Detecção do mal-posicionamento rotacional de dedos em dispositivos de captura de impressões digitais multivista sem toque utilizando redes neurais artificiais

Dissertação (mestrado)—Universidade de Brasília, Faculdade de Tecnologia, Departamento de Engenharia Mecânica, 2016.

Texture-Detail Preservation Measurement in Camera Phones: An Updated Approach

Recent advances in mobile phone cameras have poised them to take over compact hand-held cameras as the consumer’s preferred camera option. Along with advances in the number of pixels, motion blur removal, face-tracking, and noise reduction algorithms have significant roles in the internal processing of the devices. An undesired effect of severe noise reduction is the loss of texture (i.e. low-contrast fine details) of the original scene. Current established methods for resolution measurement fail to accurately portray the texture loss incurred in a camera system. The development of an accurate objective method to identify the texture preservation or texture reproduction capability of a camera device is important in this regard. The ‘Dead Leaves’ target has been used extensively as a method to measure the modulation transfer function (MTF) of cameras that employ highly non-linear noise-reduction methods. This stochastic model consists of a series of overlapping circles with radii r distributed as r−3, and having uniformly distributed gray level, which gives an accurate model of occlusion in a natural setting and hence mimics a natural scene. This target can be used to model the texture transfer through a camera system when a natural scene is captured. In the first part of our study we identify various factors that affect the MTF measured using the ‘Dead Leaves’ chart. These include variations in illumination, distance, exposure time and ISO sensitivity among others. We discuss the main differences of this method with the existing resolution measurement techniques and identify the advantages. In the second part of this study, we propose an improvement to the current texture MTF measurement algorithm. High frequency residual noise in the processed image contains the same frequency content as fine texture detail, and is sometimes reported as such, thereby leading to inaccurate results. A wavelet thresholding based denoising technique is utilized for modeling the noise present in the final captured image. This updated noise model is then used for calculating an accurate texture MTF. We present comparative results for both algorithms under various image capture conditions.

Repetibilidade da mensuração de imagens das características de carcaça obtidas por ultrassonografia em fêmeas Nelore

Avaliou-se a repetibilidade da mensuração de imagens de ultrassom da área do músculo longissimus dorsi (AOL) e das espessuras de gordura subcutânea do lombo (EGL) e da garupa (EGG). Imagens de ultrassom tomadas no lombo (entre a 12ª e a 13ª costela) e na garupa (entre os músculos gluteus medium e biceps femoris) de novilhas Nelore de 14 a 22 meses de idade foram classificadas em aceitáveis, marginais e rejeitáveis. As imagens aceitáveis e marginais foram mensuradas duas vezes por três técnicos em diferentes níveis de treinamento. Foram estimadas as repetibilidades entre e dentro de técnicos por classe de qualidade da imagem, para determinação do efeito da qualidade da imagem e do técnico no valor absoluto da diferença entre a primeira e a segunda mensuração dessas características. A repetibilidade para as imagens aceitáveis foi maior que para imagens marginais, tanto entre como dentro de técnicos. Na análise da diferença absoluta entre a primeira e a segunda interpretação, foram significativos os efeitos de técnico para AOL e EGL e de classe de qualidade da imagem para AOL. Em geral, o técnico com maior experiência apresentou maiores valores de repetibilidade. É recomendável que a mensuração de imagens de animais de mesmo grupo contemporâneo seja feita por um único técnico.

Denisiuk-type reflection holography display with sillenite crystals for imaging and interferometry of small objects

A novel setup for imaging and interferometry through reflection holography with Bi12TiPO20(BTO) sillenite photorefractive crystals is proposed. A variation of the lensless Denisiuk arrangement was developed resulting in a compact, robust and simple interferometer. A red He-Ne laser was used as light source and the holographic recording occurred by diffusion with the grating vector parallel to the crystal [0 0 1]-axis. In order to enhance the holographic image quality and reduce noise a polarizing beam splitter (PBS) was positioned at the BTO input and the crystal was tilted around the [0 0 1]-axis. This enabled the orthogonally polarized transmission and diffracted beams to be separated by the PBS, providing the holographic image only. The possibility of performing deformation and strain analysis as well as vibration measurement of small objects was demonstrated. (C) 2007 Elsevier B.V. All rights reserved.

