Objective assessment of image quality. II. Fisher information, Fourier crosstalk, and figures of merit for task performance

Response surface methodology for predicting optimal conditions in very low-dose chest CT imaging

OBJECTIVES

Dose reduction techniques, such as new reconstruction algorithms and automated exposure control systems vary with manufacturer and scanner models, complicating the optimization and standardization procedures. We investigated the feasibility of using the design of experiments in CT protocols optimization.

MATERIALS & METHODS

A Doehlert matrix was used to define the experiments to carry out. Measurements were conducted on a 128-slice CT scanner using an anthropomorphic chest phantom with a 5 mm diameter lesion that has a HU of -800. CT images were reconstructed using iterative (ASIR-V) and deep learning-based reconstruction techniques at low (DLIR-L) and high (DLIR-H) strengths. Lesion detectability was assessed using two self-supervised learning-based model observers and six human observers. Second-order polynomial functions have been established to model the combined effect of noise index (NI) and percentage of ASIR-V on dose and model observers' performances. The analysis of agreement between model and human observers was performed using correlation coefficients and Bland-Altman test.

RESULTS

The optimal conditions predicted by this method were NI = 64, % ASIR-V = 60 and DLIR-H reconstruction. They were found in good agreement with the experimental results obtained by the average human observer, as showed by the Bland-Altman plot with a mean absolute difference of -0.01 ± 3.16. Compared to 60 % ASIR-V, these results suggested an approximately 64 % dose reduction potential for DLIR-H without compromising lesion detection.

CONCLUSION

The proposed method can predict the optimal conditions that ensure diagnostic quality of low-dose chest CT examinations, while minimizing the number of experiments to carry out.

Applying Self-Supervised Learning to Image Quality Assessment in Chest CT Imaging

Many new reconstruction techniques have been deployed to allow low-dose CT examinations. Such reconstruction techniques exhibit nonlinear properties, which strengthen the need for a task-based measure of image quality. The Hotelling observer (HO) is the optimal linear observer and provides a lower bound of the Bayesian ideal observer detection performance. However, its computational complexity impedes its widespread practical usage. To address this issue, we proposed a self-supervised learning (SSL)-based model observer to provide accurate estimates of HO performance in very low-dose chest CT images. Our approach involved a two-stage model combining a convolutional denoising auto-encoder (CDAE) for feature extraction and dimensionality reduction and a support vector machine for classification. To evaluate this approach, we conducted signal detection tasks employing chest CT images with different noise structures generated by computer-based simulations. We compared this approach with two supervised learning-based methods: a single-layer neural network (SLNN) and a convolutional neural network (CNN). The results showed that the CDAE-based model was able to achieve similar detection performance to the HO. In addition, it outperformed both SLNN and CNN when a reduced number of training images was considered. The proposed approach holds promise for optimizing low-dose CT protocols across scanner platforms.

Need for Objective Task-Based Evaluation of Image Segmentation Algorithms for Quantitative PET: A Study with ACRIN 6668/RTOG 0235 Multicenter Clinical Trial Data

