Theodor D. Sterling

Neoplasia Following Therapeutic Irradiation for Benign Conditions In Childhood

The question of whether or not radiation can be indicted as the principal causative factor in the induction of neoplasia following radiation exposure for either diagnostic or therapeutic purposes has been of increasing interest over the past several years (4, 12, 18, 19, 20). The present investigation was initiated in 1956, in the Cincinnati area, which appeared to offer abundant material for evaluation. The first case of x-irradiation for treatment of “enlarged thymus” was reported from Cincinnati by Friedlander in 1907 (8), and both roentgen diagnosis and therapy of thymic enlargement enjoyed considerable local popularity until recently. Observations from this city were published by Lange in 1914 (11). Since the majority of patients in the series to be reported here were from a group of private and municipal hospitals, none of which is particularly noted as a cancer center, the possibility of bias being introduced by selection of a neoplasia-prone population was thought to be minimal. Historical Review ...

A Practical Procedure for Automating Radiation Treatment Planning

A simple method for the translation of isodose graphs into numerical grids is described and the use of these grids for automatic dose computation of multiple field treatments is demonstrated. Details of the preparations of grid values, I.B.M. cards, and programme for one particular computer are given in detail.

PLANNING RADIATION TREATMENT ON THE COMPUTER

The abundance of artificially produced sources of ionizing radiation has made radiation therapy increasingly convenient and effective. This method has the obvious advantage of avoiding many of the severe traumas resulting from surgical or chemical extraction of tumors. On the other hand, to be effective, the radiologist must deliver an adequate dose to the tumor without destroying vital tissues. In fact, the dose delivered to the tumor is limited by that amount of radiation the surrounding normal tissues may be expected to tolerate. The total shape and pattern of low and high dose areas within the treatment field will thus determine the efficacy of the treatment. Combining multiple fields in planning a course of therapy so that dose to tumor is high and dose to normal tissues low are the major problems that test the skill of the physician. The preparation and evaluation of individual treatment plans for patients has been one of the most vexing problems in radiation therapy. The usual treatment plan is as follows. A contour is first obtained of the body cross section of interest and is then drawn on transparent paper. From radiographs and physical examinations, the tumor is localized and drawn in its proper location, along with the important normal structures such as the spinal cord. To destroy the tumor it is necessary to deliver a lethal dose to it without, however, damaging adjacent vital tissues. Since the effects of radiation are additive, it is possible to build up a large dose at the tumor and smaller doses at other locations by aiming beams through the body at different angles and having them intersect at the location of the malignancy. FIGURE 1 shows an example of a treatment field for a patient (with carcinoma of the esophagus) who was to be treated with Cobalt 60 teletherapy; 80 em. S.S.D. (skin to source distance) and four 8 X 15 em. fields. The fields were to enter at angles of 40° with the vertical. The total dose at any point in the region of interest is the sum of the contributions from each treatment field at that point. The difficulty in evaluating a treatment plan lies in finding the values of combined doses for a large enough number of points in the treatment field so that the effect of the treatment on the tumor, the surrounding area, and specific vital tissues can be assessed.

Use of the computer to teach introductory statistics

It has always been obvious that the aid to calculation offered by the computer forces a change in the curricula of mathematics, statistics, physics, engineering and other courses. Not so obvious are the many pedagogic aids the computer can offer in teaching the subject matter. The possibilities of giving the student a better technical as well as conceptual understanding of statistics were explored for a number of years at the College of Medicine of the University of Cincinnati and are reported here.

A BIOLOGICALLY‐ORIENTED COMPUTER LANGUAGE*

When the life sciences turned towards computers, their major initial concern was with instrumentation and construction of interfaces between tissue and recording equipment or between recording and digital hardware. During these years of learning, the concept of the computer-based robot slowly emerged. This is basically the idea of the digital computer controlling much of the procedures of data processing or the workings of peripheral instrumentation. However, even some modest accomplishments in this work demand a sophistication in software which simply does not exist in present day computer science development. This is true especially for the environment in which this robot operates or the languages with which it can be instructed. Historically (if such a short period of time as we refer to here can be graced by the term history a t all) we saw impressive developments from crude machine languages to the more vernacular-based and much easier to manipulate problem-oriented language. These languages have answered the need of the scientist who designs a context within which he wants his robot to operate. Our existing vernacular-oriented languages (we prefer this term to problem oriented) such a COBOL, FORTRAN, LISP, etc., still operate on the constructional or comprehensive instructional level. The investigator is still forced to instruct the machine a t each step and build a program of what he wants the computer to do. That is, the person preparing the program still must concern himself with the individual mathematical, logical, or iterative steps required to synthesize a given program or data processing procedure. If the life science investigator prepares the program, he must be willing to spell out the detailed specifications and formats to be used for transmitting his prepared data to the computer and for formatting the computer output in acceptable form. If the life science investigator is also a mathematician and wishes to explore new mathematical and computational techniques, then he may have to spell out and solve in detail computational procedures, such as programming problems. The logical extention of problem oriented languages as well as a fulfillment of the need created by the complex work of the biologist lies in supplying a multi-level language system and environment to the medical and biological investigator, which permits him to instruct the computer to use certain procedures rather than to build these procedures himself. We are speaking

The role of the blind in data processing

Imagine that you are asked to evaluate a candidate for a job as a programmer. The applicant appears to be extremely intelligent. He has shown himself to be very inventive. He seems to have a number of outstanding capabilities which point to him as an especially gifted programmer. There is no question but that he is highly motivated to do well in his profession. He can show that he is well trained in all aspects of computation and related topics. He also appears to have a good general education besides being personable. Yet, despite these desirable qualifications, the applicant may be disqualified because he is blind.