Using a unique data format is an often overlooked area of security. Simply packing bytes and using bit fields not only makes the data packet smaller, but does prevent all bit the technically savvy from monitoring the data.
I certainly see your point, Alan. Current trends are taking technology usage out of the hands of specialists and into those of less technical users who may inadvertently create a security breach that allows sensitive information to either be accessed or corrupted. Networking is definitely on the increase in medical applications for easier sharing of data - I can see how this increases the need for security...
Thanks all for your comments on the article. Apologies for the delay in responding. I lost track of when the article was going live. I think there are three main contributors to the cost of medical devices. One is definitely the higher development costs to meet regulatory and safety requirements. However the "typical" medical device takes much less overall cost to develop than the iPhone for example. Liability is another big piece, and I think the third important piece is the relatively low volumes. At Logic PD I have worked with a variety of very recognizable device manufacturers, and volumes for some of the devices that we've worked on with large-scale development efforts can be in the 100's per year. If you sell only about 1,000 in ten years every million you spend on development adds $1000 to the cost of each unit. Such low volumes also increases the cost of the components that go into the device (buying a million memory chips is a much lower unit cost than buying 100).
Nancy, first off apologies for the delay in responding. I lost track of when the article was going live. Absolutely most of what I described is generally applicable to consumer devices as well as medical devices. Device and information security is generally a fairly mature and active area of development, and I was trying to illustrate some of the areas where medical device designers and manufacturers should be paying more attention to security, in an environment that historically has had fewer security concerns (non-networked devices, used in controlled environments, by trained health care professionals).
Greg, first off apologies for the delay in responding. I lost track of when the article was going live. I agree completely. The focus of the article was intended to be FDA regulated devices, not so much data protection that would be governed by HIPAA. In fact I was motivated to write because of the reports in 2011 that an insulin pump had been successfully hacked, and was able to be programmed maliciously over a wireless connection.
Charles, first off apologies for the delay in responding. I lost track of when the article was going live. The level of level of security depends on the safetly classification of the device. In cases of lowest patient risk something like SELinux or SEAndroid (Security Ehanced) may be appropriate. In cases of higher risk most closed source OS options that offer packages specifically for medical device development will be closed-source, and provide an appropriate level of security as a starting point. In terms of networked devices one aspect of security outside scope of my post is IT policy. The range and nature of devices that connect to your network, and whether or not persistent storage is all encrypted, and whether it's possible to install new apps, etc all contribute to overall security.
From a patient safety standpoint, I'm not as concerned with the pirating of medical information as I am about a hacker who infiltrates the medical device with malicious intent. I think we should consider ways to mitigate hacker risk if a medical device is connected to a network and could be vulnerable to an attack on its operating system (where applicable).
The high cost of medical devices is due in part to a longer history of liability problems than of leaked data, a much more recent concern. Other factors like very high performance and the high cost of middlemen no doubt contribute yet more cost. But I think Cabe's point about leaked data is a good one--that's probably going to be a contributing factor to higher device costs in the near future.
In an age of globalization and rapid changes through scientific progress, two of our societies' (and economies') main concerns are to satisfy the needs and wishes of the individual and to save precious resources. Cloud computing caters to both of these.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.