Section 4.2.6 MYCIN PROJECT Justification for Continued Use The rule-based framework for the MYCIN consultation program has demonstrated its applicability to medical domains. In addition, other SUMEX-AIM projects depend on our continued maintenance and development of the framework (see section on Collaboration). Because of Dr. Shortliffe's commitment, future developments of the program will be based on clinical problems which he encounters. We have not identified the precise areas at this time, however. Computing Requirements We perceive no additional requirements either from the system or from outside. Recommendations Community Development - Create a "visiting scientist" position for individuals who want to spend a fen months at Stanford learning about artificial intelligence in medicine. If it carries a stipend, or partial salary support for sabbatical leave, finding qualified individuals should not be difficult. Resource Development - Undertake, as a long range goal, solution of the problem of making large research programs available (in some form) to practicing physicians and biomedical scientists. This would require persons who understand the complexities of these programs and of the people using them, and could involve purchase and support of small computers as well. J. Lederberg €& E. Feigenbaum 176 PROTEIN STRUCTURE PROJECT Section 4.2.7 4.2.7 PROTEIN STRUCTURE PROJECT Protein Structure Modeling Project Prof. E. Feigenbaum and Or. R. Engelmore (Computer Science, Stanford) T. Summary of research program A. Technical goals The goals of the protein structure modeling project are to 1) identify critical tasks in protein structure elucidation which may benefit by the application of AI problem-solving techniques, and 2) design and implement programs to perform those tasks. We have identified two principal areas which have both practical and theoretical interest to both protein crystal lographers and computer scientists working in AI. The first is the problem of interpreting a three-dimensional electron density map. The second is the problem of determining a plausible structure in the absence of phase information normally inferred from experimental isomorphous replacement data. Current emphasis is on the implementation of a program for interpreting electron density (Ce.d.) maps. B. Medical relevance and collaboration The biomedical relevance of protein crystallography has been well stated in an excellent textbook on the subject (Blundell & Johnson, Protein Crystallography, Academic Press, 1976): "Protein Crystallography is the application of the techniques of X-ray diffraction ... to crystals of one of the most important classes of biological molecules, the proteins. ... It is known that the diverse biological functions of these complex molecules are determined by and are dependent upon their three-dimensional structure and upon the ability of these structures to respond to other molecules by changes in shape. At the present time X-ray analysis of protein crystals forms the only method by which detailed structural information Cin terms of the spatial coordinates of the atoms) may be obtained. The results of these analyses have provided firm structural evidence which, together with biochemical and chemical studies, immediately suggests proposals concerning the molecular basis of biological activity." The project is a collaboration of computer scientists at Stanford University and crystallographers at the University of California at San Diego (under the direction of Prof. Joseph Kraut) and at Oak Ridge National Laboratories (Dr. Carroll Johnson). Our principal collaborator at UCSD is Dr. Stephan Freer. 177 J. Lederberg & E. Feigenbaum Section 4.2.7 PROTEIN STRUCTURE PROJECT C. Progress summary During the past year we have continued with the design and implementation of a system of programs for interpreting three-dimensional e.d. maps. Progress has been made by attacking the problem from two directions: working upward from the primary data Ci.e., the array of e.d. values) to higher level symbolic abstractions, and working downward from the given amino acid sequence and ather experimental information to generate candidate structures which can then be confirmed by the abstracted data. Research which emphasized the “bottom-up” approach yielded results in several areas: | 1) A new skeletonization technique, based on Greer's work, was implemented, making it considerably easier to generate an optimum skeletal representation of the map. Experience in using this method on several different e.d. maps has led to heuristics for selecting grid size, minimum density threshold, contour levels, etc. Moreover, some additional post-processing of the skeleton has been implemented to indicate connectivity, to produce a list of "model atoms" for map refinement (see below) and to provide input for graphical display. 