Zenon Protopapas

Fast and effective retrieval of medical tumor shapes

Investigates the problem of retrieving similar shapes from a large database; in particular, we focus on medical tumor shapes (finding tumors that are similar to a given pattern). We use a natural similarity function for shape matching, based on concepts from mathematical morphology, and we show how it can be lower-bounded by a set of shape features for safely pruning candidates, thus giving fast and correct output. These features can be organized in a spatial access method, leading to fast indexing for range queries and nearest-neighbor queries. In addition to the lower-bounding, our second contribution is the design of a fast algorithm for nearest-neighbor searching, achieving significant speedup while provably guaranteeing correctness. Our experiments demonstrate that roughly 90% of the candidates can be pruned using these techniques, resulting in up to 27 times better performance compared to sequential scanning.

Impact of filmless imaging on the frequency of clinician review of radiology images.

The purpose of this study was to determine the impact of filmless imaging on the frequency with which physicians access radiology images and to assess clinician perception of image accessibility using a hospital-wide Picture Archival and Communication System (PACS). Quantitative data were collected at the Baltimore VA Medical Center (BVAMC), prior to and after conversion to filmless imaging, to determine the frequency with which clinicians access radiology images. Survey data were also collected to assess physician preferences of image accessibility, time management, and overall patient care when comparing filmless and film-based modes of operation. In general, there was a significant increase in the average number of radiology images reviewed by clinicians throughout the hospital. However, the one area in the hospital where this trend was not observed was in the intensive care unit (ICU), where the frequency of image access was similar between film and filmless operations. Ninety-eight percent of clinicians surveyed reported improved accessibility of images in a filmless environment resulting in improved time management. The mean clinician estimate of time saved due to the use of PACS was 44 minutes. The study documented a combination of clinician perception of improved accessibility and substantial time savings with the use of a hospital-wide PACS, which was supported by objective measurements. The increased frequency of image review by clinicians and rapid image access should provide a further impetus to radiologists to decrease report turnaround time to provide “added value” for patient care.

Picture archiving and communication system training for physicians: Lessons learned at the Baltimore VA Medical Center

Physicians practicing at the “filmless” Baltimore VA Medical Center need to be proficient in the use of the picture archiving and communication system (PACS) to be able to view radiologic images and accompanying reports. PACS training is necessary to assure optimal patient care and to satisfy potential medicolegal requirements. Providing such training is the responsibility of both the Imaging Department and the hospital. Training in the use of the PACS at the Baltimore VA is conducted by an on-site application specialist. Data were collected from interviews with the trainer, training log sheets, and physician surveys. Although 100% of radiologists received formal training, only 22% of nonradiologists were formally trained; 32% of these physicians identified themselves as having been trained by their peers and 41% stated they were self-trained. We identified two goals of a PACS training program. The first is to teach physicians how to retrieve images and reports from current as well as prior studies and display them on a computer workstation. Secondly, the training should include instruction on the use of the various workstation tools to enhance image interpretation. Imaging requirements and usage by different physician groups vary, and PACS training should be tailored accordingly. Difficulties in the scheduling of training sessions during working hours and the widespread use of a “generic” log-on identification have contributed to the low (22%) compliance of nonradiologists with the formal training program. Although we believe that one-on-one training is most effective and can be best tailored to the needs and computer expertise of an individual particular physician, computer based training (both on and off-line) may provide an acceptable, and in some cases, a preferred alternative.

Picture archiving and communication systems and vascular surgery: clinical impressions and suggestions for improvement.

The purpose of this study was to determine the acceptance and clinical utility of a large scale picture archiving and communication system (PACS) for vascular surgery. Questionnaires and one-on-one interviews were conducted with physicians and nurses in the department of vascular surgery at the Baltimore VA Medical Center where PACS has been in routine, hospital-wide use for more than 21/2 years. The perceptions of the clinical staff were assessed to determine the efficacy of PACS in comparison to the conventional film based alternative for the practice of vascular surgery and suggestions for improvements were solicited. There was consensus among the vascular surgery staff members that the use of PACS enhanced their clinical practices, both in and out of the operating room (OR). Vascular surgeons heavily rely on image display in the OR as a “road map” to help determine their operative approach and to guide their surgery. PACS offers unique intraoperative imaging capabilities including rapid image retrieval and improved archival, cine review, the ability to modify image contrast, and the ability to obtain direct quantitative measurements of the degree of vascular stenosis. The increased accessibility and availability of images throughout the hospital enables improvements in time management and in patient care.

