The Stanford University Medical EXperi-
mental Computer (SUMEX) was established in
January, 1974, to provide the first shared na-
tional computing facility for medical research.
Directed by Dr. Joshua Lederberg, Professor and
Chairman of the Department of Genetics,
SUMEX is an innovative effort to help biomedi-
cal scientists meet today’s research requirements
and to explore computer applications in many
health fields ranging from basic research to
bedside care. The project is funded by a grant
from the Division of Research Resources of the
National Institutes of Health (Biotechnology
Resources Branch) for an initial term that ex-
pires in July, 1978.

At present, SUMEX consists of a powerful
PDP-10 computer available to approved users
throughout the United States over a computer
communications network on a time-shared basis.
The project’s goals over the next 5 years are: 1)
the encouragement of applications of artificial
intelligence in medicine (AIM), and 2) the man-
agerial, administrative and technical demonstra-
tion of a nationally-shared technological re-
source for health research.

Such a resource offers scientists both a signifi-
cant economic advantage in sharing expensive
equipment and a greater opportunity to share
ideas about their research. This is especially true
in computer science, a field whose intellectual
and technological complexity has made it diffi-
cult to avert the development of relatively iso-
lated research groups. Each group may then tend
to pursue its line of investigation with limited
convergence on working programs available from
others. The SUMEX-AIM project seeks to lower
these barriers to scientific cooperation in the
field of artificial intelligence applied to health
research.
ARTIFICIAL INTELLIGENCE

The term “artificial intelligence” (AI) refers
to research efforts aimed at studying and mecha-
nizing information processing tasks that general-
ly have been considered to require human intel-
ligence. The current emphasis in the field is to
understand the underlying principles in efficient
acquisition and utilization of material know-
ledge and representation of conceptual abstrac-
tions in reasoning, deductive and problem-
solving activities. AI systems are characterized
by complex computational processes that are
primarily non-numeric, e.g., graph-searching and
symbolic pattern analysis. They involve proce-
dures whose execution is controlled by different
types and forms of knowledge about a given task
domain, such as models, fragments of “advice”
and systems of constraints or heuristic rules.
Unlike conventional algorithms commonly based
on a well-tailored method for a given task, AI
procedures typically use a multiplicity of meth-
ods ina highly conditional manner—depending on
the specific data in the task and a variety of
sources of relevant information. The tangible
objective of this approach is the production of
computer programs which, using formal and
informal knowledge together with mechanized
hypothesis formation and problem-solving pro-
cedures, will be more general and effective con-
sultative tools for the clinician and medical
scientist.

Each authorized project in the SUMEX-AIM
community is concerned in some way with the
application of these principles to medical prob-
lems. This type of ‘‘intelligent” assistance by
computer program is perhaps best illustrated by
the following brief descriptions of some
SUMEX-AIM projects.
DENDRAL

The DENDRAL project at Stanford, under
the direction of Dr. Lederberg, Professor
Edward Feigenbaum, Computer Science, and
Professor Carl Djerassi, Chemistry, is aimed at
assisting the biochemist in interpreting molecu-
lar structures from mass spectral and other
chemical information. In cases where the charac-
teristic spectrum of a compound is not cata-
logued in a library, the DENDRAL programs
carry out the rather laborious processes a
chemist must go through to interpret the spec-
trum from “first principles.”” By symbolically
generating ‘‘reasonable’’ candidate structures
from hints within the spectrum and a knowledge
of organic chemistry and mass spectrometry, the
program infers the unknown structure to be the
one which best explains the observed spectrum.
There is no direct algorithmic path available to
determine such a molecular structure from the
spectral data—only the inferential process of
hypothesis generation and testing within the
domain of reasonable solutions defined by a
knowledge of organic and physical chemistry.

This process, as implemented in the com-
puter, is a simplified example of the cycle of
inductive hypothesis—deductive verification that
is often taught as a model of the scientific
method (Whether this is a faithful description
of contemporary science is arguable, and how it
may be implemented in the human brain is
unknown. Regardless, these are useful leads
rather than absolute preconditions for the prag-
matic improvement of mechanized intelligence
for more efficient problem-solving.). The elabo-
ration of these approaches with existing hard-
ware and software technologies is the most
promising approach to enhancing computer
application to the vaguely structured problems
that dominate our task domains.
THE RUTGERS PROJECT
COMPUTERS IN BIOMEDICINE

Professor Saul Amarel, a Rutgers University
computer scientist, directs several research
efforts designed to introduce advanced methods
in computer science—particularly in artificial in-
telligence and interactive data base systems—into
specific areas of biomedical research.