2) A new technique for improving the quality of medium resolution (about 2.5 Angstroms) e.d. maps was formulated and tested on simple structures and on a real protein currently under investigation at UCSD. The technique utilizes the quasi-~atomic structure produced by skeletonizing the map. The grid points in the map Which are not removed by the skeletonization process are treated as zeroth- order model atoms. A difference map is then computed, using the difference between observed and calculated structure amplitudes, and calculated phases, for the Fourier coefficients. The difference map is then used to move the model atoms tonard regions of higher density. The process is repeated until the shifts in atomic coordinates become insignificant. The standard crystallographic R value is used as one measure of quality of the map. Another measure is a subjective one of displaying the map and inspecting it in regions where the protein structure has been partially determined. This technique is currently being used as one step in the determination of the structure of cytochrome c peroxidase. The method is similar to one formulated by Agarwal in its use of a set of model atoms, whose positions are progressively refined. The use of a difference map is based on earlier methods of refinement developed by Freer et al. 3) We investigated the utility of graphics systems as a tool for acquiring model building heuristics. One system is a grey-scale display used in image understanding research at the Stanford Artificial Intelligence Laboratory. A second system is the Picture System used by Prof. Langridge's group at the UCSF jedical Center. We found the former system to be difficult to use as an interpretation aid, due to the necessity of visually synthesizing tuo-dimensional slices of the e.d. map. We attempted to incorporate depth into the display by making the grey-scale intensity a function of both electron density and depth, but the technique wasn't helpful. Other schemes were considered, but not implemented, due to the inordinate amount of programming effort that would be required. The Picture System at the UCSF Medical Center is similar in many respects to the system used by our UCSD collaborators, and its proximity to Stanford makes it attractive. We found the system to be a potentially valuable aid in comparing the skeletal representation of the e.d. map with the more J. Lederberg & E. Feigenbaum 178 PROTEIN STRUCTURE PROJECT Section 4.2.7 conventional contour map representation, as well as comparing the skeleton with known protein structures. At present, however, there is no working softnrare for displaying a contour map and a skeletal model simultaneously, but an effort is underway to put this capability into the system. The development of our "top-down" system (named CRYSALIS) for inferring the molecular structure from the amino acid sequence, symbolic abstractions of the e.d. map and stereochemical knowledge, is continuing. Recent additions to the system include rules for identifying main chains, side chains and bridge segments, using knowledge of expected topological properties of the skeleton and peak distributions. A new report, which focusses on the design of the CRYSALIS system, will be cut shortly. The purpose of that report is to summarize the current state of the system, and to critically review it with respect to its design specifications. Although the system does make inferences from the data about some structural features of the model, it has been difficult to extend the power of the system beyond its present level. The design review was primarily motivated by a desire to see which features of the system are worth preserving and which ones need redesigning in the next version of CRYSALIS. References: Agarwal, R. C., and Isaacs, N. W., "Method for obtaining a high resolution protein map starting from a low resolution map", Proc. Natl. Acad. Sci. USA, Vol. 74, pp 2835-2839 (1977). Freer, S. 1T.,, Alden, R. A.» Carter, C. W. and Kraut, J., "Crystallographic Structure Refinement of Chromatium High Potential Iron Protein at Two Angstroms Resolution", J. Biol. Chem., Vol. 250,46 €1974). Greer, J., "Three-dimensional Pattern Recognition: An Approach to Automated Interpretation of Electron Density Maps of Proteins", J. Mol. Biol., Vol. 82, pp. 279-301 (1974). D. List of Publications 1} Robert S. Engelmore and H. Penny Nii, "A Knowledge-Based System for the Interpretation of Protein X-Ray Crystallographic Data," Heuristic Programming Project Memo HPP-77-2, January, 1977. (Alternate identification: STAN-CS-77- 589) 2) E.A. Feigenbaum, R.S. Engelmore, C.K. Johnson, "A Correlation Between Crystallographic Computing and Artificial Intelligence," in Acta