J Jacobs

Unenhanced brain CT is useful to decide on further imaging in suspected venous sinus thrombosis

AIM To assess the value of unenhanced brain computed tomography (CT) in the diagnosis of cerebral venous sinus thrombosis (CVST). MATERIALS AND METHODS Unenhanced brain CT images of 26 patients with proven CVST were mixed with 26 age and sex-matched images from patients without CVST. Four readers reviewed the 52 brain CT images and were asked to score the examinations for the absence or presence of CVST on a scale from 0 to 4. The mean density in the different venous sinuses was measured by one radiologist. RESULTS The sensitivity of unenhanced brain CT for the diagnosis of CVST was 73%. There were no false-positive readings. A receiver-operating characteristic (ROC) analysis on these data resulted in an area under the curve of 0.86. Density measurements proved to be helpful, but could not detect all cases of CVST. CONCLUSION Unenhanced brain CT is a valid initial radiological examination in the diagnosis of CVST. Due to the absence of false-positives in the present series, unenhanced CT can be used to decide whether further imaging with CT angiography or magnetic resonance angiography is required.

Simulation of 3D objects into breast tomosynthesis images.

Digital breast tomosynthesis is a new three-dimensional (3D) breast-imaging modality that produces images of cross-sectional planes parallel to the detector plane from a limited number of X-ray projections over a limited angular range. Several technical and clinical parameters have not yet been completely optimised. Some of the open questions could be addressed experimentally; other parameter settings cannot be easily realised in practice and the associated optimisation process requires therefore a theoretical approach. Rather than simulating the complete 3D imaging chain, it is hypothesised that the simulation of small lesions into clinical (or test object) images can be of help in the optimisation process. In the present study, small 3D objects have been simulated into real projection images. Subsequently, these hybrid projection images are reconstructed using the routine clinical reconstruction tools. In this study, the validation of this simulation framework is reported through the comparison between simulated and real objects in reconstructed planes. The results confirm that there is no statistically significant difference between the simulated and the real objects. This suggests that other small mathematical or physiological objects could be simulated with the same approach.

Comparison of visual grading and free-response ROC analyses for assessment of image processing algorithms in digital mammography

OBJECTIVE To compare two methods for assessment of image-processing algorithms in digital mammography: free-response receiver operating characteristic (FROC) for the specific task of microcalcification detection and visual grading analysis (VGA). METHODS The FROC study was conducted prior to the VGA study reported here. 200 raw data files of low breast density (Breast Imaging-Reporting and Data System I-II) mammograms (Novation DR, Siemens, Germany)-100 of which abnormal-were processed by four image-processing algorithms: Raffaello (IMS, Bologna, Italy), Sigmoid (Sectra, Linköping, Sweden), and OpView v. 2 and v. 1 (Siemens, Erlangen, Germany). Four radiologists assessed the mammograms for the detection of microcalcifications. 8 months after the FROC study, a subset (200) of the 800 images was reinterpreted by the same radiologists, using the VGA methodology in a side-by-side approach. The VGA grading was based on noise, saturation, contrast, sharpness and confidence with the image in terms of normal structures. Ordinal logistic regression was applied; OpView v. 1 was the reference processing algorithm. RESULTS In the FROC study all algorithms performed better than OpView v. 1. From the current VGA study and for confidence with the image, Sigmoid and Raffaello were significantly worse (p<0.001) than OpView v. 1; OpView v. 2 was significantly better (p=0.01). For the image quality criteria, results were mixed; Raffaello and Sigmoid for example were better than OpView v. 1 for sharpness and contrast (although not always significantly). CONCLUSION VGA and FROC discordant results should be attributed to the different clinical task addressed. ADVANCES IN KNOWLEDGE The method to use for image-processing assessment depends on the clinical task tested.

A novel platform to simplify human observer performance experiments in clinical reading environments

Human observer performance experiments (HOPE) are frequently carried out in controlled environments in order to maximize the influence of the performance parameter under study. As an example, the amount of ambient reading variables can be kept as low as possible during HOPE. This is contrasting with the dynamic nature of a clinical reading environment that may therefore be suboptimal for the majority of the experiments. The aim of current work was to extend our previously developed software platform Sara² to cope with the influences of the reading environment on HOPE experiments. Generic modules for ROC, LROC, FROC, MAFC and visual grading analysis/image quality criteria (VGA/IQC) experiments were developed for 2D and 3D input images. Additional modules were included in the platform for finding unexpected interruptions due to clinical emergencies by means of idle time and for mouse trajectory monitoring. Also a generic approach towards the inclusion of reading questionnaires and a RFID enabled secured login system was added. Next, we created a sensor network consisting of off-the-shelf components which continuously monitor ambient reading conditions like: temperature, ambient lighting, humidity, ambient noise levels and observer reading distance. These measured parameters can be synchronized with the reading findings. Finally we included a link to incorporate the use of specialized 3rd party PACS viewers in our software framework. Using the proposed software and hardware solution, we could simplify the setup and the performing of HOPE in clinical reading environments and we can now properly control our reading experiments.

