George Panayiotakis

Atlas-Based Segmentation of Degenerated Lumbar Intervertebral Discs From MR Images of the Spine

Intervertebral disc degeneration is an age-associated condition related to chronic back pain, while its consequences are responsible for over 90% of spine surgical procedures. In clinical practice, MRI is the modality of reference for diagnosing disc degeneration. In this study, we worked toward 2-D semiautomatic segmentation of both normal and degenerated lumbar intervertebral discs from T2-weighted midsagittal MR images of the spine. This task is challenged by partial volume effects and overlapping gray-level values between neighboring tissue classes. To overcome these problems three variations of atlas-based segmentation using a probabilistic atlas of the intervertebral disc were developed and their accuracies were quantitatively evaluated against manually segmented data. The best overall performance, when considering the tradeoff between segmentation accuracy and time efficiency, was accomplished by the atlas-robust-fuzzy c-means approach, which combines prior anatomical knowledge by means of a rigidly registered probabilistic disc atlas with fuzzy clustering techniques incorporating smoothness constraints. The dice similarity indexes of this method were 91.6% for normal and 87.2% for degenerated discs. Research in progress utilizes the proposed approach as part of a computer-aided diagnosis system for quantification and characterization of disc degeneration severity. Moreover, this approach could be exploited in computer-assisted spine surgery.

A wavelet-based spatially adaptive method for mammographic contrast enhancement.

A method aimed at minimizing image noise while optimizing contrast of image features is presented. The method is generic and it is based on local modification of multiscale gradient magnitude values provided by the redundant dyadic wavelet transform. Denoising is accomplished by a spatially adaptive thresholding strategy, taking into account local signal and noise standard deviation. Noise standard deviation is estimated from the background of the mammogram. Contrast enhancement is accomplished by applying a local linear mapping operator on denoised wavelet magnitude values. The operator normalizes local gradient magnitude maxima to the global maximum of the first scale magnitude subimage. Coefficient mapping is controlled by a local gain limit parameter. The processed image is derived by reconstruction from the modified wavelet coefficients. The method is demonstrated with a simulated image with added Gaussian noise, while an initial quantitative performance evaluation using 22 images from the DDSM database was performed. Enhancement was applied globally to each mammogram, using the same local gain limit value. Quantitative contrast and noise metrics were used to evaluate the quality of processed image regions containing verified lesions. Results suggest that the method offers significantly improved performance over conventional and previously reported global wavelet contrast enhancement methods. The average contrast improvement, noise amplification and contrast-to-noise ratio improvement indices were measured as 9.04, 4.86 and 3.04, respectively. In addition, in a pilot preference study, the proposed method demonstrated the highest ranking, among the methods compared. The method was implemented in C++ and integrated into a medical image visualization tool.

Breast Cancer Diagnosis: Analyzing Texture of Tissue Surrounding Microcalcifications

The current study investigates texture properties of the tissue surrounding microcalcification (MC) clusters on mammograms for breast cancer diagnosis. The case sample analyzed consists of 85 dense mammographic images, originating from the digital database for screening mammography. mammograms analyzed contain 100 subtle MC clusters (46 benign and 54 malignant). The tissue surrounding MCs is defined on original and wavelet decomposed images, based on a redundant discrete wavelet transform. Gray-level texture and wavelet coefficient texture features at three decomposition levels are extracted from surrounding tissue regions of interest (ST-ROIs). Specifically, gray-level first-order statistics, gray-level cooccurrence matrices features, and Lawspsila texture energy measures are extracted from original image ST-ROIs. Wavelet coefficient first-order statistics and wavelet coefficient cooccurrence matrices features are extracted from subimages ST-ROIs. The ability of each feature set in differentiating malignant from benign tissue is investigated using a probabilistic neural network. Classification outputs of most discriminating feature sets are combined using a majority voting rule. The proposed combined scheme achieved an area under receiver operating characteristic curve (Az) of 0.989. Results suggest that MCspsila ST texture analysis can contribute to computer-aided diagnosis of breast cancer.

A multiple projection method for digital tomosynthesis

A new method of optimized efficiency for the retrospective reconstruction of tomograms is presented. The method has been developed for use with isocentric fluoroscopic units and is capable of performing digital tomosynthesis of anatomical planes of user selected orientation and distance from the isocenter. Optimization of efficiency has been achieved by segmenting the reconstruction process into discrete transformations that are specific to groups of pixels, rather than performing pixel by pixel operations. These involve a number of projections of the acquired image matrices as well as parallel translations and summing. Application of this method has resulted in a significant reduction of computing time. The proposed algorithm has been experimentally tested on a radiotherapy simulator unit with the use of a phantom and the obtained results are reported and discussed.

