Three-dimensional simulation of lung nodules for paediatric multidetector array CT

Size: px
Start display at page:

Download "Three-dimensional simulation of lung nodules for paediatric multidetector array CT"

Transcription

1 Te Britis Journal of Radiology, 82 (2009), Tree-dimensional simulation of lung nodules for paediatric multidetector array CT 1,2 X LI, MS, 1,2,3 E SAMEI, PD, 4 D M DELONG, PD, 5 R P JONES, BA, 6 A M GACA, MD, 6 C L HOLLINGSWORTH, MD, 6 C M MAXFIELD, MD, 6 C W T CARRICO, MD and 1,6 D P FRUSH, MD 1 Medical Pysics Graduate Program, 2 Carl E. Ravin Advanced Imaging Laboratories, Department of Radiology, 3 Departments of Pysics and Biomedical Engineering, 4 Department of Biostatistics and Bioinformatics, 5 Scool of Medicine and 6 Division of Pediatric Radiology, Department of Radiology, Duke University, Duram, NC, 27705, USA ABSTRACT. Te purpose of tis study was to develop and validate a tecnique for tree-dimensional (3D) modelling of small lung nodules on paediatric multidetector array computed tomograpy (MDCT) images. Clinical images were selected from 21 patients (,18 years old) wo underwent MDCT examinations. Sixteen of te patients ad one or more real lung nodules wit diameters between 2.5 and 6 mm. A matematical simulation tecnique was developed to emulate te 3D caracteristics of te real nodules. To validate tis tecnique, MDCT images of 34 real nodules and 55 simulated nodules were randomised and rated independently by four experienced paediatric radiologists on a continuous scale of appearance between 0 (definitely not real) and 100 (definitely real). Receiver operating caracteristic (ROC) analysis, t-test, and equivalence test were performed to assess te radiologists ability to distinguis between simulated and real nodules. Te two types of nodules were also compared in terms of measured sape and contrast profile irregularities. Te areas under te ROC curves were 0.59, 0.60, 0.40, and 0.63 for te four observers. Mean score differences between simulated and real nodules were 28, 211, 13, and 24 for te four observers wit p-values of 0.17, 0.06, 0.17, and 0.26, respectively. Te simulated and real nodules were perceptually equivalent and ad comparable sape and contrast profile irregularities. In conclusion, matematical simulation is a feasible tecnique for creating realistic small lung nodules on paediatric MDCT images. Received 14 Marc 2008 Revised 15 May 2008 Accepted 18 July 2008 DOI: /bjr/ Te Britis Institute of Radiology Detection of even one small lung nodule can ave profound implications in te prognosis and treatment of paediatric cancer. Owing to te superior resolutions of modern multidetector array computed tomograpy (MDCT), CT of te cest is often standard for paediatric cancer staging and surveillance. Wen compared wit oter imaging modalities, owever, CT delivers a iger radiation dose to te patients [1], and cildren are especially sensitive to te detrimental effect of radiation. Terefore, care must be taken to optimise radiation dose wile maintaining diagnostic accuracy. Te influence of dose reduction on lung nodule detection as been studied in adult patients by performing repeated CT scans on te same subjects at different tube current levels [2, 3]. Suc an approac is not suitable for cildren. An alternative approac is to develop tecniques to simulate lung nodules for paediatric cest MDCT. Tree types of tecniques ave been developed in te past to simulate lung nodules. Type I tecnique adds syntetic nodules into a pantom [4, 5] and as te obvious disadvantage of using unrealistic backgrounds, making it difficult to extrapolate researc results to patients. Type II tecnique digitally inserts nodules segmented from clinical images into clinical images [6, Address correspondence to: E Samei, Carl E. Ravin Advanced Imaging Laboratories, Duke University, Duram, NC 27705, USA. samei@duke.edu 7], and tus may be optimal in terms of image and nodule realism. Te occurrence of solitary small lung nodules in te paediatric population is low, owever, limiting te available nodule morpologies and making repeated use problematic. Type III tecnique adds computer-generated nodule-like objects to clinical images [8 11]. It offers te maximum flexibility and is also capable of acieving a ig degree of realism, as demonstrated by two recent studies [10, 11]. Te nodule caracteristics modelled in tese two studies, owever, were typical of nodules larger tan tose of interest in paediatric MDCT (3 5 mm); lung nodules larger tan 5 mm are rarely missed by radiologists in paediatric MDCT. Te purpose of our investigation was to develop and validate a type III tecnique for modelling small lung nodules in tree dimensions on paediatric MDCT images, and to provide a framework by wic focal lesions in oter organs can be simulated for dose and tecnique optimisation researc in MDCT. Metods and materials Our institutional review board determined tat te study was in compliance wit te Healt Insurance Portability and Accountability Act, and did not require informed consent. Te Britis Journal of Radiology, May

2 X Li, E Samei, D M DeLong et al Image selection Images of 21 patients were retrospectively selected from our clinical (teacing) database to represent infants or cildren up to 18 years old wo ad undergone unenanced MDCT (LigtSpeed 16, LigtSpeed QX/i, or LigtSpeed VCT; GE Healtcare, Waukesa, WI) examinations. All patient identifiers were removed from te images prior to te study. 16 of te patients ad sarcomas or Wilms tumour, and nodules were presumed to represent metastatic disease. Te images were acquired using tube potentials of kvp, tube currents of ma, elical pitces of 0.75 or 1.375, gantry rotation periods of s, slice ticknesses of 5 mm, reconstruction intervals of 2.5 mm or 5 mm, and in-plane resolutions of mm. Tese imaging parameters were typical of te current clinical protocols for paediatric MDCT at our institution. 34 small (real) lung nodules were identified by a paediatric radiologist wo ad 15 years of experience wit paediatric CT. Nodule caracterisation Te real nodules in our study ad diameters between 2.5 and 6 mm and were clearly visible on 1 4 contiguous MDCT slices. Nodule size and contrast varied across slices; frequently, a central slice wit maximum nodule size and contrast was accompanied by adjacent slices wit reduced sizes and contrasts (Figure 1). Altoug small nodules tended to appear round, a variety of sapes existed tat deviated clearly from a perfect spere (Figure 2). Te sape of a nodule varied from slice to slice, but strong resemblance in sape was often observed across slices (Figure 1). Te margins of nodules varied from well-defined margins to diffused/azy margins (Figure 2). Nodule simulation metod We developed a tecnique to simulate lung nodules tat ave realistic caracteristics. A 3D nodule was modelled as multiple 2D masks on sequential MDCT slices. All computer codes were written in an interactive program (Matlab, R2007a; Matworks, Natick, MA). For a given 3D nodule, te radii and peak contrasts on contiguous CT slices can vary, depending on te vertical Figure 2. Variety of nodule appearances as exemplified by 18 real nodules in our study. Only te central slice of eac nodule is sown. All regions of interest are 20 mm in size wit nodules located centrally. Figure 1. Tree-dimensional caracteristics of small lung nodules on paediatric MDCT images as exemplified by a nodule in our study. Te nodule was clearly visible on tree contiguous CT slices wit slice tickness of 5 mm and reconstruction interval of 2.5 mm. Te regions of interest are 30 mm in size wit te nodule located centrally. 402 Te Britis Journal of Radiology, May 2009

3 Tree-dimensional simulation of lung nodules for paediatric MDCT (z-direction) offsets of te nodule centre relative to te CT slices (Figure 3). On eac CT slice, te 2D mask can be defined by a contrast-profile equation proposed by Samei et al. [12] and reformulated by Burgess et al. [13] as cr ðþ~c 1{ r 2 n ð1þ R were R and C are te radius and peak contrast of te 2D mask, and te exponent n is a positive number inversely related to te steepness of te contrast profile, reflecting edge caracteristics. Due to its radial symmetry, Equation (1) defines perfectly circular masks and as been adopted by Hoe et al [14] to model small liver lesions on paediatric MDCT images. To reflect te variability of real nodule sapes, we introduced radial asymmetry by defining radius as a function of polar angle and te exponent as a function of radius, i.e. Figure 3. (a) For a 3D sperical nodule wit diameter 2R o of 6 mm, te radii on a 5-mm tick CT slice are 3 mm, 3 mm and 2.83 mm for te vertical (z-direction) locations 1, 2, and 3, respectively. Assuming te peak contrast of te nodule to be proportional to its diameter subtended by te CT slice, and defining its peak contrast at vertical location 1 as C o, its peak contrasts at vertical locations 2 and 3 are 4CO 5 and 2CO 5, respectively. Wen te reconstruction interval is canged from (b) alf of te slice tickness to (c) one slice tickness, te presence and appearance of a nodule on contiguous CT slices also cange. cr, ð Þ~C 1{ r! 2 nr ð Þ R nr ð Þ~n R R were R is a predefined nodule-sape function, rescaled so tat its average across all polar angles equals R, te nodule radius on current CT slice obtained from initial geometric calculation. We cose n(r ) as a linearly increasing function of R to improve te smootness of te nodule border; te rougness induced by increasing nodule size from one polar angle to anoter is ameliorated by a corresponding decrease in te steepness of te nodule contrast profile. Because te steepness of te contrast profile of a sperical nodule can be sown to increase wit te fraction of its diameter contained by a CT slice, f, we assumed a linear relationsip between n and f as ð2þ n~ n 1=6{n 1 1=6{1 ðf {1Þzn 1 ð3þ were n 1 and n 1/6 are exponents corresponding to conditions in wic all or one-sixt of te nodule diameter is contained by a CT slice, respectively. To determine n 1, Equation (1) was used to simulate sperical nodules tat are fully enclosed by a CT slice wit a sequence of n 1 values between 1.0 and 2.0. An experienced paediatric radiologist examined te nodule images and determined tat n yielded te most realistic appearances. Subsequently, Equations (1) and (3) were used to create 3D nodules visible on multiple contiguous CT slices wit a sequence of n 1/6 values between 2.0 and 2.6 and an n 1 value of 2.0. Te 3D nodule images were examined by te same radiologist, wo determined tat n 1/ yielded te most realistic appearances. n and n 1/ were used to calculate n in our subsequent simulations. Because sape variability contributes greatly to te difficulties associated wit nodule detection, any realistic nodule simulation must reflect te variety of real nodule sapes. We developed an algoritm tat generated a nodule sape function R from 12 manually specified base radii at evenly spaced polar angles wit linearly interpolated values in between. A nodule-sape function was first designed for te central CT slice containing te nodule centre. Sligt variations were ten added to four of te base radii, 90 degrees apart, to create nodule-sape functions for te inferior and superior CT slices. Using tis metod, a library of 60 sets of nodule-sape functions was created. Finally, we modelled nodules wit diffused edges by adding a second component to Equation (2) as 0 cr, ð Þ~aC@ 1{ r R ðþ 1 0! 2 1 ð A zð1{aþc@ 1{ r n R ð1þ R ðþ 2 Þ! 2 1 ð A n R ð2þ Þ ð4þ Te Britis Journal of Radiology, May

