systems. Many other medical applications, and indeed more importantly, commercial applications could be implemented in a similar fashion.”
Easy to update — New features can be added to the system and the web-based GUI can be altered quite easily.
It even works for simulated usage — It turns out that new users of the system can practice, prepare ablation sequences and even simulate the operation of the system without even having access to the equipment. All they need is access to their phone.
The SensorTag made all the difference — It made life easy for them because it was simple to use, well supported and included an extension port that conveniently connected to the hardware they had designed. Your choice of hardware can greatly impact your project, more than many realise!
Why this solution?
The Government initiative didn’t require that they use the Internet of Things for their solution. It didn’t even require them to make a mobile-accessible app. So why did they choose this route?
“It just seemed the neatest way to solve the problem in a time and cost-efficient manner. Why spend money on developing an equipment specific front panel and display, when almost everyone is carrying around a more capable, prettier and more flexible one in their pocket?”
In the end, there are many areas ripe with opportunities to tackle head on! Internet-connected smartphones provide a lot of opportunities just on their own — add in the ability to connect up to devices via Bluetooth, NFC and so on, and you’ve got a combination that can be used to build complex equipment, tools, and toys quickly and at low cost. There is clearly no limit to what’s possible — if Robert and his team could use it to try to cure cancer, the sky is truly the limit here.
Thank you to Robert for providing an inspirational example of using the Internet of Things and emerging tech in a way that has potentially life-saving results. If you’d like to find out more or get in touch with the RADLogic team, check out their website. Robert will be speaking at YOW! Connected 2016 in Melbourne on the 5th-6th October, so if you’ll be in the area — come find out more about what’s happening in emerging tech from people just like Robert! This will be my third event in a row and I’m really looking forward to it!
The post Treating cancer with the Internet of Things appeared first on Dev Diner.