Biomedical subjects

R J Rovinelli

Publications and source records attributed to R J Rovinelli.

3 recordsLinked to original sources

PubMed2001

Calibration of medium-resolution monochrome cathode ray tube displays for the purpose of board examinations.

This report discusses calibration and set-up procedures for medium-resolution monochrome cathode ray tubes (CRTs) taken in preparation of the oral portion of the board examination of the American Board of Radiology (ABR). The board examinations took place in more than 100 rooms of a hotel. There was one display-station (a computer and the associated CRT display) in each of the hotel rooms used for the examinations. The examinations covered the radiologic specialties cardiopulmonary, musculoskeletal, gastrointestinal, vascular, pediatric, and genitourinary. The software used for set-up and calibration was the VeriLUM 4.0 package from Image Smiths in Germantown, MD. The set-up included setting minimum luminance and maximum luminance, as well as positioning of the CRT in each examination room with respect to reflections of roomlights. The calibration for the grey scale rendition was done meeting the Digital Imaging and communication in Medicine (DICOM) 14 Standard Display Function. We describe these procedures, and present the calibration data in. tables and graphs, listing initial values of minimum luminance, maximum luminance, and grey scale rendition (DICOM 14 standard display function). Changes of these parameters over the duration of the examination were observed and recorded on 11 monitors in a particular room. These changes strongly suggest that all calibrated CRTs be monitored over the duration of the examination. In addition, other CRT performance data affecting image quality such as spatial resolution should be included in set-up and image quality-control procedures.

Calibration↗

PubMed2001

Assessing portal design skills in the radiation oncology interactive case management examination.

The American Board of Radiology is developing a computerized interactive case management examination to be used to evaluate the clinical skills of radiation oncologists. In the past, these skills have been evaluated by a pencil and paper written examination and an oral examination. With the increasing capabilities of computers, these skills can be easily, and perhaps even better, evaluated digitally. The aim is to develop an examination, which will be based on actual clinical cases, and be interactive so that it better mimics the clinical practice of a radiation oncologist than a written examination. It will also be less labor-intensive and less expensive than an oral examination. One of the most important skills of a radiation oncologist is the ability to design treatment portals that will encompass the entire cancer and yet minimize the irradiation of critical tissues and normal organs. Important parameters for radiation oncologists include the direction of the treatment beam, the size and shape of the portals, and the location of the margins of the field relative to patient anatomy and tumor location. In order to evaluate a physician's ability to design treatment portals, the computer-based examination has the capability to interactively construct field lines. The computer interface allows the candidate to draw field lines on a digitized x-ray image in a manner similar to practice. After the candidate illustrates the field lines, the evaluation of the response must be performed quickly to avoid interrupting the flow of the examination. The answer key is stored as a lossless compressed image. The key contains three regions consisting of (1) the must include region, which contains the tumor; (2) the must-exclude region, which contains tissues that if damaged would affect patient vitality and quality of life; and (3) the envelope of acceptable curves. Each region is assigned a unique byte code. The candidate's response is assigned a fourth byte code. Using basic logic operations, the response is swiftly evaluated. The scoring algorithm scores a candidate's action as correct if his/her drawn area encompasses all of the "must-include region" and is within the "envelope of acceptable curves." It scores a candidate's action as incorrect if his/her drawn area overlaps any part of the "exclude region" and/or exceeds at any point the "envelope of acceptable curves."

Algorithms↗

PubMed1987

Pilot study using 'dangerous answers' as scoring technique on certifying examinations.

Four medical specialty certifying boards collaborated in an exploratory project to determine whether "dangerous answers," that is, incorrect answers that would be dangerous in patient care, could be identified in written board examinations and used as a potential technique for scoring. Panels of experts in each specialty retrospectively identified in previously administered certifying examinations those multiple-choice questions that contained dangerous incorrect options. The 2,713 certification candidates who took these examinations from 1981 through 1983 were given dangerous answer scores based on their responses to these items. When these dangerous answer scores were compared with their conventional percent-correct scores, it was found that 3 percent of the candidates who failed the examinations by conventional scoring had fewer than the average dangerous incorrect answers. More important, 10 percent of the candidates who passed the examinations by conventional scoring chose a greater than average number of dangerous answers. The implications of these preliminary findings and the limitations of the study are discussed.

Achievement↗