4 X Li, E Samei, D M DeLong et al were n R ~n ðþ 1 R 1 n R 2 ðþ ~ n c ð Þ R ð1þ ð2þ R R ð2þ ð5þ radial distance for six evenly divided angular bins. Te contrast profiles were ten normalised to te peak contrast and fitted to te normalised version of Equation (1) as c ðþ~ r 1{ r 2 n ð6þ R R ð1þ ~ R b R ð2þ ~R Here, R ðþ 1 and R ðþ 2 are two nodule-sape functions, rescaled so tat teir averages equal R (1) and R (2), respectively. Peak contrast C is sared between te two components in a a:(12a) ratio. Relative to te first component, te exponent of te second component is reduced c-fold, wile te average radius of te second component is increased b-fold. Bot C and n were derived from te radius of te first component R (1). Simulations were performed for sequences of a, b, and c over te ranges of , , and , respectively. We found tat a 5 0.4, b , and c provided te most realistic nodule appearances. Equations (4) and (5) wit a 5 0.4, b 5 1.4, and c were subsequently used to create double-component nodules, eac of wic adopted two sets of nodulesape functions, randomly sampled from te library. 60 pairs of function sets were selected by an experienced paediatric radiologist and added to te library, bringing te total number of designs to 120 (i.e. 60 single-component and 60 double-component designs). Figure 4 illustrates an example simulated nodule in 3D. Comparison between real and simulated nodules: pysical caracteristics We compared real and simulated nodules in terms of measured sape and contrast profile irregularities. For te comparison, 20 real nodules wit diameters of 4 6 mm were selected, excluding nodules tat were too small to demonstrate contrast profiles or too close to vessels to allow accurate segmentation. Te largest (and brigtest) 2D mask of eac nodule was segmented wit a metod described by Hoe et al [14]. Using te maximum pixel value as peak contrast and te corresponding pixel as origin, we derived six contrast profiles from eac mask by plotting average pixel value as a function of were c9(r) 5 c(r)/c is normalised contrast. For eac nodule, six R9 and six n9 values were extracted from te fits. In a few cases were Equation (6) did not provide a good fit, we furter attempted double-component fits. Te visual span of a contrast profile was measured in terms of R 0 0:05, te distance at wic normalised contrast c9(r) equals 0.05, i.e. qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi p R 0:05 ~R 1{ ffiffiffiffiffiffiffiffiffi n 0:05 Accordingly, te visual steepness of a contrast profile was measured by pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi g~ R p 0:05 1{ n 0:05 ~ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi p R 0:5 1{ n 0:5 Te dimensionless quantity g is a slowly increasing function of n9, and tus is inversely related to te steepness of te contrast profile. Furtermore, to describe sape and contrast profile irregularities, coefficients of intranodule variation were calculated for te six R 0 0:05 and six g values of eac nodule as cov(r 0:05 )~ s(r 0:05) 100% R 0:05 covðgþ~ sg ðþ 100% g To compare te simulated masks wit real nodule masks, te 120 designs in our library were used to simulate 60 single-component and 60 double-component 2D masks on a zero background wit nodule centres assuming random vertical (z-direction) locations inside te background. Radius R o and peak contrast C o were randomly sampled from ranges determined to matc tose of te real masks. Te same fitting metod was applied to te simulated masks to extract R9 and n9 values for te calculation of cov R 0:05 and cov(g). ð7þ ð8þ ð9þ Figure 4. An example simulated (single-component) nodule in tree dimensions. Te nodule was simulated wit 2R o 5 5mm and C o HU on five contiguous CT slices wit slice tickness of 5 mm and reconstruction interval of 2.5 mm. All regions of interest are 30 mm in size wit te nodule located centrally. 404 Te Britis Journal of Radiology, May 2009

5 Tree-dimensional simulation of lung nodules for paediatric MDCT Two-sample t-tests were performed to compare sape and contrast profile irregularities of simulated and real masks. Commercial software (JMP, version 6; SAS Institute, Cary, NC) was used to perform te tests at a significance level of Comparison between real and simulated nodules: observer experiment Te quality and clinical utility of simulated nodules were evaluated in terms of te ability of radiologists to distinguis between simulated and real nodules. Sample preparation 16 of te patients ad one or more lung nodules. Images wit more tan one nodule were rendered multiple times, and eac time, all but one nodule was digitally removed using a metod described in [14]. Tis process provided 34 image sets, eac containing 4 6 contiguous CT slices wit a single real nodule visible on 1 4 slices. Anoter 55 image sets of 4 6 slices, clear of nodules, were identified from images of all 21 patients and used as backgrounds for nodule simulation. A single simulated 3D nodule wit simulation diameter 2R o of 3 7 mm (corresponding to visual diameter of mm) was inserted into eac of te 55 backgrounds at randomly selected locations in te pulmonary parencyma, not immediately adjacent to te cest wall. A peak contrast C o range of Hounsfield units (HU) was used, matcing te contrasts of real nodules. Figure 5 illustrates examples of simulated nodules in tis sample. Te ratio of single-component to double-component nodules was 5:4, but tey were oterwise grouped and analysed togeter to enlarge te scale of te study. Anoter four image sets of 4 6 slices tat were clear of nodules were combined wit te real and simulated nodule samples, yielding a total of 93 image sets for te observer experiment. Observer experiment Te 93 image sets were randomised and displayed as independent series on a GE Advantage Workstation (GE Healtcare, Waukesa, WI) set to te standard lung window (window centre 2500 HU, window widt 1500 HU) in a controlled reviewing environment. Four experienced paediatric radiologists (wit 3 12 years of experience wit paediatric CT) independently rated te image sets for nodule appearance on a continuous scale between 0 (definitely not real) and 100 (definitely real). All observers were told tat tere was zero or one nodule in eac image set; tey were not told ow many image sets ad no nodule or ow many of te nodules were real. Te observers were encouraged to use te full scale for assigning scores. Tey ad no time limit in rating eac image set, but tey were not allowed to return to an image set once rated. To minimise te effect of te learning period, te image sets were sown in numerical order to two observers, and in reverse numerical order to te oter two. Eac observer was first asked about te presence of a nodule, and if a nodule was found, te observer was ten asked to point to te location before rating it on a score seet. Data analysis Te detection rates and te rates of false (spurious nodule) detection were computed for te real and simulated nodules. As te goal of te experiment was to evaluate te appearances of te known nodules in te images, scores given to spurious (non-existing) nodules were discarded from furter analysis. Receiver operating caracteristic (ROC) analysis was performed to test eac observer s ability to distinguis between simulated and real nodules. ROC software (ROCKIT, version 1.1 B 2; Carles E. Metz, University of Cicago, IL) was used. Simulated nodules were defined as positive cases; true-positive fraction (TPF) was te likeliood of a simulated nodule being identified as simulated, wereas false-positive fraction (FPF) was te likeliood of a real nodule being identified as simulated. As our objective was to determine weter simulated nodules could emulate te appearances of real ones, our ROC analysis was targeted to determine ow close te areas under te ROC curves were to te cance value of 0.5. In tat way, our objective in tis use of te ROC analysis, following an earlier study [15], was somewat different from tat in te standard use of te ROC metodology, in wic one aims to assess te difference (and not te similarity) between two alternatives (e.g. disease present versus disease absent). Two-sample t-test was also performed for eac observer to test te null ypotesis tat te difference in population mean score between simulated and real nodules (Dm 5 Dm simulated 2 Dm real ) was zero. Commercial software (JMP, version 6) was used to perform te t-tests at a significance level of Finally, equivalence tests were performed using confidence intervals of Dm. We defined te zone of equivalence as,s(x real ). observer, te standard deviation of real nodule scores averaged across observers. As a grapical depiction of te quality of simulated nodules, istograms of te frequencies of score assignments for bot types of nodules were derived for eac observer. Results Pysical caracteristics Among te 120 contrast profiles derived from te 20 real nodule masks, 105 profiles (87.5%) fitted well to Equation 6 as exemplified by Figure 6a,b, 5 profiles (4.2%) ad an apparent two-component appearance resembling tat in Figure 6c, and 10 profiles (8.3%) ad apparent flat tops resembling tat in Figure 6d. For te 120 simulated nodule masks, te goodness-of-fit resembled tat in Figure 6a,b. 10 profiles (1.4%) ad sligt two-component appearance similar to but not as obvious as tat in Figure 6c. Equation 6 was found to provide generally reasonable fit for bot real and simulated nodule masks. Wen comparing scatter plots of visual steepness versus visual radius (Figure 7), an excellent matc was found between simulated and real masks except for te lower rigt corner (R 0 0:05., 2.75 mm or visual diameter., 5.5 mm) of te real mask scatter plot, were te sarp steepness (small g values) of some contrast profiles was not captured by te simulation. Altoug marginally significant differences in sape irregularity Te Britis Journal of Radiology, May