A quantitative method for evaluating the detectability of lesions in digital mammography

This study presents a quantitative method for evaluating the detectability of microcalcifications in digital mammography. Four hundred and twenty microcalcifications (with various morphology, size and contrast), simulated with a previously validated method, were used for the creation of image datasets. Lesions were inserted into 163 regions of interests of 59 selected raw digital mammograms with various anatomical backgrounds and acquired with a Siemens Novation DR. After processing, these composite images were scored by experienced radiologists, who located multiple simulated lesions and rated them under conditions of free-search. For statistical analysis, free-response receiver-operating characteristic curves are plotted; the use of jackknife free-response receiver-operating characteristic method has also been investigated. The main advantage of this methodology is that the exact number of inserted microcalcifications is well known and that the lesions are fully characterised in terms of pathology, size, morphology and peak contrast. A first application has been the evaluation of the effect of anatomical background on microcalcifications detection. Preliminary findings in this study indicate that this method may be a promising tool to evaluate factors that have an influence on the detectability of lesions, such as the clinical processing or the viewing conditions.

Clinical validation of a medical grade color monitor for chest radiology

Until recently, the specifications of medical grade monochrome LCD monitors outperformed those of color LCD monitors. New generations of color LCD monitors, however, show specifications that are in many respects similar to those of monochrome monitors typically used in diagnostic workstations. The aim of present study was to evaluate the impact of different medical grade monitors in terms of detection of simulated lung nodules in chest x-ray images. Specifically, we wanted to compare a new medical grade color monitor (Barco Coronis 6MP color) to a medical grade grayscale monitor (Barco Coronis 3MP monochrome) and a consumer color monitor (Philips 200VW 1.7MP color) by means of an observer performance experiment. Using the free-response acquisition data paradigm, seven radiologists were asked to detect and locate lung nodules (170 in total), simulated in half of the 200 chest X-ray images used in the experiment. The jackknife free-response receiver operating characteristic (JAFROC) analysis of the data showed a statistically significant difference between at least two monitors, F-value=3.77 and p-value =0.0481. The different Figure of Merit values were 0.727, 0.723 and 0.697 for the new color LCD monitor, the medical grade monitor and the consumer color monitor respectively. There was no difference between the needed reading times but there was a difference between the mean calculated Euclidian distances between the position marked by the observers and the center of the simulated nodule, indicating a better accuracy with both medical grade monitors. Present data suggests that the new generation of medical grade color monitors could be used as diagnostic workstations.

Patient Dosimetry for Mammography

We describe a method to conduct a dose study for mammography screening units using established methods and conversion factors. We establish a dose reference curve, giving a dose reference level for different sizes of breasts instead of establishing one reference level for an average size breast so we can compare the results with the achievable and acceptable dose levels from the European guidelines. We notice that, especially for smaller breasts, work is needed to get doses in Belgium below acceptable European levels.

Longitudinal tube modulation for chest and abdominal CT examinations: impact on effective patient doses calculations

Purpose: In multi-slice CT, manufacturers have implemented automatic tube current modulation (TCM) algorithms. These adjust tube current in the x-y plane (angular modulation) and/or along the z-axis (z-axis modulation) according to the size and attenuation of the scanned body part. Current methods for estimating effective dose (ED) values in CT do not account for such new developments. This study investigated the need to take TCM into account when calculating ED values, using clinical data. Methods: The effect of TCM algorithms as implemented on a GE BrightSpeed 16, a Philips Brilliance 64 and a Siemens Sensation 64 CT scanners was investigated. Here, only z-axis modulation was addressed, considering thorax and abdomen CT examinations collected from 534 adult patients. Commercially available CT dosimetry software (CT expo v.1.7) was used to compute EDTCM (ED accounting for TCM) as the sum of ED of successive slices. A two-step approach was chosen: first we estimated the relative contribution of each slice assuming a constant tube current. Next a weighted average was taken based upon the slice specific tube current value. EDTCM was than compared to patient ED estimated using average mA of all slices. Results and Conclusions: The proposed method is relatively simple and uses as input: the parameters of each protocol, a fitted polynomial function of weighting factors for each slice along the scan length and mA values of the individual patient examination. Results show that z-axis modulation does not have a strong impact on ED for the Siemens and the GE scanner (difference ranges from -4.1 to 3.3 percent); for the Philips scanner the effect was more important, (difference ranges from -8.5 to 6.9 percent), but still all median values approached zero (except for one case, where the median reached -5.6%), suggesting that ED calculation using average mA is in general a good approximation for EDTCM. Higher difference values for the Philips scanner are due to a stronger current modulation in respect to the other scanners investigated. It would be interesting to repeat the study by collecting patients in a prospective way, for whom the weight and height are know and use a dedicated patient dosimetry software to calculate the dose. If the use of TCM has a larger impact on calculated effective dose, appropriate correction factors should be used.

Contrast-detail comparison between unprocessed and processed CDMAM images

The purpose of this study is to compare Contrast Detail Curves (CDCs) of unprocessed and processed digital images. Images of a CDMAM (contrast detail for mammography) phantom had been acquired at 29 kV Tungsten-Rhodium anode-filter combination and 100 mAs; unprocessed images were subsequently processed using five clinically available image processing algorithms. Scoring of CDMAM images was then performed using human observers and automatic reading. Five observers conducted a four-alternative forced-choice experiment on a set of four images, for each processing condition. For the automatic analysis of CDMAM images the CDCOM software program was used. Contrast Detail Curves were then computed both for the human and automatic reading by fitting a psychometric curve, after applying a smoothing algorithm (Gaussian filter). For both types of readings the CDCs from processed and unprocessed images were compared. We verified the statistical significance of the difference Δ between contrast threshold measurements at 0.1 mm target size (Figure of Merit, FoM), for unprocessed and processed images and for each image processing algorithm separately. The non-parametric bootstrap method was used. No statistical significant difference is found between raw and processed images. This study shows that CDMAM images may not be appropriate in assessing image processing algorithms.