Page images

Mr. CALLAHAN. The notes or diagrams he makes at his study or laboratory are enlarged on the screen here as he writes. Thus, a new capability will be added to graduate study or training programs through the application of a visual dimension to basic telelecture service,

(The telewriting illustration received in the caucus room in Washington appears on facing page.)

Mr. CALLAHAN. We would like now to use both these communications media to bring you Dr. Laughlin's presentation directly from New York City.

Senator NELSON. Go ahead.
Mr. CALLAHAN. Dr. Laughlin?



Mr. CALLAHAN. We are ready for your presentation.
Dr. LAUGHLIN. Very well.

This particular example pertains to the application of both computers as well as teletype transmission in the field of radiation therapy. One of the major problems in radiation therapy is the use of radiation sources to treat cancer lesions inside the body.

A typical external radiation source, in this particular case one of our linear accelerator X-ray machines, is used to irradiate the patient externally from outside.

Now, one of the major problems facing the radiation therapist is how to localize the radiation dose inside the lesion and decrease the

[graphic][ocr errors]

radiation damage to the surrounding healthy tissues on the way to the lesion.

In order to approach this problem comprehensively, taking into full consideration the structure of the body, it has been necessary to go to i automatic computers.

The treatment planning procedure includes taking the patient's contour in relation to the size and position of the cancer and constructing dose curves which describe the extent to which the X-ray beams is attenuated as it passes through various tissues. If it is planned to direct the radiation beam from several different angles, the amount of radiah tion delivered to any given tissue must be calculated accordingly.

Now, another approach to the same situation, again assuming that we have a lesion in the lung, would be the actual physical implantation of radioactive sources within the patient. This would involve a surgical procedure by the radiation therapist in which radioactive needles are physically implanted in the tumor.

If this were done according to any particular pattern, a specific distribution of dose could be achieved. But just as in the previous case where external radiation was employed, it is particularly important that this patient receive the actual distribution of the radiation of dose as planned by the therapist.

These are the two methods which are generally employed, both of which require accurate computation.

We have indicated the objective of what we call realistic treatment planning. This is essentially with the necessity of taking into consideration the inhomogeneous structure of the body.

[ocr errors]


Programs have also been designed to determine at which angles to aim the beam to give the optimum treatment to the tumor and the least radiation to normal tissues. The major result of the program is a three-dimensional distribution of dose in the patient, produced by the radiation fields as specified by the therapist.

With the use of computers in the treatment planning, not only can more information be obtained in a short period of time, but the information is highly refined, so that the radiation therapist can know precisely how much radiation he will deliver to any specific area.

A major recent development has been the extension of the computerized treatment planning carried out in individual centers to : affiliated hospitals through the use of special interface equipment and telephone lines. The Memorial Computerized Treatment Planning Service is a pioneer effort in this respect and has been developed to make available to other hospitals the advantages of computer facilities and the associated treatment planning staff.

The need to extend the facilities of computerized treatment planning to radiation therapists in other hospitals which have limited treatment planning facilities has long been recognized. In the past, ocassional treatment plans have been hand-carried by messenger from various other hospitals in the metropolitan area for computation at Memorial Hospital. In order to provide a more convenient and practical treatment planning service for collaborating hospitals, the present teletype-linked treatment planning service was developed and put into operation in March of 1967. The initial two collaborating hospitals were Mount Zion Tumor Institute, San Francisco, Calif., and St. Luke's Hospital, New York City, N.Y. The teletype treatment planning system was demonstrated at the American College of Radiology Computer Conference at the University of Missouri in April 1967. The input and output for treatment plans are transmitted via teletype over regular communication channels. On the average about 30 treatment plans per month are being computed by Memorial Hospital for each collaborating institution.

New programs are being continually developed, according to need. The latest programs will compute dose distributions for any number of nonconverging fields, for any number of nonconverging arc rotations, for any number of wedged fields, as well as for any combination of weighted fields of different field size. Provisions are also made to incorporate corrections for any number of inhomogeneities as well as irregularities in the patient's surface. The computer programs for internal sources are capable of producing dose distributions in any plane at any desired orientation with as many as 100 implanted sources.

Mr. CALLAHAN. Thank you, Dr. Laughlin.
Mr. Chairman, do you have a question of Dr. Laughlin?

Senator NELSON. Yes. I will have a question for you in just a moment, Doctor.

Are you still on, Doctor?
Dr. LAUGHLIN. Yes; right here.

Senator NELSON. Do I understand, then, that the radiologist can send to the computer the various computations, say, as to the size

[ocr errors]
[ocr errors]
[ocr errors]
[ocr errors]
[ocr errors]

and weight and age of the patient and the location of the carcinoma i and that, then, the computer actually determines the quality of the

X-radiation that will be used? That is, it decides how much, how OT

much kilovoltage and how much milliamperage, for example, in this treatment process?