6 X Li, E Samei, D M DeLong et al Figure 5. Examples of simulated nodules used in te observer experiment: te central slices of (a) 18 single-component and (b) 18 double-component nodules. All regions of interest are 20 mm in size wit nodules located centrally. and significant differences in contrast profile irregularity were found between te two types of simulated masks and te real ones (Table 1), te tree types of masks ad comparable sape irregularities, and simulated doublecomponent masks ad contrast profile irregularities comparable to tose of te real nodules. Observer experiment Te real and simulated nodules ad similar and ig detection rates, 86.8% and 78.2%, respectively, wit similar rates of false (spurious nodule) detection, 2.2% and 1.4%, respectively (Table 2). At tese similar detection rates, te observers could not reliably distinguis simulated nodules from real ones. All ROC curves (Figure 8) were close to te cance line. No significant difference in population mean score was found between simulated and real nodules for any observer (Table 2). Altoug tree of te observers assigned sligtly lower scores to simulated nodules, te lower limits of all confidence intervals of Dm (m simulated 2 m real ) were well contained witin te zone of equivalence (Figure 9), indicating tat te two types of nodules were perceptually equivalent witin te normal variation of real nodule appearances. Te istograms of score assignments (Figure 10) sowed tat te distributions of scores were similar for te two types of nodules for all observers. Discussion Computer modelling of lung nodules as been an important area of investigation in CT [8 11]. Most studies ave not aimed to emulate te features of real lung 406 Te Britis Journal of Radiology, May 2009

7 Tree-dimensional simulation of lung nodules for paediatric MDCT Figure 6. Sample results of curve-fitting performed for normalised contrast profiles of 20 real nodule masks. Contrast profiles are represented by symbols, and teir fits to Equation 6 are represented by solid lines. All R9, R9 (1), and R9 (2) values are in mm. (a) An excellent fit wit small residues. (b) Anoter excellent fit, were te general trend of te contrast profile was well described by Equation 6, despite te relatively large residues. Tis example fit yielded large R9 and n9 values. Te corresponding R 0 0:05 and g values were 3.3 mm and 2.1, respectively, more suitable for describing te visual radius and steepness of tis contrast profile tan te original R9 and n9 values. (c,d) Two examples of not excellent but acceptable fits to Equation 6. Excellent fits were obtained, owever, wen fitting tese two contrast profiles to te double-component version of Equation 6, as demonstrated by te tick dased lines. nodules, as validated by observer studies of nodule appearances. Two recent studies ave offered an exception [10, 11]; owever, te modelled nodule caracteristics were typical of large (.5 mm) nodules, a range tat is larger tan wat would be subtle enoug for most tecnique-optimisation studies in MDCT, particularly in paediatric applications [7, 16]. Most recently, in an investigation focused on te paediatric population [7], a single real nodule extracted from a paediatric cest MDCT image was digitally resized and inserted back into te image to study te effects of added noise on nodule detection. Te nodule was presented and reviewed on a single slice only, was not validated as representing a standard nodule, and did not vary in terms of sape or contrast, making te nodule detection study less reflective of te actual clinical paradigm in wic nodules wit variable appearances must be detected. In tis work, we developed and validated a tecnique for computer modelling of small lung nodules in 3D. Our study focused on paediatric MDCT because of te fact tat te paediatric population is more sensitive to radiation-induced cancer, making investigations of dose reduction more important. At te same time, te coice of tis patient group limited our ability to investigate lower-dose tecniques systematically due to etical concerns wit multiple MDCT scans on te same patients wit tecnique variations. Our work also represents, to our knowledge, te first detailed caracterisation of small lung nodules in terms of sape and contrast profile irregularities. Te contrast profiles of most small lung nodules on paediatric MDCT images fitted well to an exponential equation originally proposed for nodules in cest radiograpy. In tose cases were fitting to tis original equation was unsatisfactory, te double-component version could be used to acieve excellent fits. Tese findings supported our application of tis equation to simulate small lung nodules. Wit te introduction of radial asymmetry, te simulation nearly matced te sape irregularities of real nodules. Even toug te objective of our study was to compare real and simulated nodules in terms of teir appearances, not of teir detection rates, it is interesting to Te Britis Journal of Radiology, May

8 X Li, E Samei, D M DeLong et al Figure 7. Scatter plots of visual steepness (inversely related to g) versus visual radius R 0 0:05 for real, simulated singlecomponent (SSC) and simulated double-component (SDC) nodule masks. examine teir detection rates and te reasons for missed nodules. Te real nodules yielded a iger detection rate, wic is not surprising given te fact tat tey were pre-selected, i.e. tey ad been detected (in te sample selection process) in te first place before tey could be included in te study. Neverteless, te detection rates of te two types of nodules were similar. Terefore, te subtlety of te simulated nodules (determined by size, sape, contrast, and location) adequately resembled tat of te real nodules, and would be appropriate for studies aiming to compare nodule detection rates between different CT tecniques. Revisiting te images in wic real or simulated nodules were missed revealed tat most of te nodules tat were missed by more tan two observers were close to vessels and tus looked like parts of te vessels. Tis igligts te importance of nodule location in detection studies tat employ nodule-simulation tecniques. One limitation of our study is tat we did not fully model te contrast-profile irregularities of real nodules. We assumed a linear increase of exponent n wit radius R. As a result, contrast profile steepness (inversely related to g) of simulated nodules sowed a decreasing trend wit nodule radius (Figure 7). Te beaviours of real nodules, owever, were more complex and variable. A closer matc was found between simulated doublecomponent and real masks because te superimposition of two components of different sizes and sapes introduced more variations. Noneteless, tis pysical discrepancy did not seem to ave provided strong visual clues to our observers, as demonstrated by te results of te observer experiment. Anoter limitation is tat our 3D nodule model did not fully account for te effects of respiratory and cardiac motions, wic could furter depend on nodule locations in te lung parencyma. By adding nodules directly to reconstructed CT slices, we ave assumed tat te noise caracteristics of te images remain te same wit or witout te nodules. For te small nodule sizes modelled in our study, tis is a valid assumption, because te perturbations of te small nodules to te attenuation properties of te images are negligible. Finally, our evaluation of te tecnique was limited to te ranges of scan protocols, slice tickness, reconstruction intervals used for te paediatric MDCT patients at our institution only. Our metod can, owever, be readily adapted to simulate small lung nodules on images acquired wit oter scan and reconstruction protocols. Our simulation metod may also enable te modelling of large nodules, wic often ave flat-topped contrast profiles. A few real nodules in our study exibited suc contrast profiles, wic did not fit well to te singlecomponent contrast-profile equation (as exemplified by te solid line Figure 6d) and contributed to te lower rigt corner of te real mask scatter plot (combinations of small g wit R 0 0:05., 2.75 mm or visual diameter., 5.5 mm). However, a doublecomponent contrast-profile equation in wic a small Table 1. Summary of t-test results comparing real and simulated nodule masks in terms of sape and contrast-profile irregularities a Case Sample size Mean (±sd) Difference (95% CI) p-value Sape irregularity cov(r 0 0:05 )b Real ( 5.2) SSC ( 3.8) SDC ( 4.4) SSC Real 23.0 (25.6, 20.4) 0.02 SDC Real 22.6 (25.2, 0.1) 0.06 Contrast profile irregularity cov(g) b Real ( 2.2) SSC ( 1.2) SDC ( 3.0) SSC Real 24.9 (26.0, 23.9), SDC Real 21.8 (23.1, 20.5) 0.01 a Te null ypotesis was tat te difference in population mean sape or contrast profile irregularity between simulated singleor double-component masks and real masks was zero. b cov(r 0 0:05 ) and cov(g) are coefficients of intranodule variation in radius and steepness of contrast profile, respectively. SSC, simulated single-component; SDC, simulated double-component; sd, standard deviation; CI, confidence interval. 408 Te Britis Journal of Radiology, May 2009

9 Tree-dimensional simulation of lung nodules for paediatric MDCT Table 2. Summary of t-test results comparing real and simulated nodule scores a Observer and case Detection rate b Mean score (±d) Difference (95% CI) p-value Observer 1 Real 97.1% 59.7 ( 25.7) Simulated 81.8% 51.7 ( 24.3) Simulated real 28.0 (219.5, 3.5) 0.17 Observer 2 Real 91.2% 69.2 ( 23.4) Simulated 83.6% 58.2 ( 27.5) Simulated real (222.6, 0.7) 0.06 Observer 3 Real 76.5% 49.5 ( 38.5) Simulated 69.1% 62.1 ( 29.6) Simulated real 12.5 (25.5, 30.6) 0.17 Observer 4 Real 82.4% 83.2 ( 15.0) Simulated 78.2% 79.3 ( 13.1) Simulated real 23.9 (210.9, 3.0) 0.26 a Te null ypotesis was tat te difference in population mean score between simulated and real nodules was zero. b Te detection rates averaged across observers were 86.8% and 78.2% for te real and simulated nodules, respectively. Te rates of missed nodules, for wic no scores were given or scores were given to spurious (non-existing) nodules, equalled one minus te corresponding detection rates. Te rates of false (spurious nodule) detection averaged across observers were 2.2% and 1.4% for te real and simulated cases, respectively. Scores given to spurious nodules were not included in te analysis. sd, standard deviation; CI, confidence interval. component is subtracted from a large component can model suc contrast profiles (tick dased line in Figure 6d), tus extending our metod to simulate larger nodules, provided tat te nodule sizes are small enoug (,,1 cm) to still consider te perturbations to te noise backgrounds negligible. In conclusion, our results demonstrated tat matematical simulation is a feasible tecnique for creating images of small lung nodules tat resemble real nodules in pysical caracteristics and detection rate, and tat are Figure 8. Results of receiver operating caracteristic analysis. Simulated nodules were defined as positive cases; truepositive fraction (TPF) was te likeliood of a simulated nodule being identified as simulated, wereas false-positive fraction (FPF) was te likeliood of a real nodule being identified as simulated. Error figures refer to standard errors. Figure 9. Results of equivalence tests based on confidence intervals of te differences in population mean score between simulated and real nodules (Dm 5 m simulated 2 m real ). Te zone of equivalence was defined as,s(x real ). observer, te standard deviation of real nodule scores averaged across observers. Te Britis Journal of Radiology, May