For example, a group of computer scientists
led by Professor Casimir Kulikowski is develop-
ing computer-based consultation systems for di-
seases of the eye in collaboration with Dr. Aran
Safir, an ophthalmologist from the Mount Sinai
School of Medicine. An important development
in this area is the establishment of a national
network of collaborators for computer diagnosis
and treatment of glaucoma. The computer
system, which includes an elaborate pathophysi-
ologic model of the disease, is being tested
through the SUMEX-AIM network at three eye
centers: Mount Sinai Hospital and Medical Cen-
ter, New York; Washington University, St.
Louis; and The Johns Hopkins University, Balti-
more. Glaucoma, in one form or another, affects
2% of all people over 40 years of age. It is a
disease in which increased pressure within the
eye may lead to irreparable optic nerve damage
and blindness. The computer-based program has
great potential for assisting clinicians and re-
searchers in understanding the disease, diag-
nosing it more accurately and improving its
treatment.

In another project, Professor Charles Schmidt,
a social psychologist, is developing a theory of
how people arrive at interpretation of the social
actions of others. The theory will be tested in
situations such as the psychiatric interview and
the legal trial. The computer system which
currently represents the theory is called ‘“‘Believ-
er.” It includes a large body of statements about
people’s motivations and actions. The SUMEX-
AIM environment provides an excellent medium
for collaboration between Dr. Schmidt’s group
and other researchers around the country in the
development and testing of this computer-based
theory.

The Rutgers project includes, in addition,
several fundamental studies in artificial intelli-
gence and system design. These provide much of
the support needed for the development of
complex systems such as the glaucoma consulta-
tion and the “Believer” programs.

MYCIN
Computer-based Consultation
in Clinical Therapeutics

Dr. Stanley Cohen, Associate Professor and
Head of the Division of Clinical Pharmacology at
Stanford, directs this research in collaboration
with Dr. Stanton Axline and with computer
scientists interested in artificial intelligence and
medical computing. An evolving computer pro-
gram developed to assist physician nonspecialists
in the selection of therapy for patients with
bacterial infections, MYCIN attempts to model
the decision processes of medical experts. It
consists of three closely integrated components:
the Consultation System asks questions, makes
conclusions and gives advice; the Explanation
System answers questions from the user to
justify the program’s advice and explain its
methods; and the Rule-Acquisition System per-
mits the user to teach the system new decision
rules or to alter pre-existing rules judged to be
inadequate or incorrect. Goals for further devel-
opment of the system include expansion of the
consultation program to deal with infections
other than bacteremias and implementation and
evaluation of the system in the clinical setting at
Stanford University Hospital.

COMPUTING APPLIED TO
PROTEIN CRYSTALLOGRAPHY

Members of the artificial intelligence project
at Stanford also are collaborating with Professor
Joseph Kraut and Dr. Stephan Freer, protein
crystallographers at the University of California,
San Diego. They are using the SUMEX-AIM
facility as the central repository for programs,
data and other information of common interest.
The general objective of the project is to apply
problem-solving techniques, which have emerged
from artificial intelligence research, to the well-
known “phase problem” of x-ray crystallo-
graphy in order to determine the three-dimen-
sional structures of proteins. The work is in-
tended to be of both practical and theoretical
value to computer science (particularly artificial
intelligence research) and protein crystallo-

graphy.

DIALOG

The DlAgnostic LOGic project, under the
direction of Dr. Harry Pople and Dr. Jack Myers
at the University of Pittsburgh, is a large-scale,
computerized medical diagnostic system uti-
lizing the methods and structures of artificial
intelligence. Unlike most computer diagnostic
programs, which are oriented to differential
diagnosis in a rather limited area, the DIALOG
system deals with the general problem of diag-
nosis in internal medicine and currently accesses
a medical data base encompassing approximately
50% of the major diseases in internal medicine.

