Doctor E tips his hat to a reader who sent a pointer to a new Web resource for all manner of medical subjects that include calculations that can be expressed as algorithms. The site is The Medical Algorithms Project . It requires registration but registration is free.

Log-in comes with the caution that:

The material at this website is intended only for the educational and personal use of health care students and professionals. It is not intended for persons who have not received appropriate medical training, and should not be used for making clinical decisions pertaining to patient diagnosis, care, or management. Algorithms predicting outcomes use data based on the original articles. Outcomes may vary between institutions and are impacted by newer developments in diagnostics and therapeutics. These should be validated prior to use

Any computations obtained from these algorithms should be compared with, and tempered by, personal clinical knowledge and judgment.

Most of us are aware of some very basic algorithms that are a part of daily medical practice. A fasting lipid panel, for example, will correct the total cholesterol number by a fixed proportion (20 percent) of the triglyceride result.

Creatinine clearance is another important number. Probably during residency you knew the formula off the top of your head, but today you’re more likely to send the timed urine sample off to the lab and wait to see the calculations.

Apropos of this blog, which focuses on topics in genetics, is the Algorithm Project’s section on medical genetics, which includes all manner of calculations.

Many of the algorithms for calculating the probability of paternity were new to me. They conjure up the possibility of real life sticky social situations and can surely provide the raw material for at least a couple episodes of “House.” Consider the avuncular index, which takes its name from the avunculus, the Latin for “uncle”. This algorithm calculates the probability that a child’s uncle is actually his biological father. Then there’s the RMNE algorithm—for Random Male Not Excluded—defined as “the probability that a random, unrelated male from the same racial background would not be excluded as the potential father of the child.

I find this stuff fascinating. Check it out.

Technorati Tags: medical genetics, paternity testing, medical algorithms, “House”

This entry was posted on Wednesday, November 15th, 2006 at 12:55 pm and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.