10 X Li, E Samei, D M DeLong et al Figure 10. Histograms of score assignments for (a) observer 1, (b) observer 2, (c) observer 3, and (d) observer 4. Scores of 100 were included in te last bin of eac istogram. perceptually indistinguisable from real nodules to experienced paediatric radiologists. Te nodule simulation tecnique can find applications in several areas of researc. Wen used in conjunction wit noise (tube current reduction) simulation tecniques operating on eiter projection or reconstructed CT images [16 19], simulated 3D nodules can be used in dose-reduction studies. As opposed to te virtually impractical process of collecting nodules from paediatric patients, simulated nodules can be inserted to normal MDCT images, allowing quick generation of a large database of images wit establised trut. Our tecnique may also prove useful in te optimisation of oter CT imaging parameters and in te training and evaluation of computeraided diagnosis systems. Finally, we believe tat te simulation tecnique can be extended to model nodules and lesions in oter organs, facilitating not only researc investigations but potentially te education and training of medical professionals [20]. References 1. Brenner DJ, Hall EJ. Computed tomograpy an increasing source of radiation exposure. N Engl J Med 2007;357: Karabulut N, Toru M, Gelebek V, Gulsun M, Ariyurek OM. Comparison of low-dose and standard-dose elical CT in te evaluation of pulmonary nodules. Eur Radiol 2002;12: Naidic DP, Marsall CH, Gribbin C, Arams RS, McCauley DI. Low-dose CT of te lungs: preliminary observations. Radiology 1990;175: Ko JP, Marcus R, Bomsztyk E, Babb JS, Stefanescu C, Kaur M, et al. Effect of blood vessels on measurement of nodule volume in a cest pantom. Radiology 2006;239: Bolte H, Muller-Hulsbeck S, Riedel C, Janke T, Inan N, Heller M, et al. Ex-vivo injection tecnique for implanting solid pulmonary nodules into porcine lungs for multi-slice CT. Rofo 2004;176: Madsen MT, Berbaum KS, Ellingson AN, Tompson BH, Mullan BF, Caldwell RT. A new software tool for removing, storing, and adding abnormalities to medical images for perception researc studies. Acad Radiol 2006;13: Punwani S, Zang J, Davies W, Greenalg R, Humpries P. Paediatric CT: te effects of increasing image noise on pulmonary nodule detection. Pediatr Radiol 2008;38: Seltzer SE, Judy PF, Adams DF, Jacobson FL, Stark P, Kikinis R, et al. Spiral CT of te cest: comparison of cine and film-based viewing. Radiology 1995;197: Rusinek H, Naidic DP, McGuinness G, Leitman BS, McCauley DI, Krinsky GA, et al. Pulmonary nodule 410 Te Britis Journal of Radiology, May 2009

11 Tree-dimensional simulation of lung nodules for paediatric MDCT detection: low-dose versus conventional CT. Radiology 1998;209: Sin HO, Blietz M, Frericks B, Baus S, Savellano D, Galanski M. Insertion of virtual pulmonary nodules in CT data of te cest: development of a software tool. Eur Radiol 2006;16: Zang X, Olcott E, Raffy P, Yu N, Cui H. Simulating solid lung nodules in MDCT images for CAD evaluation: modeling, validation, and applications. In: Giger M, Karssemeijer N, editors. Proceedings of SPIE: medical imaging computer-aided diagnosis; 2007, February 20 22; San Diego, CA, USA. Bellingam, WA: International Society for Optical Engineering. 2007;6514:65140Z. 12. Samei E, Flynn MJ, Eyler WR. Simulation of subtle lung nodules in projection cest radiograpy. Radiology 1997;202: Burgess AE, Li X, Abbey CK. Nodule detection in two component noise: toward patient structure. In: Kundel HL, editor. Proceedings of SPIE: medical imaging image perception; 1997, February 26; Newport Beac, CA, USA. Bellingam, WA: International Society for Optical Engineering. 1997;3036: Hoe CL, Samei E, Frus DP, Delong DM. Simulation of liver lesions for pediatric CT. Radiology 2006;238: Saunders R, Samei E, Baker J, Delong D. Simulation of mammograpic lesions. Acad Radiol 2006;13: Li X, Samei E, DeLong DM, Jones RP, Colser JG, Frus DP. Towards assessing te diagnostic influence of dose reduction in pediatric CT: a study based on simulated lung nodules. In: Hsie J, Samei E, editors. Proceedings of SPIE: medical imaging pysics of medical imaging; 2008, February 17 22; San Diego, CA, USA. Bellingam, WA: International Society for Optical Engineering. 2008;6913:69131L. 17. Frus DP, Slack CC, Hollingswort CL, Bisset GS, Donnelly LF, Hsie J, et al. Computer-simulated radiation dose reduction for abdominal multidetector CT of pediatric patients. AJR Am J Roentgenol 2002;179: Sennst DA, Kacelriess M, Leidecker C, Scmidt B, Watzke O, Kalender WA. An extensible software-based platform for reconstruction and evaluation of CT images. Radiograpics 2004;24: Britten AJ, Crotty M, Kiremidjian H, Grundy A, Adam EJ. Te addition of computer simulated noise to investigate radiation dose and image quality in images wit spatial correlation of statistical noise: an example application to X- ray CT of te brain. Br J Radiol 2004;77: Becker GJ. Simulation and te coming transformation of medical education and training. Radiology 2007;245:7 9. Te Britis Journal of Radiology, May

Propensity score analysis with hierarchical data

Propensity score analysis with hierarchical data Section on Statistics in Epidemiology Propensity score analysis wit ierarcical data Fan Li, Alan M. Zaslavsky, Mary Bet Landrum Department of Healt Care Policy, Harvard Medical Scool 180 Longwood Avenue,

More information

Sickle Cell. Scientific Investigation

Sickle Cell. Scientific Investigation Scientific Investigation Red blood cells are oval and ave a biconcave sape, giving tem te appearance of an inner tube witout te ole. Teir sape gives tem flexibility as tey pass into small capillaries.

More information

Acknowledgments. A Specific Diagnostic Task: Lung Nodule Detection. A Specific Diagnostic Task: Chest CT Protocols. Chest CT Protocols

Acknowledgments. A Specific Diagnostic Task: Lung Nodule Detection. A Specific Diagnostic Task: Chest CT Protocols. Chest CT Protocols Personalization of Pediatric Imaging in Terms of Needed Indication-Based Quality Per Dose Acknowledgments Duke University Medical Center Ehsan Samei, PhD Donald Frush, MD Xiang Li PhD DABR Cleveland Clinic

More information

Running head: SEPARATING DECISION AND ENCODING NOISE. Separating Decision and Encoding Noise in Signal Detection Tasks

Running head: SEPARATING DECISION AND ENCODING NOISE. Separating Decision and Encoding Noise in Signal Detection Tasks Running ead: 1 Separating Decision and Encoding Noise in Signal Detection Tasks (Psycological Review, in press) Carlos Alexander Cabrera1, Zong-Lin Lu1, and Barbara Anne Doser2 Te Oio State University1

More information

White Rose Research Online URL for this paper:

White Rose Research Online URL for this paper: Tis is an autor produced version of Meta-analysis of absolute mean differences from randomised trials wit treatment-related clustering associated wit care providers. Wite Rose Researc Online URL for tis

More information

Individual differences in the fan effect and working memory capacity q

Individual differences in the fan effect and working memory capacity q Journal of Memory and Language 51 (2004) 604 622 Journal of Memory and Language www.elsevier.com/locate/jml Individual differences in te fan effect and working memory capacity q Micael F. Bunting a, *,

More information

Homophily and minority size explain perception biases in social networks

Homophily and minority size explain perception biases in social networks 3 5 6 7 8 9 3 5 Homopily and minority size explain perception biases in social networks Eun Lee,+,*, Fariba Karimi,3,+,*, Claudia Wagner,3, Hang-Hyun Jo,5,6, Markus Stromaier,7, and Mirta Galesic 8,9 Department

More information

Locomotor and feeding activity rhythms in a light-entrained diurnal rodent, Octodon degus

Locomotor and feeding activity rhythms in a light-entrained diurnal rodent, Octodon degus Locomotor and feeding activity rytms in a ligt-entrained diurnal rodent, Octodon degus R. GARCÍA-ALLEGUE, 1 P. LAX, 1 A. M. MADARIAGA, 2 AND J. A. MADRID 1 1 Department of Pysiology and Parmacology (Animal

More information

Name: Key: E = brown eye color (note that blue eye color is still represented by the letter e, but a lower case one...this is very important)

Name: Key: E = brown eye color (note that blue eye color is still represented by the letter e, but a lower case one...this is very important) MONOHYBRID CROSS v2 Name: Introduction A gamete is te egg or sperm cell tat is produced by meiosis. A gamete contains te aploid number of cromosomes (in a uman tis number is 23). In eac of tese cromosomes