Modeling granular phosphor screens by Monte Carlo methods

The intrinsic phosphor properties are of significant importance for the performance of phosphor screens used in medical imaging systems. In previous analytical-theoretical and Monte Carlo studies on granular phosphor materials, values of optical properties, and light interaction cross sections were found by fitting to experimental data. These values were then employed for the assessment of phosphor screen imaging performance. However, it was found that, depending on the experimental technique and fitting methodology, the optical parameters of a specific phosphor material varied within a wide range of values, i.e., variations of light scattering with respect to light absorption coefficients were often observed for the same phosphor material. In this study, x-ray and light transport within granular phosphor materials was studied by developing a computational model using Monte Carlo methods. The model was based on the intrinsic physical characteristics of the phosphor. Input values required to feed the model can be easily obtained from tabulated data. The complex refractive index was introduced and microscopic probabilities for light interactions were produced, using Mie scattering theory. Model validation was carried out by comparing model results on x-ray and light parameters (x-ray absorption, statistical fluctuations in the x-ray to light conversion process, number of emitted light photons, output light spatial distribution) with previous published experimental data on Gd2O2S: Tb phosphor material (Kodak Min-R screen). Results showed the dependence of the modulation transfer function (MTF) on phosphor grain size and material packing density. It was predicted that granular Gd2O2S: Tb screens of high packing density and small grain size may exhibit considerably better resolution and light emission properties than the conventional Gd2O2S: Tb screens, under similar conditions (x-ray incident energy, screen thickness).

Radiation exposure of patients and coronary arteries in the stent era: A prospective study

Previous studies have investigated the radiation dose to doctors and patients during coronary angiography and angioplasty, but most of them were retrospective, conducted in the prestent era, and results have not been consistent. Effective dose of 57 patients undergoing coronary angiography and/or angioplasty was assessed by using a dose‐area product (DAP) to effective dose conversion factor. Radiation exposure risks to patients were then calculated for each procedure. Thermoluminescent dosimeters, mounted on a specially designed catheter that was advanced to the left or right sinus of Valsalva, were used to measure the dose received by the coronary arteries. Mean effective dose received by patients were 5.0 ± 0.5 mSv for coronary angiography, 6.6 ± 1.0 mSv for angioplasty, 10.2 ± 1.5 mSv for angioplasty followed by stent implantation, 13.6 ± 2.5 mSv for angiography followed by ad hoc angioplasty, and 16.7 ± 2.8 mSv for angiography followed by ad hoc angioplasty and stent implantation. Patient risk of developing cancer after each procedure was 0.025%, 0.033%, 0.051%, 0.068%, and 0.084%, respectively. Corresponding mean coronary irradiation doses were 24 ± 2.5, 31.0 ± 3.6, 43.6 ± 7.2, 55.0 ± 7.5, and 64.7 ± 5.6 mGy, respectively. A linear relationship of the DAP and the dose at the coronary arteries was found: DAP = 1,273 (cm2) × coronary dose (mGy). Radiation exposure to coronary arteries and associated risk to patients are relatively low, even following complicated, multivessel angioplasty with stent implantation. Our method can be used for calculation of radiation risk to patients and radiation dose to coronary arteries by using external dosimeters. Cathet. Cardiovasc. Intervent. 51:259–264, 2000.

Patient and staff dosimetry in vertebroplasty.

Study Design. Eleven vertebroplasty operations were studied in terms of radiation dose. Objective. Doses to patients and staff associated with vertebroplasty were measured. Occupational doses were compared with the annual dose limits, and the effectiveness of the used radiation protection means was estimated. Patient dose was estimated by means of both surface and effective dose, and the radiation-induced risk was evaluated. Summary of Background Data. Vertebroplasty is a recent minimally invasive technique for the restoration of vertebral body fractures. It involves fluoroscopic exposure, and so, it demands dose measurements for both patient and staff exposed to radiation. Methods. Thermoluminescent dosimeters (TLDs) were placed on the medical personnel and the effective dose was derived. Slow films were placed to patients’ skin to measure entrance surface dose. Furthermore, a Rando phantom loaded with TLDs was irradiated under conditions simulating vertebroplasty, in order to estimate effective dose to the patient. Results. Mean fluoroscopy time was 27.7 minutes. Patient’s mean skin dose was 688 mGy, while effective dose was calculated to be 34.45 mGy. It was estimated that the primary operator can perform about 150 vertebroplasty operations annually without exceeding the annual dose constraints, whereas occupational dose can be reduced by 76% using mobile shielding. Conclusions. Measures have to be taken to reduce patient’s skin dose, which, in extreme cases, may be close to deterministic effects threshold. The highest dose rates, recorded during the procedure, were found for primary operator’s hands and chest when no shielding was used.