Providing Integrity and Authenticity in DICOM Images: A Novel Approach

The increasing adoption of information systems in healthcare has led to a scenario where patient information security is more and more being regarded as a critical issue. Allowing patient information to be in jeopardy may lead to irreparable damage, physically, morally, and socially to the patient, potentially shaking the credibility of the healthcare institution. Medical images play a crucial role in such context, given their importance in diagnosis, treatment, and research. Therefore, it is vital to take measures in order to prevent tampering and determine their provenance. This demands adoption of security mechanisms to assure information integrity and authenticity. There are a number of works done in this field, based on two major approaches: use of metadata and use of watermarking. However, there still are limitations for both approaches that must be properly addressed. This paper presents a new method using cryptographic means to improve trustworthiness of medical images, providing a stronger link between the image and the information on its integrity and authenticity, without compromising image quality to the end user. Use of Digital Imaging and Communications in Medicine structures is also an advantage for ease of development and deployment.

A wavelet visible difference predictor

In this paper, we describe a model of the human visual system (HVS) based on the wavelet transform. This model is largely based on a previously proposed model, but has a number of modifications that make it more amenable to potential integration into a wavelet based image compression scheme. These modifications include the use of a separable wavelet transform instead of the cortex transform, the application of a wavelet contrast sensitivity function (CSP), and a simplified definition of subband contrast that allows us to predict noise visibility directly from wavelet coefficients. Initially, we outline the luminance, frequency, and masking sensitivities of the HVS and discuss how these can be incorporated into the wavelet transform. We then outline a number of limitations of the wavelet transform as a model of the HVS, namely the lack of translational invariance and poor orientation sensitivity. In order to investigate the efficacy of this wavelet based model, a wavelet visible difference predictor (WVDP) is described. The WVDP is then used to predict visible differences between an original and compressed (or noisy) image. Results are presented to emphasize the limitations of commonly used measures of image quality and to demonstrate the performance of the WVDP, The paper concludes with suggestions on bow the WVDP can be used to determine a visually optimal quantization strategy for wavelet coefficients and produce a quantitative measure of image quality.

The HI Parkes All Sky Survey: southern observations, calibration and robust imaging

The acquisition of HI Parkes All Shy Survey (HIPASS) southern sky data commenced at the Australia Telescope National Facility's Parkes 64-m telescope in 1997 February, and was completed in 2000 March. HIPASS is the deepest HI survey yet of the sky south of declination +2 degrees, and is sensitive to emission out to 170 h(75)(-1) Mpc. The characteristic root mean square noise in the survey images is 13.3 mJy. This paper describes the survey observations, which comprise 23 020 eight-degree scans of 9-min duration, and details the techniques used to calibrate and image the data. The processing algorithms are successfully designed to be statistically robust to the presence of interference signals, and are particular to imaging point (or nearly point) sources. Specifically, a major improvement in image quality is obtained by designing a median-gridding algorithm which uses the median estimator in place of the mean estimator.

Wavefront analysis and modulation transfer function of three multifocal intraocular lenses

Purpose: To evaluate wavefront performance and modulation transfer function (MTF) in the human eye aft er the implantation of diffractive or refractive multifocal intraocular lenses (IOLs). Materials and Methods: This was a prospective, interventional, comparative, nonrandomized clinical study. Uncorrected distance and near visual acuity, and wavefront analysis including MTF curves (iTrace aberrometer, Tracey Technologies, Houston, TX, USA) were measured in 60 patients aft er bilateral IOL implantation with 6 months of follow-up. Forty eyes received the diffractive ReSTOR (Alcon), 40 eyes received the refractive ReZoom (Advanced Medical Optics) and 40 eyes, the Tecnis ZM900 (Advanced Medical Optics). The comparison of MTF and aberration between the intraocular lenses was performed using analysis of variance (ANOVA), followed by the Dunn test when necessary. Results: The mean uncorrected distance visual acuity was similar in all three groups of multifocal IOLs. The ReSTOR group provided better uncorrected near visual acuity than the ReZoom group (P < 0.001), but similar to the Tecnis group. Spherical aberration was significantly higher in the ReZoom group (P = 0.007). Similar MTF curves were found for the aspheric multifocal IOL Tecnis and the spheric multifocal IOL ReSTOR, and both performed better than the multifocal IOL ReZoom in a 5 mm pupil (P < 0.001 at all spatial frequencies). Conclusions: Diffractive IOLs studied presented similar MTF curves for a 5 mm pupil diameter. Both diffractive IOLs showed similar spherical aberration, which was significantly better with the full-diffractive IOL Tecnis than with the refractive IOL ReZoom.