More information

EXPERTISE, UNDERUSE, AND OVERUSE IN HEALTHCARE * Amitabh Chandra Harvard and the NBER. Douglas O. Staiger Dartmouth and the NBER

EXPERTISE, UNDERUSE, AND OVERUSE IN HEALTHCARE * Amitabh Chandra Harvard and the NBER. Douglas O. Staiger Dartmouth and the NBER PRELIMINARY & INCOMPLETE DO NOT CITE OR DISTRIBUTE EXPERTISE, UNDERUSE, AND OVERUSE IN HEALTHCARE * Amitab Candra Harvard and te NBER Douglas O. Staiger Dartmout and te NBER Version: Marc 27, 2011 Abstract

More information

Journal of Theoretical Biology

Journal of Theoretical Biology Journal of Teoretical Biology 269 (2011) 131 137 Contents lists available at ScienceDirect Journal of Teoretical Biology journal omepage: www.elsevier.com/locate/yjtbi Life istory and mating systems select

More information

Unbiased MMSE vs. Biased MMSE Equalizers

Unbiased MMSE vs. Biased MMSE Equalizers Tamkang Journal of Science and Engineering, Vol. 12, No. 1, pp. 45 56 (2009) 45 Unbiased MMSE vs. iased MMSE Equalizers Rainfield Y. Yen Department of Electrical Engineering, Tamkang University, Tamsui,

More information

Performance of Fractured Horizontal Wells in High-Permeability Reservoirs P. Valkó, SPE and M. J. Economides, SPE, Texas A&M University

Performance of Fractured Horizontal Wells in High-Permeability Reservoirs P. Valkó, SPE and M. J. Economides, SPE, Texas A&M University SPE 349 Performance of Fractured Horizontal Wells in Hig-Permeability Reservoirs P. Valkó, SPE and M. J. Economides, SPE, Texas A&M University Copyrigt 995, Society of Petroleum Engineers, Inc. Tis paper

More information

Correcting for Lead Time and Length Bias in Estimating the Effect of Screen Detection on Cancer Survival

Correcting for Lead Time and Length Bias in Estimating the Effect of Screen Detection on Cancer Survival American Journal of Epidemiology ª Te Autor 2008. Publised by te Jons Hopkins Bloomberg Scool of Public Healt. All rigts reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org.

More information

Each year is replete with occasions to give gifts. From

Each year is replete with occasions to give gifts. From Journal of Consumer Researc Advance Access publised December 29, 2016 Experiential Gifts Foster Stronger Social Relationsips tan Material Gifts CINDY CHAN CASSIE MOGILNER Eac year is replete wit occasions

More information

Lothian Palliative Care Guidelines patient information

Lothian Palliative Care Guidelines patient information Lotian Palliative Care Guidelines patient information Q. How will I know if te morpine is not going to work for some of my pain? A. You may still ave pain despite taking bigger doses of morpine and may

More information

Doses from pediatric CT examinations in Norway Are pediatric scan protocols developed and in daily use?

Doses from pediatric CT examinations in Norway Are pediatric scan protocols developed and in daily use? Doses from pediatric CT examinations in Norway Are pediatric scan protocols developed and in daily use? Eva Godske Friberg * Norwegian Radiation Protection Authority, P.O. Box, Østerås, Norway Abstract.

More information

arxiv: v2 [cs.ro] 31 Jul 2018

arxiv: v2 [cs.ro] 31 Jul 2018 Design and Development of Effective Transmission Mecanisms on a Tendon Driven Hand Ortosis for Stroke Patients arxiv:182.6131v2 [cs.ro] 31 Jul 218 Sangwoo Park1, Lynne Weber2, Lauri Bisop2, Joel Stein2,3

More information

Modeling H1N1 Vaccination Rates. N. Ganesh, Kennon R. Copeland, Nicholas D. Davis, National Opinion Research Center at the University of Chicago

Modeling H1N1 Vaccination Rates. N. Ganesh, Kennon R. Copeland, Nicholas D. Davis, National Opinion Research Center at the University of Chicago Modeling H1N1 Vaccination Rates N. Ganes, Kennon R. Copeland, Nicolas D. Davis, National Opinion Researc Center at e University of Cicago James A. Singleton, Tammy Santibanez, Centers for Disease Control

More information

Public Assessment Report. Scientific discussion. Mebeverine HCl Aurobindo Retard 200 mg modified release capsules, hard. (mebeverine hydrochloride)

Public Assessment Report. Scientific discussion. Mebeverine HCl Aurobindo Retard 200 mg modified release capsules, hard. (mebeverine hydrochloride) Public Assessment Report Scientific discussion ebeverine HCl Aurobindo Retard 200 mg modified release capsules, ard (mebeverine ydrocloride) NL/H/3750/001/DC Date: 18 September 2017 Tis module reflects

More information

Public Assessment Report. Scientific discussion. Orlyelle 0.02 mg/3 mg and 0.03 mg/3 mg film-coated tablets. (Ethinylestradiol/Drospirenone)

Public Assessment Report. Scientific discussion. Orlyelle 0.02 mg/3 mg and 0.03 mg/3 mg film-coated tablets. (Ethinylestradiol/Drospirenone) Public Assessment Report Scientific discussion Orlyelle 0.02 mg/3 mg and 0.03 mg/3 mg film-coated tablets (Etinylestradiol/Drospirenone) NL/H/2890/001-002/DC Date: 18 June 2014 Tis module reflects te scientific

More information

Widespread use of pure and impure placebo interventions by GPs in Germany

Widespread use of pure and impure placebo interventions by GPs in Germany Family Practice 2012; 29:79 85 doi:10.1093/fampra/cmr045 Advance Access publised on 1 August 2011 Ó Te Autor 2011. Publised by Oxford University Press. All rigts reserved. For permissions, please e-mail:

More information

Public Assessment Report. Scientific discussion. Amoxiclav Aristo 500 mg/125 mg and 875 mg/125 mg film-coated tablets

Public Assessment Report. Scientific discussion. Amoxiclav Aristo 500 mg/125 mg and 875 mg/125 mg film-coated tablets Public Assessment Report Scientific discussion C Amoxiclav Aristo 500 mg/125 mg and 875 mg/125 mg film-coated tablets (amoxicillin triydrate and potassium clavulanate) NL/H/3468/001-002/DC Date: 12 July

More information

Public Assessment Report. Scientific discussion. Ramipril Teva 1.25 mg, 2.5 mg, 5 mg and 10 mg tablets Ramipril DK/H/2130/ /DC.

Public Assessment Report. Scientific discussion. Ramipril Teva 1.25 mg, 2.5 mg, 5 mg and 10 mg tablets Ramipril DK/H/2130/ /DC. Public Assessment Report Scientific discussion Ramipril Teva 1.25 mg, 2.5 mg, 5 mg and 10 mg tablets Ramipril DK/H/2130/001-004/DC 3 April 2014 Tis module reflects te scientific discussion for te approval

More information

How to Develop CT Protocols for Children

How to Develop CT Protocols for Children How to Develop CT Protocols for Children Introduction Prior to 2001 the vast majority of CT imaging of children was conducted using the same or similar techniques used for adult imaging. In 2001, several

More information

Magnetic Resonance Imaging in Acute Hamstring Injury: Can We Provide a Return to Play Prognosis?

Magnetic Resonance Imaging in Acute Hamstring Injury: Can We Provide a Return to Play Prognosis? Sports Med (2015) 45:133 146 DOI 10.1007/s40279-014-0243-1 SYSTEMATIC REVIEW Magnetic Resonance Imaging in Acute Hamstring Injury: Can We Provide a Return to Play Prognosis? Gustaaf Reurink Elisabet G.

More information

International Journal of Health Sciences and Research ISSN:

International Journal of Health Sciences and Research   ISSN: International Journal of ealt Sciences and Researc www.ijsr.org ISSN: 2249-9571 Review rticle ombay lood Group: Distinct Descent Samarpita Pramanik Senior Tutor, Scool of Nursing, Cristian ospital erampur,

More information

Summary. Introduction. Methods

Summary. Introduction. Methods Osteoartritis and Cartilage (2001) 9, 771 778 2001 OsteoArtritis Researc Society International 1063 4584/01/080771+08 $35.00/0 doi:10.1053/joca.2001.0474, available online at ttp://www.idealibrary.com

More information

A Platoon-Level Model of Communication Flow and the Effects on Operator Performance

A Platoon-Level Model of Communication Flow and the Effects on Operator Performance A Platoon-Level Model of Communication Flow and te Effects on Operator Performance by Patricia W. Kilduff, Jennifer C. Swoboda, and Josua Katz ARL-MR-0656 November 2006 Approved for public release; distribution

More information

Spiral of Silence in Recommender Systems

Spiral of Silence in Recommender Systems Spiral of Silence in Recommender Systems Dugang Liu, Cen Lin Department of Computer Science, Xiamen University Xiamen, Cina ABSTRACT Yangua Xiao Scool of Computer Science, Fudan University Sangai, Cina

More information

Public Assessment Report. Scientific discussion. Carbidopa/Levodopa Bristol 10 mg/100 mg, 12.5 mg/50 mg, 25 mg/100 mg and 25 mg/250 mg tablets

Public Assessment Report. Scientific discussion. Carbidopa/Levodopa Bristol 10 mg/100 mg, 12.5 mg/50 mg, 25 mg/100 mg and 25 mg/250 mg tablets Public Assessment Report Scientific discussion Carbidopa/Levodopa Bristol 10 mg/100 mg, 12.5 mg/50 mg, 25 mg/100 mg and 25 mg/250 mg tablets (carbidopa/levodopa) NL/H/3044/001-004/DC Date: 25 February

More information

A VIRTUAL TRAINING SYSTEM FOR CHEST RADIOGRAM INTERPRETATIONS USING ANATOMICAL HUMAN STRUCTURES IN HIGH-RESOLUTION CT IMAGES

A VIRTUAL TRAINING SYSTEM FOR CHEST RADIOGRAM INTERPRETATIONS USING ANATOMICAL HUMAN STRUCTURES IN HIGH-RESOLUTION CT IMAGES A VIRTUAL TRAINING SYSTEM FOR CHEST RADIOGRAM INTERPRETATIONS USING ANATOMICAL HUMAN STRUCTURES IN HIGH-RESOLUTION CT IMAGES T. Hara*, X. Zhou*, H. Fujita*, I. Kurimoto*, T. Kiryu**, R. Yokoyama**, H.