Adult patient radiation doses from non-cardiac CT examinations: a review of published results

OBJECTIVES CT is a valuable tool in diagnostic radiology but it is also associated with higher patient radiation doses compared with planar radiography. The aim of this article is to review patient dose for the most common types of CT examinations reported during the past 19 years. METHODS Reported dosimetric quantities were compared with the European diagnostic reference levels (DRLs). Effective doses were assessed with respect to the publication year and scanner technology (i.e. single-slice vs multislice). RESULTS Considerable variation of reported values among studies was attributed to variations in both examination protocol and scanner design. Median weighted CT dose index (CTDI(w)) and dose length product (DLP) are below the proposed DRLs; however, for individual studies the DRLs are exceeded. Median reported effective doses for the most frequent CT examinations were: head, 1.9 mSv (0.3-8.2 mSv); chest, 7.5 mSv (0.3-26.0 mSv); abdomen, 7.9 mSv (1.4-31.2 mSv); and pelvis, 7.6 mSv (2.5-36.5 mSv). CONCLUSION The introduction of mechanisms for dose reduction resulted in significantly lower patient effective doses for CT examinations of the head, chest and abdomen reported by studies published after 1995. Owing to the limited number of studies reporting patient doses for multislice CT examinations the statistical power to detect differences with single-slice scanners is not yet adequate.

Experimental and Theoretical Evaluation of a High Resolution CMOS Based Detector Under X-Ray Imaging Conditions

Fundamental imaging performance in terms of Modulation Transfer Function (MTF), Noise Power Spectrum (NPS) and Detective Quantum Efficiency (DQE) was investigated for a high resolution CMOS based imaging sensor. The device consists of a 33.91 mg/cm2 Gd2O2S:Tb scintillator screen, placed in direct contact with a CMOS photodiode array. The CMOS photodiode array, featuring 1200×1600 pixels with a pixel pitch of 22.5 μm, was used as an optical photon detector. In addition to the conventional frequency dependent parameters characterizing image quality, image information content was assessed through the application of information capacity (IC). The MTF was measured using the slanted-edge method to avoid aliasing while the Normalized NPS (NNPS) was determined by two-dimensional (2D) Fourier transforming of uniformly exposed images. Both measurements were performed under the representative radiation quality (RQA) settings, RQA-5 (70 kVp digital-radiography) and RQA-M2 (28 kVp digital-mammography) recommended by the International Electrotechnical Commission Reports 62220-1 and 62220-1-2 respectively. The DQE was assessed from the measured MTF, NPS and the direct entrance surface air-Kerma (ESAK) obtained from X-ray spectra measurement with a portable cadmium telluride (CdTe) detector. The ESAK values ranged between 11-87 μGy for RQA-5 and 6-40 μGy for RQA-M2. Additionally the output electrons per X-ray photon of the detector and its signal transfer characteristics were assessed via an analytical model, within the framework of the linear cascaded systems (LCS) theory. It was found that the detector response function was linear for the exposure ranges under investigation. Additionally our results showed that for the same RQA quality the output electrons per X-ray photon, as well as the measured and analytically predicted MTF, were not significantly affected by ESAK. MTF and DQE where found better compared to previously published data for other CCD and CMOS sensors, while the NNPS appeared to be comparable in the frequency range under investigation (0-10 cycles/mm).

Luminescence efficiency of Gd/sub 2/SiO/sub 5/:Ce scintillator under X-ray excitation

Gd/sub 2/SiO/sub 5/:Ce (GSO:Ce) is a high-Z, nonhygroscopic fast emitting scintillator used in detectors of positron emission tomography (PET) systems. The purpose of this study was to evaluate the luminescence efficiency of GSO:Ce scintillator under irradiation conditions employed in X-ray medical imaging, since findings may be of value in novel applications such as computed tomography breast imaging or modern fast image producing X-ray computed tomography. To this aim the absolute luminescence efficiency (emitted light flux over X-ray exposure), the spectral matching factor (compatibility to optical detectors), and the effective efficiency (the product of absolute efficiency and spectral compatibility) were determined for X-ray tube voltages ranging from 20 to 140 kV. The efficiency of GSO:Ce was found to increase with increasing X-ray tube voltage, while the GSO:Ce spectrum, peaking at 440 and 490 nm, was found compatible to most optical detectors (photodiodes, photocathodes, and charge-coupled devices).