Crystallographica, A33:13, (1977). (Alternate identification: HPP-77-25) 179 J. Lederberg & E. Feigenbaum Section 4.2.7? PROTEIN STRUCTURE PROJECT E. Funding status Grant title: The Automation of Scientific Inference: Heuristic Computing Applied to Protein Crystallography Principal Investigator: Prof. Edward A. Feigenbaum Funding Agency: National Science Foundation Grant identification number: MCS 74-23461-A01 Term of award: May 1, 4977 through April 30, 1979 Amount of award: $150,200 Cincluding indirect costs) II. Interaction with the SUMEX-AIM resource A. Collaborations The protein structure modeling project has been a collaborative effort since its inception, involving co-workers at Stanford and UCSD Cand, more recently, at Oak Ridge). The SUMEX facility has provided a focus for the communication of knowledge, programs and data. Without the special facilities provided by SUMEX the research would be seriously impeded. Computer networking has been especially effective in facilitating the transfer of information. For example, the more traditional computational analyses of the UCSD crystallographic data are made at the CDC 7600 facility at Berkeley. As the processed data, specifically the e.d maps and their Fourier transforms, become available, they are transferred to SUMEX via the FTP facility of the ARPA net, with a minimum of fuss. (Unfortunately, other methods of data transfer are often necessary as well -- see below.) Programs developed at SUMEX, or transferred to SUMEX from other laboratories, are shared directly among the collaborators. Indeed, with some of the programs which have originated at UCSD and elsewhere, our off-campus collaborators frequently find it easier to use the SUMEX versions because of the interactive computing environment and ease of access. Advice,» progress reports, new ideas, general information, etc. are communicated via the message and/or bulletin board facilities. B. Interaction with other SUMEX-AINM projects Our interactions with other SUMEX-AIM projects have been mostly in the form of personal contacts. We have strong ties to the DENDRAL, Meta-DENDRAL and MOLGEN projects and keep abreast of research in those areas on a regular basis through informal discussions. The SUMEX-AIM workshops provide an excellent opportunity to survey all the projects in the community. Common research themes, e.g. knowledge-based systems, as well as alternate problem-solving methodologies were particularly valuable to share. J. Lederberg & E. Feigenbaum 180 PROTEIN STRUCTURE PROJECT Section 4.2.7 C. Critique of Resource services On the whole the services provided by SUMEX have been excellent, considering the large demand on its resources. With the important exceptions of high peaks in the weekday prime-time load average, the ratio of CPU time to total wait time during pregram execution is usually acceptable. The facility provides a wide spectrum of computing services which are genuinely useful to our project -- message handling, file management, Interlisp, Fortran and text editors come immediately to mind. Moreover, the staff, particularly the operators, are to be commended for their willingness to help solve special problems (e.g., reading tapes) or providing extra service (e.g., and immediate retrieval of an archived file). Such cooperative behavior is rare in computer centers. The most serious deficiency, from our point of vien, is the lack of a file transfer facility between SUMEX and the computing system in the UCSD Chemistry Department. Our day-to-day collaboration with Dr. Freer at UCSD would be greatly ennanced by a reasonably fast Ceven 1200 baud would suffice) channel for transmitting proposed protein models, generated at SUMEX, to the Picture System at UCSD. III. Use of SUMEX during the follow-on grant period (8778 - 7783) A. Long-range goals Our current research grant extends through April, 1979. During that time we intend to bring the structure modeling system to a level of performance that permits reliable qualitative interpretation of high resolution e.d. maps, derived from real data and a correct amino acid sequence. We also plan to exploit the flexibility of the rute-based control structure to permit investigation of alternate problem-solving strategies and modes of explanation of the program's reasoning steps. Beyond the next two years, emphasis will be placed on expanding and generalizing the system to relax the constraints of resolution and accuracy in the input data. B. Justification for continued use of SUMEX The biomedical relevance of the protein structure modeling project, coupled with the need for building a computational system with a significant component of symbolic inference, qualifies the project as an AlM-relevant endeavor. SUMEX provides an excellent computing environment for creating and debugging programs Cin a variety of languages), for sharing and distributing information among geographically dispersed co-workers, and for keeping up with current research in other AIM areas. Our project is clearly too small to justify an independent computing facility, and other large computer centers that are conveniently accessible do not fulfill our requisites. Consequently SUMEX has been and hopefully will continue to be an integral research tool in this project. 