« January 2006 | Main | March 2006 »

February 28, 2006

Data Input: voice recognition

Techguy (John) recently commented he believes that Voice Recognition is the way of the future. He may be right—at least in part. I first saw VR many years ago and was unimpressed. Fortunately, I read positive reviews of Dragon Naturally Speaking 8 and decided to invest—now I’m impressed.

As a first test, I read from a book on computerising work—which uses a good number of unusual words—and was astounded to see it was more than 80 percent accurate straight out of the box. Mind, you have to "teach" the software to become more accurate, which is a bit of a chore taking several hours of dictation.

Nonetheless, I find it an excellent way of dictating first drafts into MS Word at a speed much faster than my typing. It also reduces the risk of repetitive strain injury, but the headphones give me a sore ear if I wear them for too long!

This feature on E-Health-Insider gives a good overview of the voice options available to healthcare, together with the pros and cons. The outsourcing of transcription from digital dictation to countries like India, it seems, can save money. But, this presents a public sector organisation like the NHS (the UK's largest employer) with a dilemma because using such a service would effectively outsource jobs overseas.

Nice entry here on KrellMD about data input which uses the theme of future crime prevention from the film Minority Report as an apposite analogy for future healthcare.

Lastly, after my trashing the computer keyboard read this entry on Slashdot about a new keyboard with 53 keys and an alphabetical key layout. Hmmm.

February 26, 2006

Close Communities or Remote Monitoring?

I have just come back from visiting my Mother, who lives in the north of England. On the outbound train I found myself sitting next to a recently-qualified GP, and we talked about primary care. I explained to Helen (let’s call her that) I thought using IT for remote monitoring would be the key to healthcare in Britain’s aging population. She thought that closer communities and families were needed.

Helen is Greek and is accustomed to extended families. She is struck by the lack of them and close communities in the UK. Issues daughters and sons might consult their fathers, mothers and grandparents about are now often taken to their GP. In addition, the aged are left to live alone, some unvisited for weeks. Consequently, she supported community nursing and nurse practitioners and perhaps the recreation of local community hospitals.

All of this took the wind out of my sails. We will be able to manage remotely long-term conditions with a combination of ICT and better sensors, and telecare may be a practical way to make efficient use of increasingly scarce qualified staff. But what of the human touch? Would I like my Mother to be alone with only technology to keep her company?

February 22, 2006

Computer Input by Thought

After my tirade against the computer keyboard, I had it in mind to write about other computer input devices. However, in the case of the tablet PC, Dr. Bill Crounse has already done a good job on the Healthblog here. Nonetheless, there is still some scope left.

So let's begin at the top: what about input by thought? In Scientific American Mind: Train Your Brain mental exercises with neurofeedback may ease the symptoms of attention-deficit disorder and epilepsy. A child manipulates a computer-generated jet by increasing the potential of certain brain waves.

Another SciAm article Thinking Out Loud describes how patients unable to speak or gesture can communicate by controlling a cursor through a brain-computer interface by manipulating their brain waves.

From rattling the plastics to racking our brains?

February 19, 2006

Patient Choice: nightingale or nightmare?

Peter writes for FHIT as a guest author

I was working with a group that provide call centre services for the NHS Choose and Book system a few weeks back. Casually I enquired: “Do patients exercise choice when asking for health services?”

The person I asked appeared to swell a little and then launched into a bit of a rant:

“Well, strangely enough, a lot of them do not. Most choose the nearest local facility. It is almost universally a myth that people in Kent would choose to go to Aberdeen for an operation. In quite a few cases, they actually find the notion of choice to be an irritation. In fact, just the other day, we had to report the case of a patient who had complained bitterly to the Department of Health about the need to make a choice. As far as she was concerned, she was so concerned about her health that it was just one more hurdle too far to decide whether she wanted to have her care delivered in Milton Keynes, Manchester or Milan!”

So, what's the problem with this I wondered? Surely the skilled call-centre person could head off the complaint by explaining that they could simply opt for the first available local service or let the call centre operator decide for them? How wrong could I be?

On further discussion, it appeared that the call centre staff have no choice but to offer choices. They are simply not allowed to apply common sense to a given situation, and are hidebound by a complex rulebook that prevents them from saying certain words and making certain suggestions. What can we conclude from this?

  • There's no doubt that choice is a good thing when it is important to make a choice.

  • A system based on choice must provide the operators of that system with the flexibility to offer the choice that the patient desires. If the patient does not want to make a choice, then this is a choice.

Peter Cook is the author of “Sex, Leadership and Rock 'n' Roll - Leadership Lessons from the Academy of Rock”it explores Leadership through the metaphor of music. Top Business Guru Tom Peters recently said of it:

“None would doubt that we live in a Rock 'n' Roll Age -- so what makes more sense than a brilliant, original, rockin' Rock 'n' Roll model of business management and leadership? Sex, Leadership and Rock 'n' Roll is a marvellous book, which closes the door on the tidy, hierarchical, know-your-place 'Orchestral Age' and ushers in a new, creative era of challenge and change. Hooray!”

February 16, 2006

Dirty QWERTY

keyboard.jpgWhy QWERTY? Why indeed.

This keyboard layout was designed in 1874 by Christopher Scholes. An alphabetical layout caused the machine's levers to jam, so he positioned the most frequently used letters as far apart as possible. To assist salesmen to demonstrate, Scholes craftily arranged all of the letters needed for the word "typewriter" to be in the top row.

I think this is an awful input device. I have been using it for many years and learned to touch type when I was completing a Masters dissertation. Time and again I watch users hunting and pecking their way across the keyboard. Imagine the time wasted in this painful process.