More information

Derivation of Nutrient Prices from Household level Consumption Data: Methodology and Application*

Derivation of Nutrient Prices from Household level Consumption Data: Methodology and Application* Derivation of Nutrient Prices from Houseold level Consumption Data: Metodology and Application* by Dipankor Coondoo Economic Researc Unit, Indian Statistical Institute, Kolkata, India dcoondoo@isical.ac.in

More information

Allergy: the unmet need

Allergy: the unmet need Allergy: te unmet need A blueprint for better patient care A report of te Royal College of Pysicians Working Party on te provision of allergy services in te UK Royal College of Pysicians June 2003 Royal

More information

Mammography limitations. Clinical performance of digital breast tomosynthesis compared to digital mammography: blinded multi-reader study

Mammography limitations. Clinical performance of digital breast tomosynthesis compared to digital mammography: blinded multi-reader study Clinical performance of digital breast tomosynthesis compared to digital mammography: blinded multi-reader study G. Gennaro (1), A. Toledano (2), E. Baldan (1), E. Bezzon (1), C. di Maggio (1), M. La Grassa

More information

Radiation Dose Reduction: Should You Use a Bismuth Breast Shield?

Radiation Dose Reduction: Should You Use a Bismuth Breast Shield? Radiation Dose Reduction: Should You Use a Bismuth Breast Shield? Lincoln L. Berland, M.D., F.A.C.R. Michael V. Yester, Ph.D. University of Alabama at Birmingham Breast Radiation on CT Use of chest CT

More information

Public Assessment Report Scientific discussion. Kagitz (quetiapine) SE/H/1589/01, 04-05/DC

Public Assessment Report Scientific discussion. Kagitz (quetiapine) SE/H/1589/01, 04-05/DC Public Assessment Report Scientific discussion Kagitz (quetiapine) SE/H/1589/01, 0405/DC Tis module reflects te scientific discussion for te approval of Kagitz. Te procedure was finalised on 20161221.

More information

On the feasibility of speckle reduction in echocardiography using strain compounding

On the feasibility of speckle reduction in echocardiography using strain compounding Title On the feasibility of speckle reduction in echocardiography using strain compounding Author(s) Guo, Y; Lee, W Citation The 2014 IEEE International Ultrasonics Symposium (IUS 2014), Chicago, IL.,

More information

the risk of heart disease and stroke in alabama: burden document

the risk of heart disease and stroke in alabama: burden document te risk of eart disease and stroke in alabama: burden document finding te Pat to CardioVasCUlar ealt 21 table of Contents executive Summary... 1 demograpic caracteristics of Alabama... 2 Leading causes

More information

A GEOMETRICAL OPTIMIZATION PROBLEM ASSOCIATED WITH FRUITS OF POPPY FLOWER. Muradiye, Manisa, Turkey. Muradiye, Manisa, Turkey.

A GEOMETRICAL OPTIMIZATION PROBLEM ASSOCIATED WITH FRUITS OF POPPY FLOWER. Muradiye, Manisa, Turkey. Muradiye, Manisa, Turkey. Matematical and Computational Applications, Vol. 14, No. 3, pp. 169-175, 009. Association for Scientific Researc A GEOMETRICAL OPTIMIZATION PROBLEM ASSOCIATED WITH FRUITS OF POPPY FLOWER Gözde Değer 1,

More information

Copyright 2008 Society of Photo Optical Instrumentation Engineers. This paper was published in Proceedings of SPIE, vol. 6915, Medical Imaging 2008:

Copyright 2008 Society of Photo Optical Instrumentation Engineers. This paper was published in Proceedings of SPIE, vol. 6915, Medical Imaging 2008: Copyright 2008 Society of Photo Optical Instrumentation Engineers. This paper was published in Proceedings of SPIE, vol. 6915, Medical Imaging 2008: Computer Aided Diagnosis and is made available as an

More information

A NON INVASIVE COMPUTER AIDED DIAGNOSIS SYSTEM FOR EARLY DETECTION OF LUNG CARCINOMA IN CT MEDICAL IMAGES

A NON INVASIVE COMPUTER AIDED DIAGNOSIS SYSTEM FOR EARLY DETECTION OF LUNG CARCINOMA IN CT MEDICAL IMAGES International Journal of Latest Trends in Engineering and Tecnology Vol.(8)Issue(4-1), pp.125-130 DOI: ttp://dx.doi.org/10.21172/1.841.22 e-issn:2278-621x NON INVSIVE COMPUTER IDED DIGNOSIS SYSTEM FOR

More information

LUNG CANCER continues to rank as the leading cause

LUNG CANCER continues to rank as the leading cause 1138 IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 24, NO. 9, SEPTEMBER 2005 Computer-Aided Diagnostic Scheme for Distinction Between Benign and Malignant Nodules in Thoracic Low-Dose CT by Use of Massive

More information

Two optimal treatments of HIV infection model

Two optimal treatments of HIV infection model SSN 746-733 England UK World Journal of Modelling and Simulation Vol. 8 () No. pp. 7-35 Two optimal treatments of HV infection model Kalid Hattaf Noura Yousfi Laboratory Analysis Modeling and Simulation

More information

Copyright 2008 Society of Photo Optical Instrumentation Engineers. This paper was published in Proceedings of SPIE, vol. 6915, Medical Imaging 2008:

Copyright 2008 Society of Photo Optical Instrumentation Engineers. This paper was published in Proceedings of SPIE, vol. 6915, Medical Imaging 2008: Copyright 2008 Society of Photo Optical Instrumentation Engineers. This paper was published in Proceedings of SPIE, vol. 6915, Medical Imaging 2008: Computer Aided Diagnosis and is made available as an

More information

Improved image quality of low-dose thoracic CT examinations with a new postprocessing software*

Improved image quality of low-dose thoracic CT examinations with a new postprocessing software* JOURNAL OF APPLIED CLINICAL MEDICAL PHYSICS, VOLUME 11, NUMBER 3, Summer 2010 Improved image quality of low-dose thoracic CT examinations with a new postprocessing software* Anne Catrine Traegde Martinsen,

More information

Ask EuroSafe Imaging. Tips & Tricks. Paediatric Imaging Working Group. Shielding in pediatric CT

Ask EuroSafe Imaging. Tips & Tricks. Paediatric Imaging Working Group. Shielding in pediatric CT Ask EuroSafe Imaging Tips & Tricks Paediatric Imaging Working Group Shielding in pediatric CT Claudio Granata (IRCCS Istituto Giannina Gaslini, IT) Joana Santos (ESTeSC-Coimbra Health School, PT) Elina

More information

State of the art and future development for standardized estimation of organ doses in CT

State of the art and future development for standardized estimation of organ doses in CT State of the art and future development for standardized estimation of organ doses in CT March 2015 William J. O Connel, Dr. Ph, Senior Medical Physicist Imagination at work. Agenda Introduction Duke Florida

More information

Citation Knight J, Andrade M (2018) Genes and chromosomes 4: common genetic conditions. Nursing Times [online]; 114: 10,

Citation Knight J, Andrade M (2018) Genes and chromosomes 4: common genetic conditions. Nursing Times [online]; 114: 10, Genetics Keywords Nondisjunction/Aneuploidy/ Trisomy/Autosomal/Recessive Tis article as been double-blind peer reviewed In tis article... Definitions of nondisjunction and aneuploidy Clinical features

More information

Improvement of Image Quality with ß-Blocker Premedication on ECG-Gated 16-MDCT Coronary Angiography

Improvement of Image Quality with ß-Blocker Premedication on ECG-Gated 16-MDCT Coronary Angiography 16-MDCT Coronary Angiography Shim et al. 16-MDCT Coronary Angiography Sung Shine Shim 1 Yookyung Kim Soo Mee Lim Received December 1, 2003; accepted after revision June 1, 2004. 1 All authors: Department

More information

To Shield or Not to Shield? Lincoln L. Berland, M.D.

To Shield or Not to Shield? Lincoln L. Berland, M.D. To Shield or Not to Shield? Lincoln L. Berland, M.D. Disclosures Consultant to: Nuance, Inc. Page 2 Breast Radiation on CT Use of chest CT has increased in women vulnerable to cancer induction by radiation.

More information

Ultralow Dose Chest CT with MBIR

Ultralow Dose Chest CT with MBIR Ultralow Dose Chest CT with MBIR Ella A. Kazerooni, M.D. Professor & Director Cardiothoracic Radiology Associate Chair for Clinical Affairs University of Michigan Disclosures Consultant: GE Healthcare

More information

Update of Digital Breast Tomosynthesis. Susan Orel Roth, MD

Update of Digital Breast Tomosynthesis. Susan Orel Roth, MD Update of Digital Breast Tomosynthesis Susan Orel Roth, MD NCI estimates that : Why DBT? Approximately 20% of breast cancers are missed at mammography screening Average recall rates approximately 10%

More information

Original Article Detection of lymph node metastases in cholangiocanma by fourier transform infrared spectroscopy

Original Article Detection of lymph node metastases in cholangiocanma by fourier transform infrared spectroscopy Int J Clin Exp Med 207;0(2):2925-293 www.ijcem.com /ISSN:940-590/IJCEM004693 Original Article Detection of lymp node metastases in colangiocanma by fourier transform infrared spectroscopy Min Wu *, Huan-Rong

More information

Computer Assisted Radiology and Surgery

Computer Assisted Radiology and Surgery Computer Assisted Radiology and Surgery How Can a Massive Training Artificial Neural Network (MTANN) Be Trained With a Small Number of Cases in the Distinction Between Nodules and Vessels in Thoracic CT?

