Grant No. pRoos.es Section TH}. 8 INDIVIDUAL USER PROJECT DESCRIPTION INVESTIGATOR: DEPARTMENT INSTITUTION: Stanford Comput Emmanuel Mesel, M.D, Pediatric Cardiology Stanford Medi FIELD OF INVESTIGATION PROJECT TIT Direct Measurement of Intracardiac VSD Blood Flow. AMOUNT OF RESOURCE USAGE. 15 52% PROJECT DESCRIPTION (Approximately 300 words) Project VSD is concerned with blood flow through ventricular septal defects (VSD) surgically produced in dogs. Two major sets of comparisons are made: the pattern of flow through the VSD is compared with the pattern of differentia pressure between the left and right ventricles and with the electrocardiogram (ecg); and flow measured by an electromagnetic flow probe (which we consider a primary standard) is compared with flow measured by other techniques used on people (Flick, dye dilution). During the experiment, VSD flow, left and right ventricular pressures, and the ecg are recorded on tape. The more interesting data are selected for A to D conversion and for computation of the differential pressure by program WORKHORSE, Program LISTING lists digitized data, which, when graphed, permits comparison of the pattern of flow with the pattern of differential pressure. As might be expected, we have found that these patterns are very similar even under varying conditions (eg, ectopic beats), with flow slightly delayed with respect to pressure. Program cathlog produces a file which summarizes all our VSD experiments. ~ Future effort will be directed towards the incorporation and use of programs developed in project carcat for pattern recognition of pressure and flow contours. , Section fi- B INDIVIDUAL USER PROJECT DESCRIPTICN INVESTIGATOR. DEPARTMENT INSTITUTION. Emmanuel Mesel Pediatrics Stanford Medical School oy] { ay ae Ae Tord Comtutat FIELD OF HIVESTICATION PROLECT Medical Diagnosis WER AMOUNT OF RESOURCE USAGE. 49,504 PROJECT DESCRIPTION (Approximately 300 words) The project is an investigation of mathematical modeling techniques applicable to medical diagrams. The plan is ultimately to apply the cause- effect modeling techniques developed in reference 1 in an environment that allows online interaction between physician and computer model. Currently programmed is the congenital heart disease model of Warner . a : * and his collaborators“. Also programmed are text editor routines that are being used to speed the preparation of reference 1. Though a program has been written to implement the cause-effect modeling techniques of reference 1 using a Burroughs B5500 computer, adapting even that program to ACME will require considerable effect as the program depends heavily on the nearly unique ability of the B5500 to efficiently handle re- cursion and treat overlay automatically. It is felt that the ability to experiment with the models constructed in a way available only in an online system and that the increased interest and criticism that will result from testing the models produced in a clinical environment justify the effort, l. W.F. Rousseau, A Method for Computing Probabilities in Complex Situations, Doctoral Dissertation, Stanford University (in pteparation). 2. H.R. Warner, A.F. Toronto, L.G, Veasy, R. Stephenson, "A Mathematical Approach to Medical Diagnosis," JAMA , Vol. 177, July 22, 1961, pp 177-183. Grant Ho. pp. OOBL1OG. Section Hl-B INVESTIGATO | DEPARTINT. NSTITUTION ; stant ord Medical Center Emmanuel Mesel, 1D, Pediatrics pe bamzors Camput on Center FIELD OF | { INVE Sj IGA TION ! PROJGCT TITLE: On-line analysis of cardiac catheterize- Carcat tion data. AMOUNT OF RESOURCE USAGE. 186, 106 PROJECT DESCRIPTION (Appfoximately 300 words) Project "'carcat!' analyzes cardiac catheterization pressure tracings in children. From catheters in the right and left heart, pressure tracings are transmitted to the ACME computer, converted to digital data, and analyzed to determine atrial, ventricular, arterial, venous and wedge pressures. Currently the values in millimeter of mercury are calculated for the a and u waves, X and y troughs, and mean pressures in the artia and great veins, for systolic and end-diastolic pressures in the ventricles, for systolic, diastolic and mean pressures in the great arteries, and for mean pressures for the wedge positions. These values are calculated immediately and printed out on the computer terminal in the catheterization room. At this time, efforts are under way to improve and ascertain the accuracy of the algorithms used in pattern recognition for atrial and ventricular pressure tracings. The basic data acquisition and analysis system that has been set up will also be used to store data acquisition and analysis sytem that has been set up will also be used to store data for additional calculations and for the preparation of reports. As data is accumulated in storage from cardiac catheteriza- tions and from other sources of clinical information, it will be possible to analyze large amounts in clinical data rapidly using eh ACME computer. Research into methods of storing and recalling data for analysts ‘of clinical information will be an important part of our future efforts. INDIVIDUAL USER PROJECT DESC INVESTIGATOR INSTITUTION: S mol iitee : a Seonlev ld 7 Stan® ord Computati FIELD OF 5 NGATION PROJECT TITL EL - USAGE. 