181 J. Lederberg & E. Feigenbaum Section 4.3. PILOT AIM PROJECTS 4.3 PILOT AIM PROJECTS The following are descriptions of the informal pilot projects currently using the AIM portion of the SUMEX-AIM resource pending funding, and full review and authorization. J. Lederberg & E. Feigenbaum 182 COMMUNICATION ENHANCEMENT PROJECT Section 4.3.1 4.3.1 COMMUNICATION ENHANCEMENT PROJECT Communication Enhancement Project John 8. Eulenberg, Ph.D. and Carl V. Page, Ph.D. Department of Computer Science Michigan State University 1) Summary of research program. A) Technical goals. The major goal of this research is the design of intelligent speech prostheses for persons who experience severe communication handicaps. Essential subgoals are: (1) Design of input devices which can be used by persons whose movement is greatly restricted. (2) Development of software for text-to-speech production. (3) Research in knowledge representations for syntax and semantics of spoken English in restricted real world domains. (4) Development of micro-computer based portable speech prostheses. B) Medical Relevance and Collaboration. Members of our group are in touch with Dr. Kenneth Colby and his group at UCLA who are working on similar protlems for a domain of people who have aphasia. The need for such technology in the medical area is very great. Millions ef people around the world lead isolated existences unable to communicate because of stroke, traumatic brain injury, cerebral palsy or other causes. The availability of inexpensive micro-processors and voice synthesizers allous development of complex experimental systems to study human communication. The knowledge gained from these experimental systems should lead in a few years to prototypes of very low cost which will permit many people to engage in the vital acts of communication required for a "normal" life in human society. We have organized institutes to bring together the many professionals who have an interest in this area. Together with the Tufts New England Medical Center, the TRACE Center for Research and Development for the Severely Communicatively impaired of U. of Wisconsin, and the Children’s Hospital at Stanford (Maurice LeBlanc). We have begun the first newsletter for dissemination in this area. Called "Communication Outlook", the first issue will be published in April, 1978. Subscribers and contributors to the Newsletter come from a vide variety of disciplines and from many countries. John B. Eulenberg helped to organize the first Federal workshop for governmental agencies who have some 183 J. Lederberg & E. Feigenbaum Section 4.3.1 COMMUNICATION ENHANCEMENT PROJECT interest in funding work in these areas. Represented were the Bureau of Education for the Handicapped, The Veterans Administration, The Civil Service Commission, NIH, NSF, and others. We have also been in touch with United Cerebral Palsy associations at the state and national levels. Much of our effort has been in educating those medical, educational, and governmental communities with an interest in this area on the available technology since most of them are not accustomed to funding the development of high-technology systems. C) Progress summary. Although some facets of the research have been underway at MSU for several years, we have been using SUMEX-AIM for only about a year at this time, having received our password in March, 1977. During the last year, we have: 1) Designed and built hardware and software allowing us to transmit files to SUMEX from our Nova 2/710 at 300 baud. 2) Organized a research team of 4 students possessing background in artificial intelligence lead by Dr. Carl V. Page to start a semantics- speech generator. This group had a very primitive prototype (written in running in June, 1977. The system uses statistical, grammatical and semantic information to generate sentences by anticipation. We are organizing a similar group again this month (we have a seasonal supply AI students) to expand the semantics. 3) Converted a large program (Orthophone) for English text to speech synthesizer codes to SAIL from Algol. 