Reliable performance of PET segmentation algorithms on clinically relevant tasks is required for their clinical translation. However, these algorithms are typically evaluated using figures of merit (FoMs) that are not explicitly designed to correlate with clinical task performance. Such FoMs include the Dice similarity coefficient (DSC), the Jaccard similarity coefficient (JSC), and the Hausdorff distance (HD). The objective of this study was to investigate whether evaluating PET segmentation algorithms using these task-agnostic FoMs yields interpretations consistent with evaluation on clinically relevant quantitative tasks. Methods: We conducted a retrospective study to assess the concordance in the evaluation of segmentation algorithms using the DSC, JSC, and HD and on the tasks of estimating the metabolic tumor volume (MTV) and total lesion glycolysis (TLG) of primary tumors from PET images of patients with non-small cell lung cancer. The PET images were collected from the American College of Radiology Imaging Network 6668/Radiation Therapy Oncology Group 0235 multicenter clinical trial data. The study was conducted in 2 contexts: (1) evaluating conventional segmentation algorithms, namely those based on thresholding (SUVmax40% and SUVmax50%), boundary detection (Snakes), and stochastic modeling (Markov random field-Gaussian mixture model); (2) evaluating the impact of network depth and loss function on the performance of a state-of-the-art U-net-based segmentation algorithm. Results: Evaluation of conventional segmentation algorithms based on the DSC, JSC, and HD showed that SUVmax40% significantly outperformed SUVmax50%. However, SUVmax40% yielded lower accuracy on the tasks of estimating MTV and TLG, with a 51% and 54% increase, respectively, in the ensemble normalized bias. Similarly, the Markov random field-Gaussian mixture model significantly outperformed Snakes on the basis of the task-agnostic FoMs but yielded a 24% increased bias in estimated MTV. For the U-net-based algorithm, our evaluation showed that although the network depth did not significantly alter the DSC, JSC, and HD values, a deeper network yielded substantially higher accuracy in the estimated MTV and TLG, with a decreased bias of 91% and 87%, respectively. Additionally, whereas there was no significant difference in the DSC, JSC, and HD values for different loss functions, up to a 73% and 58% difference in the bias of the estimated MTV and TLG, respectively, existed. Conclusion: Evaluation of PET segmentation algorithms using task-agnostic FoMs could yield findings discordant with evaluation on clinically relevant quantitative tasks. This study emphasizes the need for objective task-based evaluation of image segmentation algorithms for quantitative PET.

Emission Ghost Imaging: reconstruction with data augmentation

Ghost Imaging enables 2D reconstruction of an object even though particles transmitted or emitted by the object of interest are detected with a single pixel detector without spatial resolution. This is possible because for the particular implementation of ghost imaging presented here, the incident beam is spatially modulated with a non-configurable attenuating mask whose orientation is varied (e.g. via transverse displacement or rotation) in the course of the ghost imaging experiment. Each orientation yields a distinct spatial pattern in the attenuated beam. In many cases, ghost imaging reconstructions can be dramatically improved by factoring the measurement matrix which consists of measured attenuated incident radiation for each of many orientations of the mask at each pixel to be reconstructed as the product of an orthonormal matrix Q and an upper triangular matrix R provided that the number of orientations of the mask (N ) is greater than or equal to the number of pixels (P ) reconstructed. For the N < P case, we present a data augmentation method that enables QR factorization of the measurement matrix. To suppress noise in the reconstruction, we determine the Moore-Penrose pseudoinverse of the measurement matrix with a truncated singular value decomposition approach. Since the resulting reconstruction is still noisy, we denoise it with the Adaptive Weights Smoothing method. In simulation experiments, our method outperforms a modification of an existing alternative orthogonalization method where rows of the measurement matrix are orthogonalized by the Gram-Schmidt method. We apply our ghost imaging methods to experimental X-ray fluorescence data acquired at Brookhaven National Laboratory.

Characterization and Assessment of Projection Probability Density Function and Enhanced Sampling in Self-Collimation SPECT

We have recently reported a self-collimation SPECT (SC-SPECT) design concept that constructs sensitive detectors in a multi-ring interspaced mosaic architecture to simultaneously improve system spatial resolution and sensitivity. In this work, through numerical and Monte-Carlo simulation studies, we investigate this new design concept by analyzing its projection probability density functions (PPDF) and the effects of enhanced sampling, i.e. having rotational and translational object movements during imaging. We first quantitatively characterize PPDFs by their widths and edge slopes. Then we compare the PPDFs of an SC-SPECT and a series of multiple-pinhole SPECT (MPH-SPECT) systems and assess the impact of PPDFs - combined with enhanced sampling - on image contrast recovery coefficient and variance through phantom studies. We show the PPDFs of SC- SPECT have steeper edges and a wider range of width, and these attributes enable SC-SPECT to achieve better performance.

Need for objective task-based evaluation of deep learning-based denoising methods: A study in the context of myocardial perfusion SPECT

BACKGROUND

Artificial intelligence-based methods have generated substantial interest in nuclear medicine. An area of significant interest has been the use of deep-learning (DL)-based approaches for denoising images acquired with lower doses, shorter acquisition times, or both. Objective evaluation of these approaches is essential for clinical application.