More information

Modifi ed CT perfusion contrast injection protocols for improved CBF quantifi cation with lower temporal sampling

Modifi ed CT perfusion contrast injection protocols for improved CBF quantifi cation with lower temporal sampling Investigations and research Modifi ed CT perfusion contrast injection protocols for improved CBF quantifi cation with lower temporal sampling J. Wang Z. Ying V. Yao L. Ciancibello S. Premraj S. Pohlman

More information

Public Assessment Report. Scientific discussion. Efavirenz/Emtricitabine/Tenofovirdisoproxil Teva, film-coated tablets

Public Assessment Report. Scientific discussion. Efavirenz/Emtricitabine/Tenofovirdisoproxil Teva, film-coated tablets Public Assessment Report Scientific discussion Efavirenz/Emtricitabine/Tenofovirdisoproxil Teva, film-coated tablets (efavirenz/emtricitabine/tenofovir disoproxil) NL/H/3602/001/DC Date: 28 September 2017

More information

New Horizons in the Imaging of the Lung

New Horizons in the Imaging of the Lung New Horizons in the Imaging of the Lung Postprocessing. How to do it and when do we need it? Peter M.A. van Ooijen, MSc, PhD Principal Investigator, Radiology, UMCG Discipline Leader Medical Imaging Informatics

More information

Copyright 2007 IEEE. Reprinted from 4th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, April 2007.

Copyright 2007 IEEE. Reprinted from 4th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, April 2007. Copyright 27 IEEE. Reprinted from 4th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, April 27. This material is posted here with permission of the IEEE. Such permission of the

More information

B-Flow, Power Doppler and Color Doppler Ultrasound in the Assessment of Carotid Stenosis: Comparison with 64-MD-CT Angiography

B-Flow, Power Doppler and Color Doppler Ultrasound in the Assessment of Carotid Stenosis: Comparison with 64-MD-CT Angiography Med. J. Cairo Univ., Vol. 85, No. 2, March: 805-809, 2017 www.medicaljournalofcairouniversity.net B-Flow, Power Doppler and Color Doppler Ultrasound in the Assessment of Carotid Stenosis: Comparison with

More information

Pediatric chest HRCT using the idose 4 Hybrid Iterative Reconstruction Algorithm: Which idose level to choose?

Pediatric chest HRCT using the idose 4 Hybrid Iterative Reconstruction Algorithm: Which idose level to choose? Journal of Physics: Conference Series PAPER OPEN ACCESS Pediatric chest HRCT using the idose 4 Hybrid Iterative Reconstruction Algorithm: Which idose level to choose? To cite this article: M Smarda et

More information

ttp://www.sis.se ttp://www.sis.se ttp://www.sis.se ttp://www.sis.se ttp://www.sis.se Provläsningsexemplar / Preview Teknisk rapport SIS-CEN ISO/TR 25107:2006 Utgåva 1 Oktober 2006 Oförstörande provning

More information

8/10/2016. PET/CT Radiomics for Tumor. Anatomic Tumor Response Assessment in CT or MRI. Metabolic Tumor Response Assessment in FDG-PET

8/10/2016. PET/CT Radiomics for Tumor. Anatomic Tumor Response Assessment in CT or MRI. Metabolic Tumor Response Assessment in FDG-PET PET/CT Radiomics for Tumor Response Evaluation August 1, 2016 Wei Lu, PhD Department of Medical Physics www.mskcc.org Department of Radiation Oncology www.umaryland.edu Anatomic Tumor Response Assessment

More information

Chest CT with Ultra-High Resolution Collimator for Submillimeter Fat Plane Detection: A Phantom Study

Chest CT with Ultra-High Resolution Collimator for Submillimeter Fat Plane Detection: A Phantom Study Chest CT with Ultra-High Resolution Collimator for Submillimeter Fat Plane Detection: A Phantom Study Poster No.: C-1207 Congress: ECR 2013 Type: Scientific Exhibit Authors: Y. Shimomiya, M. Kondo, M.

More information

The Gold Standard. ASDIN 2014 Scientific Meeting. When is an AVF mature? Longitudinal Assessment of AVF Maturation with Ultrasound.

The Gold Standard. ASDIN 2014 Scientific Meeting. When is an AVF mature? Longitudinal Assessment of AVF Maturation with Ultrasound. Wen is an AVF mature? Longitudinal Assessment of AVF Maturation wit Ultrasound CONFLICT of INTEREST DECLARATION St Micael s Hospital olds patent on use of ig frequency probes for evaluation of venous structures

More information

A TREATMENT PLANNING STUDY COMPARING VMAT WITH 3D CONFORMAL RADIOTHERAPY FOR PROSTATE CANCER USING PINNACLE PLANNING SYSTEM *

A TREATMENT PLANNING STUDY COMPARING VMAT WITH 3D CONFORMAL RADIOTHERAPY FOR PROSTATE CANCER USING PINNACLE PLANNING SYSTEM * Romanian Reports in Physics, Vol. 66, No. 2, P. 394 400, 2014 A TREATMENT PLANNING STUDY COMPARING VMAT WITH 3D CONFORMAL RADIOTHERAPY FOR PROSTATE CANCER USING PINNACLE PLANNING SYSTEM * D. ADAM 1,2,

More information

Dual Energy CT Aortography: Can We Reduce Iodine Dose??

Dual Energy CT Aortography: Can We Reduce Iodine Dose?? Dual Energy CT Aortography: Can We Reduce Iodine Dose?? William P. Shuman MD, FACR FSCBTMR Department of Radiology University of Washington SCBTMR Annual Course Boston, October 10, 2012 Conflict of Interest

More information

Automatic Definition of Planning Target Volume in Computer-Assisted Radiotherapy

Automatic Definition of Planning Target Volume in Computer-Assisted Radiotherapy Automatic Definition of Planning Target Volume in Computer-Assisted Radiotherapy Angelo Zizzari Department of Cybernetics, School of Systems Engineering The University of Reading, Whiteknights, PO Box

More information

THE TUFFEST STUFF CT REGISTRY REVIEW Live Lecture Seminar SATURDAY CURRICULUM

THE TUFFEST STUFF CT REGISTRY REVIEW Live Lecture Seminar SATURDAY CURRICULUM 1. The CT Imaging Chain-10 major components & their functions a. The x-ray tube b. Generator c. Filter d. Pre-patient collimator e. Pre-detector collimator f. Detector system g. Analog to digital converter

More information

FOR CMS (MEDICARE) MEMBERS ONLY NATIONAL COVERAGE DETERMINATION (NCD) FOR MAGNETIC RESONANCE IMAGING:

FOR CMS (MEDICARE) MEMBERS ONLY NATIONAL COVERAGE DETERMINATION (NCD) FOR MAGNETIC RESONANCE IMAGING: National Imaging Associates, Inc. Clinical guidelines BONE MARROW MRI Original Date: July 2008 Page 1 of 5 CPT Codes: 77084 Last Review Date: September 2014 NCD 220.2 MRI Last Effective Date: July 2011

More information

Anatomical and Functional MRI of the Pancreas

Anatomical and Functional MRI of the Pancreas Anatomical and Functional MRI of the Pancreas MA Bali, MD, T Metens, PhD Erasme Hospital Free University of Brussels Belgium mbali@ulb.ac.be Introduction The use of MRI to investigate the pancreas has

More information

Measurement error of spiral CT Volumetry:

Measurement error of spiral CT Volumetry: Measurement error of spiral CT Volumetry: Influence of Low Dose CT Technique 1 Tae Gyu Lee, M.D. 2, Myung Jin Chung, M.D., Sung Bum Cho, M.D. 2, Jae Min Cho, M.D., Seog Joon Kim, M.D. 2, Sang Hyun Baik,

More information

Small Pulmonary Nodules: Our Preliminary Experience in Volumetric Analysis of Doubling Times

Small Pulmonary Nodules: Our Preliminary Experience in Volumetric Analysis of Doubling Times Small Pulmonary Nodules: Our Preliminary Experience in Volumetric Analysis of Doubling Times Andrea Borghesi, MD Davide Farina, MD Roberto Maroldi, MD Department of Radiology University of Brescia Brescia,

More information

Public Assessment Report. Scientific discussion. (Atorvastatin calcium) SE/H/757/01-03/DC

Public Assessment Report. Scientific discussion. (Atorvastatin calcium) SE/H/757/01-03/DC Public Assessment Report Scientific discussion Atovans 10mg, 20mg and 40mg film coated tablets (Atorvastatin calcium) SE/H/757/01-03/DC Tis module reflects te scientific discussion for te approval of Atovans.