MISL

The Medical Information Systems Laboratory
at the University of Illinois at Chicago Circle has
been established under the direction of Dr.
Bruce McCormick, Information Engineering, in
collaboration with Dr. Morton Goldberg, an
ophthalmologist at the U of | medical school.
The project explores inferential relationships
between analytic data and the natural history of
selected eye diseases both in treated and un-
treated forms. SUMEX-AIM will be utilized to
build a data base to be used as a test bed for the
development of clinical decision support algo-
rithms.
SUMEX-AIM Management

A significant part of the SUMEX-AIM experi-
ment is the development of a management struc-
ture to maximize the utility of the computer
capability for a national community.

Users of the SUMEX facility are divided for
administrative purposes into two groups: 1)
those at Stanford University School of Medicine,
and 2) those elsewhere in the United States. The
facility resources (computing capacity and con-
sulting support) are allocated in equal portions
to the two groups. As Principal Investigator for
the SUMEX grant, Dr. Lederberg reviews Stan-
ford medical school projects with the assistance
of a local advisory committee. National users
may gain access to the facility resources through
an advisory panel for a national program in
artificial intelligence in medicine (AIM). The
AIM Advisory Group consists of members-at-
large of the AI and medical communities, facili-
ty users and the Principal Investigator of
SUMEX as an ex-officio member. A representa-
tive of the National Institutes of Health-Biotech-
nology Resources Branch (NIH-BRB) serves as
Executive Secretary.

The SUMEX-AIM computing resource is ini-
tially allocated to qualified users without fee.
This, of course, entails a careful review of the
merits and priorities of proposed applications.
At the direction of the Advisory Group, ex-
penses related to communcations and transpor-
tation to allow specific users to visit the facility
may be covered as well.

SUMEX-AIM is aware of the necessity of
making the facility available for trial use by
potential users and collaborators. A GUEST
mechanism has been established for those who
have an indicated requirement for brief access to
certain programs. Those who have been given an
appropriate telephone number and login proce-
dure can dial up SUMEX-AIM to exercise these
programs on a trial! basis.
USER QUALIFICATIONS

Applications for use of the SUMEX-AIM facil-

ity are judged on the basis of:

1) The scientific interest and merit of the
proposed research.

2) The relevance of the research to the artifici-
al intelligence approach of SUMEX-AIM as
opposed to other computing alternatives.

3) The user’s prospective contributions and
role in the community, e.g., developing and
sharing new systems or application pro-
grams, sharing use of special hardware, etc.

4) The user’s capability and intentions of ope-
rating in a community-effective style for
mutual advantage. Besides the program-
ming innovations that some participants
may contribute, all are expected to furnish
expert knowledge and advice about the
existing art in their fields of interest.

5) The quantitative allocation of specific ele-
ments of the SUMEX-AIM resource based
on a concept of mean and ceiling planned
expectations.

FACILITY INFORMATION

The computer facility, consisting of a DEC
Model KI-10 CPU running under the TENEX
operating system, has 256K words (36-bit) of
high-speed memory, 1.6M words of swapping
storage, 70M words of disk storage, two 9-track
800 bpi industry-compatible tape units, a dual
DEC-tape unit, a line printer, and communica-
tions-network interfaces providing user terminal
access, SUMEX is available through TYMNET
and as a host over the ARPANET communica-
tions network.

Program (software) support will evolve from
the basic system as dictated by the research
goals and needs of the user. Initially, available
programs include a variety of TENEX user,
utility and text editor programs. Major user
languages include INTERLISP, SNOBOL, SAIL,
FORTRAN-10, BLISS-10, BASIC, Macro-10,
OMNIGRAPH and MLAB.

Revised May 1975
 

 

POTENTIAL USERS
For further information, write:

Elliott Levinthal, Ph.D.

AIM User Liaison

SUMEX-AIM Computer Project
c/o Department of Genetics, S047
Stanford University Medical Center
Stanford, California 94305

Procedures for access to SUMEX-AIM are
governed by the:

Biotechnology Resources Branch
Division of Research Resources
National Institutes of Health
Building 31, Room 5B19

9000 Rockville Pike

Bethesda, Maryland 20014