OFFICE MEMORANDUM © STANFORD UNIVERSITY © OFFICE MEMORANDUM ®© STANFORD UNIVERSITY © OFFICE MEMORANDUM

To

FROM

SusJect:

S19 (244-_ KC) roB 251975

Date: February 25, 1975

Joshua Lederberg

Robert Greenes

Proposal by Frank Kuo for a Distributed Data Base System for Chronic Diseases

Dear Josh,

Thanks for giving me the opportunity to look at Kuo's Proposal. I think
this is an interesting idea, and it certainly fits with the objective of
utilizing SUMEX as a resource for medical applications of artificial in-
telligence. Although I would agree with the concept that multiple models
and levels of specificity need to be used to deal with real world medical
problems, it is not clear from this proposal how this use of multiple
models will actually come about. Who is going to do the work? How is
the expertise to integrate the various models with the data base to occur?
If an effort such as this were to gain much impetus, it might be reasonable
to consider a project staff supported at Stanford to actually make these
multiple decision-making support programs available as a library.

It appears that the use of the distributed network is somewhat contrived
in the sense that the number of computers involved in the proposed network
certainly does not seem to be required. Ina way, this has the appearance
of being technological overkill in order to dramatize a concept. On the
other hand, I know some of the work that has been going on at Gilbert's
shop at Straub Clinic, and although I don't know Nordyke, I would expect
that this group has a fairly high degree of competence.

One concern I have in the description of the project is the location of
the data base for gathering statistics on collected experience. For a
sophisticated consultation program to operate effectively and to improve
over time based on its own experience, I would much prefer to see a data
base management system integrated with SUMEX developed as part of the r2-
source capabilities located here, rather than being maintained at one of
the remote user sites. However, this is certainly not a criticism of tha
proposal but rather of the lack of activity in this area being done hers.
I think that the integration of a data base capability to help evaluate
methodologies such as this would be a valuable effort to undertake at
Stanford, particularly because of the data base capabilities that alread;
exist locally. Parenthetically, this kind of data base capability, when
coupled with some of the technological developments going on here, would
be a useful part of a technology assessment program, if we were to propose
that kind of emphasis for a Health Services Research Center at Stanford.

Regards,

LG
fof

WOGONVYOWIW 3291ddO © ALISHBAINN GYOINVILS «

ALISUJAINN GYOINVIS

ALISYZAINN GHOINVIS @ WNONVYOWAW 33tddO

WNONVYOWIW Fd1ddO