4) Obtained local support for terminals and tie-lines to use the SUMEX-AIM facility. We requested these in our original proposal but were not granted them. At present, the lack of a dedicated tie-line from East Lansing to Tymshare in Ann Arbor or Detroit is a problem for us during 0600 to 0900 PST. 5) During the past year, Dr. Reid of our project refined a wheel-chair portable personal communication system for a 10 year old boy who has cerebral palsy. It is micro-computer based and can accept inputs via an adaptive switch from a series of menus displayed on a TV screen, via Morse code, or by a keyboard. Its outputs can be TV display, hard copy, spoken English, Morse code, or musical sounds. As the memory available for small systems will soon be substantial, we will need to specify the content and connection of the choice menus using the knowledge gained in our SUMEX-AIM project. 6) A Doctoral thesis in the association of knowledge sources (letter and word frequencies, syntactics, semantics, pragmatics and belief systems) for the generation of speech has been started by one of our students, Mr. James Soddy (Supervised by Carl V. Page). Mr. Soddy will use the SUMEX-AIM system during Summer 1978 to program some examples for his thesis, as a means of obtaining current information from the AI community, and to communicate with Dr. Page who will be working in Sunnyvale, Calif. for the summer . J. Lederberg €& E. Feigenbaum 184 COMMUNICATION ENHANCEMENT PROJECT Section 4.3.1 7) We have built and tested a myoelectric interface and used it (together with a miniature FM transmitter) for input of changing muscle potentials into a computer. There is reason to believe that this means of input may provide a higher bit rate than other known means for those people who possess severe cerebral palsy. 8) We have developed software for teaching basic educational concepts to severely impaired persons. For example we have developed a "talking" system for drilling students in Bliss symbolics. Another system we have developed teaches spelling using a voice synthesizer and TV screen. BD) Up-to-date list of publications. (1976 to date) By John B. Eulenberg "Technical Systems Development, Headin", Interim Report, April, 1976, Experimental Applications of Two-Way Cable Delivery, NSF Grant No. APR 75- 14286, "Interactive Nen Hired Information Access System with Both Voice and Hard Copy Output: User's Guide to NHQUERRY", April 11, 1976 (With Steven Kludt and Jerome Jackson (Artificial Language Laboratory Report AEB 041176)) "Language Individualization in a Computer-Based Speech Prosthesis System", National Computer Conference, New York, June 9, 1976. "Individualization in a Speech Prosthesis System", Proceedings of 1976 Conference on Systems and Devices for the Disabled, June 10, 1976. "The LEAF Language", Interim Report, September, 1976, NSF Grant No. APR 75-14286. "Microprocessor-Based Artificial Language for Communication Prostheses", with M. R. Rahimi, Proc. of the National Electronics Conference, Vol. XXXI, October, 1977. "A programmable Multi-Channel Modem Output Switch", September 22, 1976, with Joseph C. Gehman and Juha Koljonen (Artificial Language Laboratory Report AEB 092276) "SMPTE Time Code Interface and Computer-Controlled Video Switcher", with Michael Gorbutt and Dennis Phillips, Interim Report, March, 1977 NSF Grant APR 75- 14286. "Representation of Language Space in Speech Prostheses", with R. Reid and MN. Rahimi, Proc. of Fourth Annual Conference on Systems and Devices for the Disabled, June, 1977. "Administration and Management of a Computer-Based Communication Enhancement Program", with M. R. Rahimi and L. Neiswander, Proc. of Amer. Acad. for Cerebral Palsy and Developmental Medicine, October, 1977. "“WNhen [-VOICE] becomes [+VOICE]- The Phonological Competence of People Who Cannot Speak", with Carol Myers Scotton, Proceedings of the Annual Confer. of the Linguistic Soc. of America, December, 1977. 185 J. Lederberg & E. Feigenbaum Section 4.3.1 COMMUNICATION ENHANCEMENT PROJECT By Carl V. Page: "Heuristics for Signature Table Analysis as a Pattern Recognition Technique", IEEE Transactions on Systems, Man and Cybernetics,Vol. SMC-7, No. 2, February 1977. "Discriminant Grammars, an Alternative to Parsing". with Alan Filipski, Proceedings of the IEEE Workshop on Picture Processing, Computer Graphics, and Pattern Recognition, April 22, 1977. "Pattern Recognition and Data structures", Chapter in "Data Structures in Computer Graphics and Pattern Recognition" Edited by Allen Klinger, Academic Press, 1977. "A Survey of Artificial Intelligence in Computer- Aided Instruction", with Alice Gable (Submitted to the International Journal of Man-Machine Systems, March, 1978) E) Funding Status. 