Seong Ki Mun

Improving the distinction between benign and malignant breast lesions : the value of sonographic texture analysis

To improve the ability of ultrasound to distinguish benign from malignant breast lesions, we used quantitative analysis of ultrasound image texture. Eight cancers, 22 cysts, 28 fibroadenomata, and 22 fibrocystic nodules were studied. The true nature of each lesion was determined by aspiration (for some cysts) or by open biopsy. Analysis of image texture was performed on digitized video output from the ultrasound scanner using fractal analysis and statistical texture analysis methods. The most useful features were those derived from co-occurrence matrices of the images. Using two features together (contrast of a co-occurrence matrix taken in an oblique direction, and correlation of a co-occurrence matrix taken in the horizontal direction), it was possible to exclude 78% of fibroadenomata, 73% of cysts, and 91% of fibrocystic nodules while maintaining 100% sensitivity for cancer. These findings suggest that ultrasonic image texture analysis is a simple way to markedly reduce the number of benign lesion biopsies without missing additional cancers.

Three-dimensional computer-simulated prostate models: lateral prostate biopsies increase the detection rate of prostate cancer.

OBJECTIVES Urologists routinely use the systematic sextant needle biopsy technique to detect prostate cancer. However, recent evidence suggests that this technique has a significant sampling error. We developed a novel three-dimensional (3D) computer-assisted prostate biopsy simulator based on whole-mounted step-sectioned radical prostatectomy specimens to compare the diagnostic accuracy of various prostate needle biopsy protocols. METHODS We obtained digital images of 201 step-sectioned whole-mounted radical prostatectomy specimens. 3D computer simulation software was developed to accurately depict the anatomy of the prostate and all individual tumor foci. Additional peripheral devices were incorporated into the system to perform interactive prostate biopsies. We obtained 18 biopsies of each prostate model to determine the detection rates of various biopsy protocols. RESULTS The 10- and 12-pattern biopsy protocols had a 99.0% detection rate; the traditional sextant biopsy protocol rate was only 72.6%. The 5-region biopsy protocol had a 90.5% detection rate and the 14-pattern, which includes all the biopsies used in the patterns above, only added 1 additional positive case (99.5%). Transitional zone and seminal vesicle biopsies did not result in a significantly increased detection rate when added to the patterns above. Only one positive model was obtained when the transitional zone biopsies were added. The lateral sextant pattern had a detection rate of 95.5%, and the 4-pattern lateral biopsy protocol had a 93.5% detection rate. CONCLUSIONS Our results suggest that all the biopsy protocols that use laterally placed biopsies based on the 5-region anatomic model are superior to the routinely used sextant prostate biopsy pattern. Lateral biopsies in the mid and apical zones of the gland are the most important.

Automatic lung nodule detection using profile matching and back-propagation neural network techniques.

The potential advantages of using digital techniques instead of film-based radiography have been discussed extensively for the past 10 years. A major future application of digital techniques is computer-assisted diagnosis: the use of computer techniques to assist the radiologist in the diagnostic process. One aspect of this assistance is computer-assisted detection. The detection of small lung nodule has been recognized as a clinically difficult task for many years. Most of the literature has indicated that the rate for finding lung nodules (size range from 3 mm to 15 mm) is only approximately 65%, in those cases in which the undetected nodules could be found retrospectively. In recent published research, image processing techniques, such as thresholding and morphological analysis, have been used to enhance true-positive detection. However, these methods still produce many false-positive detections. We have been investigating the use of neural networks to distinguish true-positives nodule detections among those areas of interest that are generated from a signal enhanced image. The initial results show that the trained neural networks program can increase true-positive detections and moderately reduce the number of false-positive detections. The program reported here can perform three modes of lung nodule detection: thresholding, profile matching analysis, and neural network. This program is fully automatic and has been implemented in a DEC 5000/200 (Digital Equipment Corp, Maynard, MA) workstation. The total processing time for all three methods is less than 35 seconds. In this report, key image processing techniques and neural network for the lung nodule detection are described and the results of this initial study are reported.

Technology improvements for image-guided and minimally invasive spine procedures

This paper reports on technology developments aimed at improving the state of the art for image-guided minimally invasive spine procedures. Back pain is a major health problem with serious economic consequences. Minimally invasive procedures to treat back pain are rapidly growing in popularity due to improvements in technique and the substantially reduced trauma to the patient versus open spinal surgery. Image guidance is an enabling technology for minimally invasive procedures, but technical problems remain that may limit the wider applicability of these techniques. The paper begins with a discussion of low back pain and the potential shortcomings of open back surgery. The advantages of minimally invasive procedures are enumerated, followed by a list of technical problems that must be overcome to enable the more widespread dissemination of these techniques. The technical problems include improved intraoperative imaging, fusion of images from multiple modalities, the visualization of oblique paths, percutaneous spine tracking, mechanical instrument guidance, and software architectures for technology integration. Technical developments to address some of these problems are discussed next. The discussion includes intraoperative computerized tomography (CT) imaging, magnetic resonance imaging (MRI)/CT image registration, three-dimensional (3-D) visualization, optical localization, and robotics for percutaneous instrument placement. Finally, the paper concludes by presenting several representative clinical applications: biopsy, vertebroplasty, nerve and facet blocks, and shunt placement. The program presented here is a first step to developing the physician-assist systems of the future, which will incorporate visualization, tracking, and robotics to enable the precision placement and manipulation of instruments with minimal trauma to the patient.