27,925 PROJECT DESCRIPTION “Approxivately 300 words) Gre WE fil O He G04 I i. ” OF INDIVIDUAL USER PROJECT DESCRIPTION INSTITUTION: . tn me ee ee ee te te cman a es INVESTIGATOR: DEPARTM 3 stanford Medics Stenford Compus NAT ay fact GENRE TCS GY Ville: PROS: BRAIN PROTHIN BIOCHEMISTRY AMOUNT OF RESOURCE USAGE: 123,7h PROJECT DESCRIPTION (Approximately 300 words) \ 2 An inexpensive, easy to realize interface for a Packard # 3514 liquid scintillation counter - IBM 1800 was built and tested. Several support pro- grams written in 1800 Assembly Tanguage end PI/1 comclete the interface. 1 (A full description is available in ACME Note #TPA-1). The interface makes possible direct reading of data into ACME deta files from the counter outout, linst 1 PUT SPAR hi ARG po INVES Wi OATO: fix: tye FIELD OF iN. see below AMOUN ’ OF RESOURS E USAGE 51529 Up ty CT DESCRIP 300 PROJE (Aporoxin ate! HON words) The usage of this terminal under this name actually represents usage by several investigators in this department. Mr. Nye has written most of the programs and his field of usege has been calculation of equilibrium constants of antibody-hapten reactions and structural studies. Dr. Rosenberg has used it for genetic studies of complement in mice. Dr. Stocker has used it fo genetic studies in bacteria, and Dr. Awkraut for statistical studies of the immunologicbulins in man. It has also been used in a vedagosic sense by students of these men as well as for manuscript editin antages of time sharing and data files becoms more evident, and ai sted instrumentation becotrs more commonplace, it is expected that there will be yn considerably more usage by this department. CO INDIVIDUAL USER PROJECT DESCRIPTIO WES: es —— RA ANS THTUTION: Dr. Petralii | Tnfectious Diseases F IL 0 0. Aye STiCAT- GH: i? Pep ie Med-Data pace nee eee Ae ree ee es ee ee ns . ee ene rental on omen eee ce ene mat cence ee nem ob een Ae mene ne nea AMOUNT OF SOU RCE USAGE: 43, Gee PROJECT DESCRIP TIO? ty (Approxivately 300 word This project deals with the data collected in the Ho: store in a form suiteble Laboratory, quality control of the input as well as s for later analysis pital Bacterlolosy Ge 5 or o As conceived the project will proceed as follows: the secretaries will type the information at the terminal. Yhe data will be placed in a tempor: frou which it will be aualyeed for quslity vtuel, Deta moet consis Hh oat previous data will be questioned and perhaps the leboratory test repent deta will then be placed in a complete file end a sorted file, each of which may be used for later analysis. The temsorary file will be used to put out the daily laboratory reports. This step will inelude some calculations such as conversion of sensitivity zone size to "sensitive" or “resistant”. OEE Using the computer to put on daily reports allows the project to proceed without addition of personnel to type in information, The input time of the secretary will be less than the time usually required to type repoxts. The data analysis will give us informoticn about the sensitivitics of various bacteria to antibiotics, This information vill help us to decide which treatnent | to use in certain cases. We will also be able to detect significant changes in sensitivity as well as major trends. On YO Stant ay C Stantord iM Robert W. Porter LD OF UNVES TI SAT ON | ekOszar Fa Kinetics of Aspartate | ATC KIN Transcarbary tase | AMOUNT. OF RESOURCE. USAGE: 58,012 PROJECT BDESSRIFTION (Approxirate!ly 300 words) ATC_KIN contains six programs used for the study of the reaction catalyzed by the enzyme, aspartate transcarbamylase. Program LstSq simply calculates a least-squares linear fit and standard deviation. Program DataFit calculates initial rates of rcaction from experimental data. These data are time points and counts per minute of product at each time point. Initial rates are calculated by a least-squares linear fit; rates are taken from the fitted slopes, converted to molar values using a value for specific radio-activity, and also corrected for enzyme concentration, This program, like the others in Project ATC_KIN, has been written so that it can be operated easily by other workers in the research group without experience in using computers, Other programs are used to fit the various kinetic equations which describe the relation of initial rate to substrate concentration. Program HyperFit fits the simple hyperbolic equation, called the Michaelis-Menten equation, The curve fitting procedure is very crude. For the two constant parameters in this function, initial estimates are provided, with ranges to be tested for both, Ina first step, a coarse fit is obtained by testing all the combinations of the trial values for the two parameters, in coarse steps covering the two ranges. In succeeding steps, the operator provides new, smaller ranges to be tested, repeating this procedure until achieving a sufficiently defined pair of values. Next the data points are scanned for deviations from this