1) Current funding. Wayne County (Detroit) Wayne Intermediate School District $75,816 (Third year) Northville Public School District $41,333 (Third year) Jackson Intermediate School District $26,500 (Second year) Ingham Intermediate School District $23,700 (First Year) Michigan State University Division of Engineering Research $64,500 (for each of two years). Grand Rapids Public Schools $2,100 Vandervoot Foundation $5,000 Some of this money has been used to purchase equipment which is the property of WCISD for use in a demonstration classroom in an elementary school. Commitments in the grants have prevented us from using very much of these funds to support long range goals such as those communicated to SUMEX-AIM. However, the special communication devices, student and other research facilities provide the critical mass which will allow us to do the work that we have proposed. The main value of SUMEX-AIM to us is to allow experimentation with AI technology in order to develop the theory to develop intelligent speech prostheses. 2} Pending applications and reneuals. Qaktand County Intermediate School District $100,000. (Application being considered after negotiation) Genessee County Intermediate School District $100,000. Tuscola County Intermediate School District $20,000. Livingston County Intermediate School District $50,000. J. Lederberg & E. Feigenbaum 186 COMMUNICATION ENHANCEMENT PROJECT Section 4.3.1 As one can see from this list of sources, there is a lot of interest in this area from agencies which are not experienced in funding high-technology and research. II) Interactions with the SUMEX-AIM resource. A) Examples of medical collaboration and medical use of programs via SUMEX. The faculty in the MSU College of Human Medicine who teach medical decision making were shown a demonstration of the SUMEX system, MYCIN and PARRY. We have agreed to present a demonstration of PUFF to Dr. Clyde Flory, an allergist who is the most knowledgeable person in our area in pulmonary studies. We intend to explore the possibility of a table-driven program for the treatment of allergy with Dr. Flory. If we decide to undertake the development of this, we will send another proposal to SUMEX~AIM. A member of our Medical School faculty, Dr. Richard Ropple, an expert on myoelectronics, is a member of of our research group. The Dean of our College of Human medicine visited our laboratory in April, 1977 and we have been asked to apply for inclusion in the University's Clinical Center as part of the Rehabilitation Medicine Program. B) Examples of sharing, contacts, and cross-fertilization with other SUMEX-AIM projects. 1. We have met with Or. Kenneth Colby on many occasions including the SUMEX-AIM workshop in June, 1976. OQur work in many ways complements his and we have had several worthwhile interchanges of information. We are converting our major software for speech generation and adaptive inputs to the SUMEX-AIM system in part so that they can be used by Dr. Colby and his group. 2. Mr. Douglas Appelt, a doctoral student at SU-AI was our principal systems programmer last summer. He is currently doing research in the same area as ours with DR. Gary Hendrix of SRI. We have used his knowledge of your system (via the message sending routines) to assist us in starting our project. C) Critique of resource services. Our use of SUMEX-AIM has been seasonal with most programs run during Spring Term. We have used your system for work that could not be done conveniently on our computers. He have been pleased with the system and find it as easy to use as one that is close to us geographically. Dr. Page will be working in the Bay Area this summer and plans to visit your facility as well as use it to keep in touch with the work at East Lansing. 187 J. Lederberg & E. Feigenbaum Section 4.3.1 COMMUNICATION ENHANCEMENT PROJECT III) Foltlow-on SUMEX grant period (8/78-7783). A) Long-range user project goals and plans. We want to do fundamental research in artificial intelligence in the context of the generation of speech from very minimal amounts of input. This problem seems closely related to the understanding of speech. It seems that the methods of representation of knowledge used for speech or vision understanding can be used in a natural way for fluent generation of speech. Our area seems almost unique in AI in that it is socially desirable (without question). Even relatively primitive systems can improve the quality of life for hundreds of thousands of people. Major long range goals are: 1) To do research in transposing the vocal tract to another region of the body in which an individual has suitable myoelectric control for the generation of speech. 2) To define a suitable system of semantics and to encode world knowledge in that system that would be useful for the generation of speech fluently. 3) To discover primitive operations on semantics which allow new and appropriate combinations of speech to be generated. (Using other sources of knowuledge.) 