This article on Slashdot refers to a Swedish study that shows that the keyboard is a bacteria farm. It is inhabited by 33000 bacteria per square centimetre, compared to 130 on a toilet seat.

Covering a keyboard makes it easier to clean, important in these times of super bugs like MRSA. But what about ridding ourselves of it altogether? It's time for FHIT to look at other methods of data input.

February 15, 2006

FHIT Makeover

Regular visitors will notice that FHIT has had a makeover. We hope you like it. If you notice any problems or glitches please let us know. We hope you enjoy the site and will continue to contribute with your comments.

February 11, 2006

Healthcare IT: past, present and future

rbbh.JPGYesterday evening I went to a leaving do for my friend and former colleague Roger. He leaves the Royal Berkshire Hospital in Reading after 12 years in the IT Department to join NHS Connecting for Health.

On the way to a restaurant, Roger and I were reminiscing. He reminded me about the implementation of email at the hospital—at first to about 12 staff. It was received with scepticism, though the Trust secretaries found it a great way to mail papers to all of the executive at once. The implementation of an intranet met with a similarly lukewarm response.

I was to procure a hospital wide Electronic Patient Record as part of a major programme of change and rebuilding. While that procurement progressed, demand for email, intra-and internet applications grew until the two-man systems development team could barely cope. They implemented an A&E system, results reporting to GPs, protocol-based referrals and various other web-based applications.

That experience formed my view of healthcare IT. The near future of the NHS is one of considerable change, owing to an ageing population, a lack of qualified staff and increasingly better informed consumers. Major restructuring of the NHS is inevitable, moving funding away from acute to primary and community care and erasing organisational and disciplinary boundaries in the process.

To support this change, core systems like EPR, PACS, Pathology and Pharmacy will be the foundation for flexible, innovative applications, like those developed by systems development at the RBH. The future of healthcare IT in the NHS is one of centralized information and devolved innovation.

My article “Making NP fit” describes how the National Programme for IT fits with such a future. Good luck Roger!

February 08, 2006

Podcare: Care and iPod

Following from FHIT posting on a US doctor using 21st century iPods to improve students skills in the use of 19th century stethoscopes. Please read this article from Wired that shows how podcasting can be used to improve patient care.

February 06, 2006

Bayesian Machine Learning: dealing with complexity

bayes-small.jpgWhat do a spam filter and a Nonconformist minister who lived 300 years ago have in common?

About 300 hundred years ago Thomas Bayes invented Bayes Theorem. Though it was given some recognition in his lifetime, the field of Artificial Intelligence adopted it centuries later to create Bayesian Networks (BN) which became the basis of Bayesian Machine Learning.

BNs help us to manage complexity and uncertainty by inferring cause and effect, rather than having to be told it.
They modify an initial theory by applying evidence to a fixed set of rules based on Bayes Theorem. Using BN's machines learn about the real world, where the relationships between events are often uncertain.

Interrelationships in BNs can achieve mind-boggling complexity, but increases in processing power have made them practicable. They are used in data mining, operating systems, fault diagnosis and fraud detection. And, as the number of intelligent devices grows, they will become more common.

My spam filter Popfile uses Bayesian techniques. Since installing it, Popfile has become more than 90 percent accurate and helps to keep my Inbox clear. Bayesian filters identify spam from keywords such as "free", "new" and "enlargement". They also learn from my actions by associating the way I deal with an email with its characteristics.

BNs also mimic the way in which we think. Experience that is consistent with our beliefs tend to reinforce them; whereas inconsistent experience may force us to modify them or form new ones. Unlike Neural Networks (more in future FHIT entries) the workings of BNs are visible and so easy to consider and discuss.

It's easy to see how such techniques could be applied to decision and diagnostic support in healthcare. Also, clinical trials can be modified while they are in progress using Bayesian techniques; for example, dosages can be changed in the light of findings to allow more data to be collected closer to the optimal dose. This is inimical to traditional statistics whose supporters--sometimes called frequentists--say that it risks contaminating data and introducing bias.

Nonetheless, Bayesian techniques are likely to become common, allowing machines to anticipate our actions and help us to make decisions and deal with complexity--which seems perfect for the healthcare world.

If you are interested in reading more about machine learning, please see my article on the E-Health-Insider website: Cyber Care.

February 03, 2006

Forget Clinical Involvement

pacs.jpgAt a conference last week in London, UK “Successful Implementation of NPfIT 2006” engaging clinicians in the National Programme for IT was brought up time and again by speakers: let's engage them, let's involve them, let's get them on board—phrases that to me are meaningless shibboleths. I do not want to involve them at all.

I wrote an article last year for Hospital Doctor “Switching on to IT Benefits” in which I suggested—despite the assertion of some professional medical bodies—clinicians are not turned on by technology or the security of patient records or their transfer from existing systems. They are interested in what an Electronic Patient Record can do for them and their patients.

I give examples in the article of successful clinically-driven IT projects. One was the implementation of a PACS at the Royal Berkshire Hospital in Reading. A Radiologist who played a leading role in the project presented at the conference: not only had the PACS provided clinical benefit, and enabled new working processes—it had also paid for itself through savings. I can assure you that clinicians did not need to be pushed to use the PACS—indeed a recent survey reveals they find it of huge benefit.

Recently I also wrote about telemedicine projects which clinicians also enthusiastically endorsed.

All healthcare IT should aspire to the condition of PACS and Telemedicine, because I do not want clinicians to be involved, I want them to be inspired.

Image of PACS: GE Healthcare