Coronary CT angiography using 64 detector rows: methods and design of the multi-centre trial CORE-64

Multislice computed tomography (MSCT) for the noninvasive detection of coronary artery stenoses is a promising candidate for widespread clinical application because of its non-invasive nature and high sensitivity and negative predictive value as found in several previous studies using 16 to 64 simultaneous detector rows. A multi-centre study of CT coronary angiography using 16 simultaneous detector rows has shown that 16-slice CT is limited by a high number of nondiagnostic cases and a high false-positive rate. A recent meta-analysis indicated a significant interaction between the size of the study sample and the diagnostic odds ratios suggestive of small study bias, highlighting the importance of evaluating MSCT using 64 simultaneous detector rows in a multi-centre approach with a larger sample size. In this manuscript we detail the objectives and methods of the prospective ""CORE-64"" trial (""Coronary Evaluation Using Multidetector Spiral Computed Tomography Angiography using 64 Detectors""). This multi-centre trial was unique in that it assessed the diagnostic performance of 64-slice CT coronary angiography in nine centres worldwide in comparison to conventional coronary angiography. In conclusion, the multi-centre, multi-institutional and multi-continental trial CORE-64 has great potential to ultimately assess the per-patient diagnostic performance of coronary CT angiography using 64 simultaneous detector rows.

Coronary Artery Stenoses: Accuracy of 64-Detector Row CT Angiography in Segments with Mild, Moderate, or Severe Calcification-A Subanalysis of the CORE-64 Trial

Purpose: To evaluate the influence of cross-sectional arc calcification on the diagnostic accuracy of computed tomography (CT) angiography compared with conventional coronary angiography for the detection of obstructive coronary artery disease (CAD). Materials and Methods: Institutional Review Board approval and written informed consent were obtained from all centers and participants for this HIPAA-compliant study. Overall, 4511 segments from 371 symptomatic patients (279 men, 92 women; median age, 61 years [interquartile range, 53-67 years]) with clinical suspicion of CAD from the CORE-64 multi-center study were included in the analysis. Two independent blinded observers evaluated the percentage of diameter stenosis and the circumferential extent of calcium (arc calcium). The accuracy of quantitative multidetector CT angiography to depict substantial (>50%) stenoses was assessed by using quantitative coronary angiography (QCA). Cross-sectional arc calcium was rated on a segment level as follows: noncalcified or mild (<90 degrees), moderate (90 degrees-180 degrees), or severe (>180 degrees) calcification. Univariable and multivariable logistic regression, receiver operation characteristic curve, and clustering methods were used for statistical analyses. Results: A total of 1099 segments had mild calcification, 503 had moderate calcification, 338 had severe calcification, and 2571 segments were noncalcified. Calcified segments were highly associated (P < .001) with disagreement between CTA and QCA in multivariable analysis after controlling for sex, age, heart rate, and image quality. The prevalence of CAD was 5.4% in noncalcified segments, 15.0% in mildly calcified segments, 27.0% in moderately calcified segments, and 43.0% in severely calcified segments. A significant difference was found in area under the receiver operating characteristic curves (noncalcified: 0.86, mildly calcified: 0.85, moderately calcified: 0.82, severely calcified: 0.81; P < .05). Conclusion: In a symptomatic patient population, segment-based coronary artery calcification significantly decreased agreement between multidetector CT angiography and QCA to detect a coronary stenosis of at least 50%.

Quantitative and qualitative comparison of 3.0 T and 1.5 T MR imaging of the liver in patients with diffuse parenchymal liver disease

Purpose: The purpose of our study was to compare signal characteristics and image qualities of MR imaging at 3.0 T and 1.5 T in patients with diffuse parenchymal liver disease. Materials and methods: 25 consecutive patients with diffuse parenchymal liver disease underwent abdominal MR imaging at both 3.0 T and 1.5 T within a 6-month interval. A retrospective study was conducted to obtain quantitative and qualitative data from both 3.0 T and 1.5 T MRI. Quantitative image analysis was performed by measuring the signal-to-noise ratios (SNRs) and the contrast-to-noise ratios (CNRs) by the Students t-test. Qualitative image analysis was assessed by grading each sequence on a 3- and 4-point scale, regarding the presence of artifacts and image quality, respectively. Statistical analysis consisted of the Wilcoxon signed-rank test. Results: the mean SNRs and CNRs of the liver parenchyma and the portal vein were significantly higher at 3.0 T than at 1.5 T on portal and equilibrium phases of volumetric interpolated breath-hold examination (VIBE) images (P < 0.05). The mean SNRs were significantly higher at 3.0 T than at 1.5 T on T1-weighted spoiled gradient echo (SGE) images (P < 0.05). However, there were no significantly differences on T2-weighted short-inversion-time inversion recovery (STIR) images. Overall image qualities of the 1.5 T noncontrast T1- and T2-weighted sequences were significantly better than 3.0 T (P < 0.01). In contrast, overall image quality of the 3.0 T post-gadolinium VIBE sequence was significantly better than 1.5 T (P< 0.01). Conclusions: MR imaging of post-gadolinium VIBE sequence at 3.0 T has quantitative and qualitative advantages of evaluating for diffuse parenchymal liver disease. (C) 2008 Elsevier Ireland Ltd. All rights reserved.

