Main

June 03, 2007

Electronic Empathy: computers can care

Man using a computer.TV psychiatrist Professor Raj Persaud reports* the National Institute for Health and Clinical Excellence (NICE) has recommended making computer-based treatments for anxiety and depression more widely available. He argues this may be seen as another effort to reduce cost rather than meet patient needs. After all, patients want to be seen as individuals and prefer a person to a chip.

I’m not so sure.

In 1995 while researching for a MBA I came across some relevant research into the use of expert systems (considered part of artificial intelligence). Many patients who had consulted an expert system called ELIZA that did little more than ask reflective questions—for example: “Tell me more about…” or “What do you mean by..?”—responded positively. One woman left in tears saying she had never before met someone who understood her so well. Try some therapy from ELIZA.

In 2004 Whitfield and Williams asked: If the Evidence is so Good, Why Doesn’t Anyone Use Them? Surprisingly only about 5 percent of cognitive-based therapists were using computer-based self help as an alternative to face-to-face contact.

Today, if it comes to a choice between a highly personalised computer program, available day and night with no waiting list or 40 minutes with a busy human practitioner, I know which I prefer, Professor Persaud.

*Health Service Journal, 24 May 2007

September 25, 2006

Computers Better than the Docs?

Doctor examining an X-ray film.Regular readers will know, or have guessed, that I think the future of healthcare lies with computer diagnosis and intervention and the human touch. Machines will be able to calculate and infer more reliably, efficiently and accurately than us, but it will be a long time before they exceed us in empathy and in detecting non-verbal or numerical signs to our fellows's state of health.

In a study by Swiss researchers, neural networks exceeded humans in predicting epoetin responsiveness in treating anaemia in patients on dialysis. The study concludes: "In predicting the erythropoietin dose required for an individual patient and the monthly dose adjustments ANNs are superior to nephrologists' opinion. Thus, ANN may be a useful and promising tool that could be implemented in clinical wards to help nephrologists in prescribing erythropoietin."

Computers can also help in the detection of breast cancer. The Journal of Radiology reports that computers may also be more efficient and accurate in detecting breast cancer. One expert assisted by a computer could do the job just as well as two, speeding up screening for women who have waited for longer than three years. Read the BBC's report.


May 15, 2006

Medicine: humanly impossible?

My wife is studying Physiotherapy and spends her life scouring publication databases on-line. If she proposes a regime of treatment to a lecturer they ask: “How do you know? Where is your evidence?” Excellent, I say as an inveterate sceptic.

Continue reading "Medicine: humanly impossible?" »

March 28, 2006

Human Biology and Health

Nice entry here on the Nature Newsblog: 2020 Computing mostly about the integration of IT and human physiology.

Also see this entry on our US cousin with a link to articles on Nature.

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.

Continue reading "Bayesian Machine Learning: dealing with complexity" »