4) To develop means for a severely handicapped individual to program and personalize his or her own speech and environmental control system. 5) To study means of using speech output to aid blind persons both through experiments with simplified text to speech devices and through means of teaching blind persons to write in cursive. 6) To study the effect of communication aids technology on the psychological assessment of individuals previously thought to be retarded and to study the consequences of this technology for the educational system. 7) To improve the prosodic qualities of generated speech, using its semantic aspects. 8) To design portable speech prostheses which allow maximum use of state of the art knowledge in speech generation. 9} To develop an experimental base for studying how the concepts which are articulated in speech are manipulated by individuals at differing states of mental organization 10) To study the potential for speech generation systems as a means of stimulating autistic children. 11) To develop voice recognition systems which will aid individuals with limited speech to develop their full potential. J. Lederberg & E. Feigenbaum 188 COMMUNICATION ENHANCEMENT PROJECT Section 4.3.1 (We don't expect to finish all of these by 1983. ) B) Justification for continued use of SUMEX by your project. 1) We need to use many sources of knowledge represented in computers te do our work, similar to many SUMEX users. 2) We know kindred spirits in the Al community who possess goals similar to our long range goals. 3) We have substantial hardware and software expertise which we are willing to share... 4) The payoff to society of our research in terms of the improved quality of millions of human lives seems great. 5) This area does not have a traditional means of support for research separate from development which makes your support vital for our long range goals. 6) Our area is very interdisciplinary and the communication aspects of SUMEX-AIM will be increasing valuable to us. C) Plans For Other: Computational Resources. We have available to us three mini-computers: a Nova 3/10, a PDP-11734 and a PDP-11745 as well as the cDe¢ 6500 and coc 6400 of our central computer facility. Our demonstration classroom in a Detroit suburb will open soon making a Nova available to students who experience severe communication and motoric handicaps. None of our small systems possess AI software. We hope to develop prototype systems on SUMEX-AIM within the next year the can be used in our demonstration classroom. We believe that the injection of quite small amounts of Al technology into the speech generation systems can produce significant improvements in the communication and educational processes. We will be ina position to measure the effectiveness of the AI tools which we try. If we can make a case on a cost effectiveness basis, there are sources of support to acquire appropriate hardware to service a larger group of students and also support our research. At this point we do not feel the need to try to acquire a PDP-10 or other machine suited to AI research because although important to us, the number of machine cycles we require is relatively small. D) Recommendations for Future Community and Resource Development. 1) We would find it helpful if there were means within SUMEX-AIM to assist in the transfer of prototypes written for your system to various common minicomputers. 2) We have available software to assist handicapped individuals in using computer systems. We would be happy to facilitate the use of SUMEX-AIM resources by a blind programmer by modifying appropriate softnare. 189 J. Lederberg & E. Feigenbaum Section 4.3.2 COMPUTERIZED PSYCHOPHARNACOLOGY ADVISOR 4.3.2 COMPUTERTZED PSYCHQPHARMACOLOGY ADVISOR A Computerized Psychopharmacology Advisor Jon F. Heiser, M.D. Dept. of Psychiatry and Human Behavior University of California at Irvine I. Summary Research Program A. Technical Goals We propose to develop a computer-based automated system for education and consultation in clinical psychopharmacology. Our technical goals are envisioned in three phases: To develop a model of expert teaching, consulting and decision-making in clinical psychopharmacology. To implement this model on a computer system which responds in real time and communicates in natural language. To evaluate the performance of the system as a teaching and consulting aid. B. Medical Relevance and Collaboration. 1. Medical Relevance. For many years, it has been recognized that potent psychopharmacological agents are frequently used in an unsystematic manner. There are at least 50 discrete syndromes currently identified in clinical psychiatry which have unique hierarchies of plausible pharmacological treatments. Each therapeutic regimen in each hierarchy may involve several classes of drugs which can be preferentially ranked.