More information

Detection of architectural distortion using multilayer back propagation neural network

Detection of architectural distortion using multilayer back propagation neural network Available online www.jocpr.com Journal of Chemical and Pharmaceutical Research, 2015, 7(2):292-297 Research Article ISSN : 0975-7384 CODEN(USA) : JCPRC5 Detection of architectural distortion using multilayer

More information

Computed tomographic pulmonary angiography procedures: Contrast media dilution from the venous to the systemic circulation

Computed tomographic pulmonary angiography procedures: Contrast media dilution from the venous to the systemic circulation Computed tomographic pulmonary angiography procedures: Contrast media dilution from the venous to the systemic circulation Petter Bugge Askeland Project thesis at the Faculty of Medicine UNIVERSITETET

More information

AN ALGORITHM FOR EARLY BREAST CANCER DETECTION IN MAMMOGRAMS

AN ALGORITHM FOR EARLY BREAST CANCER DETECTION IN MAMMOGRAMS AN ALGORITHM FOR EARLY BREAST CANCER DETECTION IN MAMMOGRAMS Isaac N. Bankman', William A. Christens-Barryl, Irving N. Weinberg2, Dong W. Kim3, Ralph D. Semmell, and William R. Brody2 The Johns Hopkins

More information

Breast screening: understanding case difficulty and the nature of errors

Breast screening: understanding case difficulty and the nature of errors Loughborough University Institutional Repository Breast screening: understanding case difficulty and the nature of errors This item was submitted to Loughborough University's Institutional Repository by

More information

Ultrasound. Computed tomography. Case studies. Utility of IQon Spectral CT in. cardiac imaging

Ultrasound. Computed tomography. Case studies. Utility of IQon Spectral CT in. cardiac imaging Ultrasound Computed tomography Case studies Utility of IQon Spectral CT in cardiac imaging Cardiac imaging is a challenging procedure where it is necessary to image a motion-free heart. This requires a

More information

Linear Ultrasonic Wave Propagation in Biological Tissues

Linear Ultrasonic Wave Propagation in Biological Tissues Indian Journal of Biomechanics: Special Issue (NCBM 7-8 March 29) Linear Ultrasonic Wave Propagation in Biological Tissues Narendra D Londhe R. S. Anand 2, 2 Electrical Engineering Department, IIT Roorkee,

More information

Preparations for pandemic influenza. Guidance for hospital medical specialties on adaptations needed for a pandemic influenza outbreak

Preparations for pandemic influenza. Guidance for hospital medical specialties on adaptations needed for a pandemic influenza outbreak Preparations for pandemic influenza Guidance for ospital medical specialties on adaptations needed for a pandemic influenza outbreak June 2009 Te Royal College of Pysicians Te Royal College of Pysicians

More information

Paediatric Dose Reduction and Image Quality

Paediatric Dose Reduction and Image Quality Paediatric Dose Reduction and Image Quality Alan Whiteside The majority of this work was undertaken as part of MSc Thesis of Helen Dixon. Introduction Paediatric CT protocols result in a higher effective

More information

TG13 management bundles for acute cholangitis and cholecystitis

TG13 management bundles for acute cholangitis and cholecystitis J Hepatobiliary Pancreat Sci (2013) 20:55 59 DOI 10.1007/s00534-012-0562-2 GUIDELINE TG13: Updated Tokyo Guidelines for acute colangitis and acute colecystitis TG13 management bundles for acute colangitis

More information

PERSPECTIVES OF THE TOOTH RESTORATION TECHNOLOGY BASED ON THE COMPUTED TOMOGRAPHY DATA

PERSPECTIVES OF THE TOOTH RESTORATION TECHNOLOGY BASED ON THE COMPUTED TOMOGRAPHY DATA PERSPECTIVES OF THE TOOTH RESTORATION TECHNOLOGY BASED ON THE COMPUTED TOMOGRAPHY DATA Maxim Putrik *, Vladimir Ivanov, Igor Antsygin Federal State Autonomous Educational Institution of Higher Education

More information

Automatic Patient Centering for MDCT: Effect on Radiation Dose

Automatic Patient Centering for MDCT: Effect on Radiation Dose Patient Centering for MDCT CT Imaging Original Research Jianhai Li 1 Unni K. Udayasankar 1 Thomas L. Toth 2 John Seamans 2 William C. Small 1 Mannudeep K. Kalra 1,3 Li J, Udayasankar UK, Toth TL, Seamans

More information

Insights. Central Nervous System Cancers, Version

Insights. Central Nervous System Cancers, Version NCCN Guidelines Insigts Central Nervous System Cancers NCCN Guidelines Insigts Featured Updates to te NCCN Guidelines 1331 Louis Burt Nabors, MD 1,* ; Jana Portnow, MD 2,* ; Mario Ammirati, MD, MBA 3 ;

More information

UK TOBY Cooling Register

UK TOBY Cooling Register UK TOBY Cooling Register Patient identification number (PIN) Patient ospital number At centre, optional Cooling treatment provided at Name of Hospital Sex M F Mont and year of treatment M M / Y Y Gestational

More information

A Kinematic Assessment of Knee Prosthesis from Fluoroscopy Images

A Kinematic Assessment of Knee Prosthesis from Fluoroscopy Images Memoirs of the Faculty of Engineering, Kyushu University, Vol. 68, No. 1, March 2008 A Kinematic Assessment of Knee Prosthesis from Fluoroscopy Images by Mohammad Abrar HOSSAIN *, Michihiko FUKUNAGA and

More information

Current status of diagnostic imaging in dental university hospitals in Japan

Current status of diagnostic imaging in dental university hospitals in Japan Oral Radiol (2004) 20:15 21 Japanese Society for Oral and Maxillofacial Radiology and Springer-Verlag Tokyo 2004 DOI 10.1007/s11282-004-0010-3 ORIGINAL ARTICLE Takehito Sasaki Minoru Fujita Tsuguhisa Katoh

More information

doi: /

doi: / Yiting Xie ; Matthew D. Cham ; Claudia Henschke ; David Yankelevitz ; Anthony P. Reeves; Automated coronary artery calcification detection on low-dose chest CT images. Proc. SPIE 9035, Medical Imaging

More information

Manual Ultrasonic Inspection of Thin Metal Welds

Manual Ultrasonic Inspection of Thin Metal Welds 11th European Conference on Non-Destructive Testing (ECNDT 2014), October 6-10, 2014, Prague, Czech Republic Manual Ultrasonic Inspection of Thin Metal Welds More Info at Open Access Database www.ndt.net/?id=16364

More information

Radiology Rounds A Newsletter for Referring Physicians Massachusetts General Hospital Department of Radiology

Radiology Rounds A Newsletter for Referring Physicians Massachusetts General Hospital Department of Radiology Radiology Rounds A Newsletter for Referring Physicians Massachusetts General Hospital Department of Radiology Minimizing CT Radiation Dose CT examinations improve health care and are an essential part

More information

Dr. P.V. Ramaraju 1, Satti Praveen 2 Department of Electronics and Communication SRKR Engineering College Andhra Pradesh, INDIA

Dr. P.V. Ramaraju 1, Satti Praveen 2 Department of Electronics and Communication SRKR Engineering College Andhra Pradesh, INDIA Classification of lung tumour Using Geometrical and Texture Features of Chest X-ray Images Dr. P.V. Ramaraju 1, Satti Praveen 2 Department of Electronics and Communication SRKR Engineering College Andhra

More information

TRAUMATIC HIP DISLOCATION IN CHILDHOOD

TRAUMATIC HIP DISLOCATION IN CHILDHOOD Actaortop. scand. 50,549-553,1979 TRAUMATIC HIP DISLOCATION IN CHILDHOOD A Report of 26 Cases and a Review of te Literature ANTONIO B ARQUET Institute and Department of Ortopaedics and Traumatology, Montevideo,

More information

Early Detection of Lung Cancer

Early Detection of Lung Cancer Early Detection of Lung Cancer Aswathy N Iyer Dept Of Electronics And Communication Engineering Lymie Jose Dept Of Electronics And Communication Engineering Anumol Thomas Dept Of Electronics And Communication

More information

Developing a Statistical Method of Quantifying Vascular Response after Radiotherapy Co-supervised by Dr. Glenn Bauman and Dr.

Developing a Statistical Method of Quantifying Vascular Response after Radiotherapy Co-supervised by Dr. Glenn Bauman and Dr. 6 Week Project Developing a Statistical Method of Quantifying Vascular Response after Radiotherapy Co-supervised by Dr. Glenn Bauman and Dr. Slav Yartsev Michal Stankiewicz April 9, 2013 Medical Biophysics,

More information

Horizon Scanning Technology Summary. Magnetic resonance angiography (MRA) imaging for the detection of coronary artery disease

Horizon Scanning Technology Summary. Magnetic resonance angiography (MRA) imaging for the detection of coronary artery disease Horizon Scanning Technology Summary National Horizon Scanning Centre Magnetic resonance angiography (MRA) imaging for the detection of coronary artery disease April 2007 This technology summary is based

More information

Low-dose CT Lung Cancer Screening Guidelines for Pulmonary Nodules Management Version 2

Low-dose CT Lung Cancer Screening Guidelines for Pulmonary Nodules Management Version 2 Low-dose CT Lung Cancer Screening Guidelines for Pulmonary Nodules Management Version 2 The Committee for Management of CT-screening-detected Pulmonary Nodules 2009-2011 The Japanese Society of CT Screening

More information

Dual-Energy Imaging of Bone Marrow Edema on a Dedicated Multi-Source Cone-Beam CT System for the Extremities

Dual-Energy Imaging of Bone Marrow Edema on a Dedicated Multi-Source Cone-Beam CT System for the Extremities Dual-Energy Imaging of Bone Edema on a Dedicated Multi-Source Cone-Beam CT System for the Extremities W Zbijewski, 1 A Sisniega, 1 JW Stayman, 1 N Packard, 2 J Yorkston, 2 G Thawait, 3 S Demehri, 3 J Fritz,

More information

T3 Short Implants. Implants & Instrumentation

T3 Short Implants. Implants & Instrumentation T3 Sort Implants Implants & Instrumentation Implant Treatment Options for Vertical Heigt Deficiencies Te T3 Sort Implant s lengt and features are designed to provide an implant treatment option in some

More information