Variation of monitor luminance on radiologist productivity in the interpretation of skeletal radiographs using a picture archiving and communication system

T HE cathode ray tube (CRT) display used in picture archiving and communication system (PACS) workstations have been shown to have lower spatial resolution and luminance compared with conventional film-screen radiographs displayed on an illuminator. Laboratory observer studies indicate display brightness to be an important parameter in the interpretation of radiological images. This study was undertaken to evaluate the effects of varying monitor luminance on radiologist productivity and accuracy in the interpretation of skeletal radiographs with the hope of identifying the ideal monitor luminance for everyday practice. One hundred randomly selected skeletal radiographs were evatuated prospectively by three board certified radiologists with extensive PACS experience. Each examination was reviewed independently on a four-monitor, high-resolution (2048 • 1536 pixels) workstation at three different levels of monitor brightness (40, 65, and 90 footlamberts). Ambient room light was maintained at a constant level throughout the study. Interpretation

Patterns of utilization of computer workstations in a filmless environment and implications for current and future picture archiving and communication systems

ConclusionA workstation monitoring program can be implemented, which automatically creates a log of workstation utilization that can be exported for analysis on a conventional spreadsheet or database analysis program. This methodology has several drawbacks as described, which make it a relatively crude tool in the prediction of peak and average network traffic and optimal workstation deployment. A large-scale PACS should include system administration tools that permit the creation of a utilization log and subsequent analysis of all image and database accesses for each workstation. This tool should not result in measurable degradation of system performance and should be run on an ongoing basis. Such an administrative tool could provide critical information to vendors for system design, to customers for system planning, and to users for system optimization and maintenance.

A prospecive evaluation of the impact of filmless operation on the Baltimore VA Medical Center

Summary The objective of this article is to summarize the results of a 3 year prospective evaluation of the impact of filmless operation on the Baltimore VA Medical Center using a hospital-wide Picture Archival and Comnunication System (PACS) The data were collected both before and after the transition to filmless operation by direct observation, analysis of the Hospital Information System databases, physician surveys, and evaluation of the PACS database. There were substantial reductions in examination reporting times. ‘lost’ or missing images, unread studies, and film retakes as well as significant increases in productivity both by the radiologists and technologists. Clinical surveys were suggested a strong preference for filmless over conventional film-based operation. Preliminary analysis suggests that the savings from filmless operation are offset in the imaging department by the eqipment depreciation costs and of maintenance costs but that the hospital as a whole benfits financially from savings in clinican time associated with the PACS. As a conclusion, analysis of the 3 year experience with filmless operation indicates that it has been successful in improving patient care and may result in economic savings as well.

Effect of screen monitor number on radiologist productivity in the interpretation of portable chest radiographs using a picture archiving and communication system.

p [CTURE archiving and communication systetas (PACS) offer unique improvements in operational efficiency when compared with filmbased imaging by providing economical storage, rapid image retrieval, increased image accessibility, elimination of lost images, and reduction in film retakes. Radiologist produ.ctivity is dependent upon workstations facilitating fast and accurate interpretations. This study evaluates the effect of workstation monitor number on radiologist efficiency and accuracy in the interpretation of portable chest radiographs. Time-motion studies were performed evaluating radiologist interpretation of portable chest radiographs using one-, two-, and four-monitor displays (2048 x 1536 pixels). Three board certified radiologists with extensive PACS experience prospectively evaluated 100 portable chest radiographs that were randomly assigned to one, two, or four monitors. In addition to evaluating the time rcquired for image interpretation, accuracy was assessed comparing individual radiologist interpretation with those of an expert review panel. In all cases, comparison studies and reports were available at the time of image review. As the number of monitors increased, overall time requirements in image display and interpretation progressively decreased. This accentuation in time savings was of greatest magnitude when the number of comparison studies increased and was independent of the radiologist. The overall accuracy of interpretation was not affected significantly by differences in monitor number, with consistent reporting accuracy for all radiologists. Radiologists subjectively reported fatigue to be inversely proportional to monitor number. Use of a four-monitor workstation produces significant time savings when compared with oneand two-monitor workstations in the interpretation of portable chest radiographs. Although overall interpretation accuracy is not affected significantly by monitor number, there is a reported reduction in radiologist fatigue as the monitor number increases, which leads to further enhancements in radiologist productivity.

Analysis of the clinical impact of a DICOM HIS/RIS to modality interface and recommendations for improvement

One ofthe biggest challenges in either a PACS or teleradiology environment is the requirement for accurate and rapid identification ofpatient examinations. Despite the major advances in DICOM and modality interfaces, this challenge has become an important and often frustrating issue at most PACS sites. The use of a DICOM interface does not ensure that the images will be properly matched with the correct patient or study.

Impact of filmless operation on staff productivity in the CT department

Given todays competitive economic environment, maximization ofproductivity is essential to decrease operational costs and maximize patient throughput. A large scale PACS offers the potential to achieve lower operating costs, increased efficiency, and improved quality of care. A Picture Archival and Communications System (PACS) permits direct display of digital images. Such a system may reduce the need for time consuming processing associated with conventional film based systems