Building an integrated clinical and research network

At the Georgetown University Medical Center Department of Radiology we are currently involved in integrating three diverse networks into a coherent whole. We are installing a new Radiology Information System (RIS) and a new Picture Archiving and Communication System (PACS) as well as upgrading our existing research network, which provides Internet access, to add office automation tools. To accomplish this, many issues have to be resolved. Users of the different systems have different requirements and must have different levels of access to data on the various systems. For example, researchers need access to Internet resources and e-mail while data from the clinical systems must be protected from the outside world. Physicians and some other users on the non-clinical network also require fast and convenient access to the RIS and PACS for clinical uses. Parts of the network should be shielded from the heavy image traffic created by the PACS. On the other hand, because clinical data is also used for research, a connection between the networks is necessary. Our solution for providing adequate access for all users, assuring confidentiality for patient data, and managing network traffic will be described.

Noise impact on error-free image compression

Some radiological images with different levels of noise have been studied using various decomposition methods incorporated with Huffman and Lempel-Ziv coding. When more correlations exist between pixels, these techniques can be made more efficient. However, additional noise disrupts the correlation between adjacent pixels and leads to a less compressed result. Hence, prior to a systematic compression in a picture archiving and communication system (PACS), two main issues must be addressed: the true information range which exists in a specific type of radiological image, and the costs and benefits of compression for the PACS. It is shown that with laser film digitized magnetic resonance images, 10-12 b are produced, although the lower 2-4 b show the characteristics of random noise. The addition of the noise bits is shown to adversely affect the amount of compression given by various reversible compression techniques. The sensitivity of different techniques to different levels of noise is examined in order to suggest strategies for dealing with noise.

Classification of microcalcifications in digital mammograms using trend-oriented radial basis function neural network

Abstract We proposed some novel classification features for the microcalcification of mammograms, and selected the effective combined features using Karhunen–Loeve (KL) transformation followed by the restricted Euclidean distance measure, and finally applied the proposed trend-oriented radial basis function neural network (TRBF-NN) to distinguish the benign group from the malignant group and evaluate the performance with the round-robin method. The two-dimensional KL features were more distinguishable than the raw two-dimensional features. The TRBF-NN was able to define the more generalized distribution than those distributions defined by the conventional RBF-NNs. According to the receiver operating characteristic analysis, the proposed system performed better than two trained radiologists.

Lateral biopsies added to the traditional sextant prostate biopsy pattern increases the detection rate of prostate cancer

Urologists routinely use the systematic sextant needle biopsy technique to detect prostate cancer. However, recent evidence suggests that this technique has a significant sampling error and data based upon whole-mounted step-sectioned radical prostatectomy specimens using a three-dimensional computer-assisted prostate biopsy simulator suggests that an increased detection rate is possible using laterally placed biopsies. The simulated 10-core biopsy pattern (traditional sextant biopsy cores and four laterally placed biopsies in the right and left apex and mid portion of the prostate gland) was shown to be superior to the traditional sextant biopsy. The objective of this pilot study was to confirm the higher prostate cancer detection rate obtained using the 10-core biopsy pattern in patients. We reviewed data on 35 consecutive patients with a pathologic diagnosis of prostate cancer biopsied by a single urologist using the 10-core biopsy pattern. The frequency of positive biopsy was determined for each core. Additionally, the sextant and 10-core prostate biopsy patterns were compared with respect to prostate cancer detection rate. Of the 35 patients diagnosed with prostate cancer, 54.3%(19/35) were diagnosed by the sextant biopsy only. The 10-core pattern resulted in an additional 45.7%(16/35) of patients being diagnosed solely with the laterally placed biopsies. The laterally placed biopsies had the highest frequency of positive biopsies when compared to the sextant cores. In conclusion, biopsy protocols that use laterally placed biopsies based upon a five region anatomical model are superior to the routinely used sextant prostate biopsy pattern.

Optimal Biopsy Protocols for Prostate Cancer

Prostate cancer is the second leading cause of cancer-related death among American men. Biopsy for prostate cancer is a procedure known as transrectal ultrasound-guided needle biopsy. Because of the low resolution of ultrasound, the urologist cannot usually distinguish between cancerous and healthy tissue. For this reason, most biopsies follow standard protocols based on long-term experience of physicians. Recent studies indicate that these protocols have a significant rate of false negative diagnoses. In this research we use real prostate specimens removed by prostatectomy to develop a 3-D distribution map of cancer in the prostate, and use this to develop optimized biopsy procedures. The new procedures have detection rates that are significantly higher than those of current procedures, and thus have the potential to increase the rate of early detection of prostate cancer.

A Telemedicine Consultative Service for the Evaluation of Patients With Urolithiasis

OBJECTIVES A 6-month pilot teleconsultative project linking Georgetown University Medical Center (GUMC) in Washington, DC, and City Hospital in Martinsburg, West Virginia, 90 miles away, was designed to assess the effectiveness of telemedicine on the clinical decision-making process for patients with urolithiasis. METHODS The telemedicine system designed and tested for this project was based on a PC-based platform. Videoconferencing and review of the patients imaging studies were performed over an Integrated Service Digital Network (ISDN) with 3 Basic Rate (BRI) ISDN lines providing a 336-kilobytes/s bandwidth through an Inverse Multiplexor (IMUX). Treatment options were recorded for the clinical trial group and a simulated study group by the consulting urologist after the initial telephone consultation, after the telemedicine consultation, and after examination of those patients transferred to GUMC. RESULTS A total of 32 telemedicine consultations were performed: 14 in the clinical trial group and 18 in the simulated study group. The recommendation of the consulting urologist at the tertiary center was altered in 12 patients (37.5%) after the telemedicine consultation compared with the recommended treatment after the initial telephone consultation. CONCLUSIONS In the evaluation of patients with urolithiasis, this telemedicine application enhanced the clinical decision-making process by allowing for improved quality of care through immediate access and effective transfer of information between the referring urologist, the patient, and the stone center specialist.