fitted curve, and the point with the largest deviation may be rejected, at the option of the operator. If the point is rejected, the fitting process is repeated, giving new values of the two parameters for the best curve, Program DataFit 2 simply gives a least-squares linear fil for the linear equation obtained from the reciprocal form of the Michaelis-Menten equation, first calculating reciprocal values of the data ports, and also calculating the kinetic parameters from the fitted slope and intercept. Vhese values are then used as the initial estimates for use in Programm HyporFit, Program DataFit 1 fits the much more complicated equation which describes the kinetics of the two substrate reaction, or the similar equation for the kinetics in the presence of inhibitor. The cquation fitted is in the simpler reciprocal form, which predicts a family of straight lines having a common intersection, fhe programis designed to select the valucs for the coordinates of the common intersection point which gives the 1) sest value for the deviations of all the experimental points from their corresponding best lines. The fitting procedure is Similar to t.c crude trial-and-error method described for program HyperFit. It should be noted that this curve-fitting procedure requires the use of an on-line communication system. Finally, Progran: ATCase 1] is a manuscript in preparation for publication of these kinetic studies. Ay hy INVESTIGATOR, DEPART EN! INSTITUT! Walter He RS Yai FIELD or Computer i AMOUNT OF RESOURCE USAGE: 24,650 PROJECT DESCRIPTION (‘Approxirais!y 300 words} The "S007" project is a subset of the general work of the Instrumentation Re Laboratory, Cenetics Department, in the ficid of instrumentation research conceived to ansver the question, "What kind of automated beste blological instrusentation would be suitable for interplanetary probes of exobio forme?" Accusl acceso -s Of this laboratory have shed light area and have inmediate here and nov applications in conventional biological and medical research. An example is the computer-directed mass spactrenster implemented by this laboratory and reported in this labo oratory's Technical Report No. IRL 1062. A quadrupole mass spectrometer was uniquely controlled by a computer to achieve a high order of instrument efficiency. The "S007" account supports technical and enginesring development. Prograus to help in engineering design have been written and used. Two such prograns are "RCs" and “Dblfocus." The first of these exeuples vas a streightforverd electrical engineering circuit analysis aid and the second was an evaluation of the accuracy and conplexity of instrumentation n ded for a contemplated mass spectroncter purchase. Other "S007" files have experimental data uscful in the develepment of algorithirs to be used in the control or data ecauleition Be Tees TM a, " o seo les of tte taps modes of ACH. TRACE” and "PICKER" ere excmples of this types. This investicator's a ety 2, es «Leg ce ot + mi de ge compen dt in the tame-cheved instru.sn that ACH! is to develop. is the direct digital connection of the ACH compucors to laboratory dnstrwsents. All of this investisater' 2 se of ACHE S a aD has been eirectly peal. To this dste usage has been in obilicy to serve these direct ¥ instrumentation moods of this lsboratery, pelucorily in the field of ms mi os oF an 2 ere) spectrom2ters, Onee principal goal is the integration of en Associated Electronic Indus m A c 3 p> we ; te o a (AET) model MS-9 mass spectrometer date the ACHE data system, This work is being supported by NIB grant 5 ROL AM 64257-0907. ~ ON ail ic _ Br _ Section HH-B INDIVIDUAL USER PROIECT DESCRIPTION INVESTIGATOR. EPART INSTH Uy ION ‘; A.M, SAUNDERS MoD PATTOTORY stanford Medical School —s : _ ves Stanford Comoutation Contes RIEL ') oF SVEN ETAT IVE CYTOLOGY MAST CHUL YES as AMOUNT Tndividual of 1O00-420C computer. microscope and two are The computer RESOURCE E : 25,456 (Approxirral objects, cells or standard spheres, Ox in a microscope for size and flourescence intensity at a spec wave length. are measured at a meee aoe - ied Data thus tabulated forms the basis for statistical analysis by The computer is used similarly in calculating corrections when the a is used as a spectroflourimeter. Two manuscripts have been accepted in preparation using these facilities. 7 S M Le Ube also used to write the text of the er ae INVESTIGATOR: | DEPARTMENT INSTITUTION F. Me. Seudo Genetles Stanford he, otanrord FIELD OF USAGE: 1h, 655 PROJECT DESCRIPTION (Approximately 300 words) ine program tabulates the results of models for the genetical variability among populations in a linear array, with migration between edjacent colonies. Tne basic quantity is given by the symmetric recursion +E BCR. + +oF =O; Fain * Pap) + BPs * Fay)? Wg = O35 its proper, special solution has vRePorn a. a = Ajay + Ajo where Ay > Ay are very complicated algebraic functions of the parameters. The final quantity is a linear combination of F 's, d up to a few hundred. Thus a d 2 ’ with the precision of this comouter, a too large error would result from its = 3 cay direct application. K ° 9