Medpedia

Jan 14, 12 11:36AM | 0 comments
I found this visitor question on a discussion forum:" i was debating this with some doctors who say that it would be impossible to program a computer to make diagnoses as well as they can. i find this pretty ridiculous. whatever thought process/string of questions they would use to analyze the situation are the same that the computer would be programmed to use. the compute:r would then analyze all available information, ask questions, analyze the answers and assign probabilities. in fact, it seems like this would be way simpler than some of the things computers have already been programmed for. what do u think?"


So what do I think? 
My opinion, as a doctor, is that  what is input into a computer for calculation is the most important  part of the process of making a diagnosis and deciding on a treatment program to benefit the patient.   No amount of computer power or access to data storage will substitute for the physician's input of the history and the physical findings of the patient.  A computer posing questions to a patient and the patient responding will never substitute for a direct doctor-patient communication.  There are many subtleties, nuances  of a history which can never be accessed by a computer, such as body language and verbal expressions  and there is no way for a computer to perform a complete and worthy physical examination.  A robot used in surgery still requires a doctor behind it and no robot will attain the skills to inspect, auscultate, palpate and percuss and then interpret  the findings.  To me, how complete and understood is the input of data both from a patient telling a history and the doctor performing a physical is the basis for the diagnosis.   Poor input will always lead to poor output.  And, finally, it will always take a doctor to analyze the results of the computer to confirm its diagnosis.  I would agree that the doctor with knowledge and with experience and then working together with the computer can be most productive of  the correct diagnosis.

So.. what do you think?

..Maurice.

Comments

To add a comment to the original post, click here.

You must be signed in to post a comment.

Sign in now
  • (Comment from original source - Anonymous) on Jan 21, 12 02:30AM
    I don't think a computer could replace a doctor - getting an accurate history out of someone is an art form.
    I think it could be a great tool in addition though - I've seen a veterinary tool that given symptoms comes up with a list of differentials. Could help in coming up with rare conditions a doctor might not think of.
    TAM
  • (Comment from original source - borislavujo) on Jan 22, 12 09:03AM
    I think that algorithmized healthcare could and should complement current (overpriced) healthcare system and it is probably the future of western medicine, which strives for objectivity. The research is in the beginnings, for example, there have been some quite successful attempts to diagnose dermatological conditions using neural networks. I think that as we will better understand what we actually do when we try to propose a diagnosis, we will be able to put the relevant models into computers.
Editor Directory - browse by last initial
ABCDEFGHIJKLMNOPQRSTUVWXYZ
Professional Directory - browse by last initial
ABCDEFGHIJKLMNOPQRSTUVWXYZ
Cancel