3.0-T MRI evaluation of patients with chronic liver diseases: initial observations

Purpose: To describe the use of 3.0-T magnetic resonance imaging (MRI) for the evaluation of chronic liver diseases. Materials and Methods: Two groups of patients who had chronic liver diseases and underwent 3.0-T MRI for evaluation of the liver were included in the study. The first group of patients included 66 consecutive patients (33 male, 33 female; mean age +/- standard deviation, 56 +/- 11). The second group of patients included 30 consecutive patients (18 males, 12 females; mean age +/- standard deviation, 53 +/- 10) in whom Variable-Rate Selective Excitation (VERSE) pulses and improved adjustments procedure were used during the acquisitions. Imaging findings of chronic liver diseases, predetermined artifacts and image quality of all individual sequences in the first group and predetermined artifacts and image quality of T2-weighted sequences in the second group were reviewed retrospectively and independently by two reviewers. chi-Square tests were used to compare the findings between two groups of patients and individual sequences. Kappa statistics were used to determine the extent of agreement between the reviewers. Results: Fifteen dysplastic nodules in 6 of 66 (9%) patients and 12 hepatocellular carcinomas in 11 of 66 (17%) patients were detected. Excluding motion artifacts, three-dimensional (313) T1-weighted gradient-echo (GE) sequence was the least affected sequence by the artifacts. Image quality of T1-weighted 3D-GE sequences was excellent in 43 of 66 (65%) patients. In-phase and out-of-phase T1-weighted spoiled GE (SGE) images were fair in 62 of 66 (94%) and 61 of 66 (92%) patients, respectively. The image quality of short tau inversion recovery (STIR) and half-Fourier rapid acquisition with relaxation enhancement (RARE) sequences were fair in 31 of 66 (47%) and 53 of 66 (80%) patients. STIR and half-Fourier RARE sequences in the second group demonstrated significantly better image quality (P=.03 and P<.0001). Conclusion: 3.0-T MRI allows the acquisition of very high quality postgadolinium 3D-GE sequence, which permitted the detection and characterization of lesions in the setting of chronic liver diseases. The use of VERSE pulses and improved adjustments procedure improved the image quality of T2-weighted sequences. In-phase/out-of-phase SGE sequences are at present of fair quality. (C) 2008 Elsevier Inc. All rights reserved.

Water excitation MPRAGE: An alternative sequence for postcontrast imaging of the abdomen in noncooperative patients at 1.5 Tesla and 3.0 Tesla MRI

Purpose: To evaluate the diagnostic image quality of post-gadolinium water excitation-magnetization-prepared rapid gradient-echo (WE-MPRAGE) sequence in abdominal examinations of noncooperative patients at 1.5 Tesla (T) and 3.0T MRI. Materials and Methods: Eighty-nine consecutive patients (48 males and 41 females; mean age +/- standard deviation, 54.6 +/- 16.6 years) who had MRI examinations including postgadolinium WE-MPRAGE were included in the study. Of 89 patients, 33 underwent noncooperative protocol at 1.5T. 10 under-went noncooperative protocol at 3.0T, and 46 underwent cooperative protocol at 3.0T. Postgadolinium WE-MPRAGE, MPRAGE, and three-dimensional gradient-echo sequences of these three different groups were qualitatively evaluated for image quality, extent of artifacts, lesion conspicuity, and homogeneity of fat-attenuation by two reviewers retrospectively, independently, and blindly. The results were compared using Wilcoxon signed rank and Mann-Whitney U tests. Kappa statistics were used to measure the extent of agreement between the reviewers. Results: The average scores indicated that the images were diagnostic for WE-MPRAGE at 1.5T and 3.0T in noncooperative patients. WE-MPRAGE achieved homogenous fat-attenuation in 31/33 (94%) of noncooperative patients at 1.5T and 10/10 (100%) of noncooperative patients at 3.0T. WE-MPRAGE at 3.0T had better results for image quality, extent of artifacts, lesion conspicuity and homogeneity of fat-attenuation compared with WE-MPRAGE at 1.5T. in noncooperative patients (P = 0.0008, 0.0006, 0.0024, and 0.0042: respectively). Kappa statistics varied between 0.76 and 1.00, representing good to excellent agreement. Conclusion: WE-MPRAGE may be used as a T1-weighted postgadolinium fat-attenuated sequence in noncooperative patients, particularly at 3.0T MRI.