Dr. LAUGHLIN. Yes; the computer is capable of providing the optimum treatment. The way in which it is normally used, however, is that the radiologist would suggest a proposed treatment utilizing a specific radiation energy. The computer would compute this out and then the radiologist conducting the examination would say this is quite an acceptable treatment, or can you not do even better to avoid the radiation dose to a particular area. This would require a repetition of the computer analysis and would be a process of successive approximation to the optimum dose distribution.

Senator NELSON. Does the computer determine what the safety factor is and how many roentgens a patient can take over a certain period, or is that left up to the radiologist?

Dr. Laughlin. That is definitely left to the radiologist. It is his business to have to determine how many roentgens of dose should be administered. But the computer can show him how that dose is going to be distributed throughout the patient's body and warn him of excessive dose level that is going to a healthy tissue region.

Senator Nelson. Then the computer does have to have among its statistics all the factors about the size of that patient, obviously?

Dr. LAUGHLIN. Yes. All the data on the anatomy of the patient, both externally and internally, in the regions which will be exposed to radia

tion will be vital. T

Senator NELSON. In summary, what is the main value of using this computer system rather than just having the roentgenologist make all these calculations himself?

Is it a matter of time or is it a matter of time and accuracy!

Dr. LAUGHLIN. It is a matter of both. As a matter of actual practice, if the roentgenologist had to do all these calculations, he ordinarily simply would not have time to do them and the treatment of the patient in that case would have to rely somewhat on judgment. With the use of these kinds of facilities, he is able to give to all patients a very comprehensive type of analysis which would be prohibitive timewise.

Senator NELSON. Thank you very much, Doctor. It was a very informative presentation. The committee appreciates very much your giving so generously of your time.

Dr. LAUGHLIN. Thank you.
Senator NELSON. Did you have any other questions?

Mr. CALLAHAN. No, thank you, Mr. Chairman, and thank you very much, Dr. Laughlin.

Dr. LAUGHLIN. Thank you.

Mr. CALLAHAN. Mr. Chairman, we will now move to discussion of medical information collection systems, which I have listed as item No.9 on the agenda.

The communication system I will now report on is another unique and dramatic approach to collecting patient medical data. The tradi


[ocr errors]
[ocr errors]

tional method of taking and recording medical histories and the calculation of test reports involves considerable time for the physician.

As an example, a physician investigating the symptoms of allergy may have to ask the patient several hundred questions and devote several hours of his time. A computer technique now under development at the University of Wisconsin Medical School collects medical history information directly from the patient. It is so simple in operation that it can be used by the patients themselves.

A teletypewriter keyboard provides direct input to the computer. As used at Wisconsin in the investigation of symptoms of allergy, the computer questionnaire has been designed to imitate the normal physician-patient relationship as closely as possible. Simple questions are presented to the patient either on a cathode ray tube similar to a TV screen or a teletypewriter.

A typical question would be, “Have you ever had an allergic reaction to penicillin?" to which the patient replies “Yes," "No," "I do not know," or "I do not understand the question," by simply pressing one of four appropriate keys. The choice of questions presented and the order of their presentation is controlled by the patient's response.

A “Yes” response to a general question is followed by a series of specific qualifying questions. A “No” response results in the skipping of such qualifying questions and the presentation of another general question.

When the answers to questions are not known or not understood, these questions are clarified and explained with varying and simplified technology. The minimum number of questions to be answered by a patient who has no allergy and has taken none of the potentially allergenic drugs included in the program is 29. The maximum number of questions that can be answered by any patient is 320, but there are more than 500 questions in the program.

A printed summary is generated by the computer in each case and the response to every question is recorded on tape and stored in the computer.

At this point, with the assistance of my associate, Mr. Murray, we will call the computer and set it up for a demonstration which will take 2 or 3 minutes.

Senator NELSON. I understand, then, that the computer bank, for example, contains a series of questions on allergies. A patient can sit down in front of this machine and the computer can ask the questions without the necessity of the doctor's presence. The computer will, in a very comprehensive fashion, explore this particular patient relationship to any allergy problem that has been programed and the doctor can then use this information in place of an extensive consultation interview with the patient. Is that correct?

Mr. CALLAHAN. Yes, sir, and after the demonstration Dr. Meyer, who is here with us from the University of Wisconsin, will also discuss in some detail more about the medical aspects of this.

Actually, the best way to get an appreciation of the relationship that is established between patient and computer is for someone with no previous experience to play the role of the patient.


« PreviousContinue »