Refinements in stress echocardiographic techniques improve inter-institutional agreement in interpretation of dobutamine stress echocardiograms

Aims To determine the degree of inter-institutional agreement in the assessment of dobutamine stress echocardiograms using modern stress echo cardiographic technology in combination with standardized data acquisition and assessment criteria. Method and Results Among six experienced institutions, 150 dobutamine stress echocardiograms (dobutamine up to 40 mug.kg(-1) min(-1) and atropine up to I mg) were performed on patients with suspected coronary artery disease using fundamental and harmonic imaging following a consistent digital acquisition protocol. Each dobutamine stress echocardiogram was assessed at every institution regarding endocardial visibility and left ventricular wall motion without knowledge of any other data using standardized reading criteria. No patients were excluded due to poor image quality or inadequate stress level. Coronary angiography was performed within 4 weeks. Coronary angiography demonstrated significant coronary artery disease (less than or equal to50% diameter stenosis) in 87 patients. Using harmonic imaging an average of 5.2+/-0.9 institutions agreed on dobutamine stress echocardiogram results as being normal or abnormal (mean kappa 0.55; 95% CI 0.50-0.60). Agreement was higher in patients with no (equal assessment of dobutamine stress echocardiogram results by 5.5 +/- 0.8 institutions) or three-vessel coronary artery disease (5.4 +/- 0.8 institutions) and lower in one- or two- vessel disease (5.0 +/- 0.9 and 5.2 +/- 1.0 institutions, respectively-, P=0.041). Disagreement on test results was greater in only minor wall motion abnormalities. Agreement on dobutamine stress echocardiogram results was lower using fundamental imaging (mean kappa 0.49; 95% CI 0.44-0.54; P

Routine ultrasound screening in diabetic pregnancies

Objectives To assess the detection rate of congenital fetal malformations and specific problems related to routine ultrasound screening in women with pre-existing diabetes. Methods A retrospective study was carried out to assess the performance of routine ultrasound screening in women with pre-existing diabetes (Types 1 and 2) within a tertiary institution. The incidence, type and risk factors for congenital fetal malformations were determined. The detection rate of fetal anomalies for diabetic women was compared with that for the low-risk population. Factors affecting these detection rates were evaluated. Results During the study period, 12 169 low-risk pregnant women and 130 women with pre-existing diabetes had routine ultrasound screening performed within the institution. A total of 10 major anomalies (7.7%) and three minor anomalies (2.3%) were present in the fetuses of the diabetic women. Central nervous system and cardiovascular system anomalies accounted for 60% of the major anomalies. Peri-conceptional hemoglobin A 1 c of more than 9% was associated with a high prevalence of major anomalies (14311000). Women who had fetuses with major anomalies bad a significantly higher incidence of obesity (78% vs. 37%; P < 0.05). Ultrasound examination of these diabetic pregnancies showed high incidences of suboptimal image quality (37%), incomplete examinations, and repeat examinations (17%). Compared to the 'low-risk' non-diabetic population from the same institution, the relative risk for a major congenital anomaly among the diabetic women was 5.9-fold higher (95% confidence interval, 2.9-11.9). The detection rate for major fetal anomalies was significantly lower for diabetic women (30% vs. 73%; P < 0.01), and the mean body mass index for the diabetic group was significantly higher (29 vs. 23 kg/m(2); P < 0.001). Conclusion The incidence of congenital anomalies is higher in diabetic pregnancies. Unfortunately, the detection rate for fetal anomalies by antenatal ultrasound scan was significantly, worse than that for the low-risk population. This is likely to be related to the maternal body habitus and unsatisfactory examinations. Methods to overcome these difficulties are discussed.