PURPOSE

DL-based approaches for denoising nuclear-medicine images have typically been evaluated using fidelity-based figures of merit (FoMs) such as root mean squared error (RMSE) and structural similarity index measure (SSIM). However, these images are acquired for clinical tasks and thus should be evaluated based on their performance in these tasks. Our objectives were to: (1) investigate whether evaluation with these FoMs is consistent with objective clinical-task-based evaluation; (2) provide a theoretical analysis for determining the impact of denoising on signal-detection tasks; and (3) demonstrate the utility of virtual imaging trials (VITs) to evaluate DL-based methods.

METHODS

A VIT to evaluate a DL-based method for denoising myocardial perfusion SPECT (MPS) images was conducted. To conduct this evaluation study, we followed the recently published best practices for the evaluation of AI algorithms for nuclear medicine (the RELAINCE guidelines). An anthropomorphic patient population modeling clinically relevant variability was simulated. Projection data for this patient population at normal and low-dose count levels (20%, 15%, 10%, 5%) were generated using well-validated Monte Carlo-based simulations. The images were reconstructed using a 3-D ordered-subsets expectation maximization-based approach. Next, the low-dose images were denoised using a commonly used convolutional neural network-based approach. The impact of DL-based denoising was evaluated using both fidelity-based FoMs and area under the receiver operating characteristic curve (AUC), which quantified performance on the clinical task of detecting perfusion defects in MPS images as obtained using a model observer with anthropomorphic channels. We then provide a mathematical treatment to probe the impact of post-processing operations on signal-detection tasks and use this treatment to analyze the findings of this study.

RESULTS

Based on fidelity-based FoMs, denoising using the considered DL-based method led to significantly superior performance. However, based on ROC analysis, denoising did not improve, and in fact, often degraded detection-task performance. This discordance between fidelity-based FoMs and task-based evaluation was observed at all the low-dose levels and for different cardiac-defect types. Our theoretical analysis revealed that the major reason for this degraded performance was that the denoising method reduced the difference in the means of the reconstructed images and of the channel operator-extracted feature vectors between the defect-absent and defect-present cases.

CONCLUSIONS

The results show the discrepancy between the evaluation of DL-based methods with fidelity-based metrics versus the evaluation on clinical tasks. This motivates the need for objective task-based evaluation of DL-based denoising approaches. Further, this study shows how VITs provide a mechanism to conduct such evaluations computationally, in a time and resource-efficient setting, and avoid risks such as radiation dose to the patient. Finally, our theoretical treatment reveals insights into the reasons for the limited performance of the denoising approach and may be used to probe the effect of other post-processing operations on signal-detection tasks.

Comparison of supervised-learning approaches for designing a channelized observer for image quality assessment in CT

BACKGROUND

The current paradigm for evaluating computed tomography (CT) system performance relies on a task-based approach. As the Hotelling observer (HO) provides an upper bound of observer performances in specific signal detection tasks, the literature advocates HO use for optimization purposes. However, computing the HO requires calculating the inverse of the image covariance matrix, which is often intractable in medical applications. As an alternative, dimensionality reduction has been extensively investigated to extract the task-relevant features from the raw images. This can be achieved by using channels, which yields the channelized-HO (CHO). The channels are only considered efficient when the channelized observer (CO) can approximate its unconstrained counterpart. Previous work has demonstrated that supervised learning-based methods can usually benefit CO design, either for generating efficient channels using partial least squares (PLS) or for replacing the Hotelling detector with machine-learning (ML) methods.

PURPOSE

Here we investigated the efficiency of a supervised ML-algorithm used to design a CO for predicting the performance of unconstrained HO. The ML-algorithm was applied either (1) in the estimator for dimensionality reduction, or (2) in the detector function.

METHODS

A channelized support vector machine (CSVM) was employed and compared against the CHO in terms of ability to predict HO performances. Both the CSVM and the CHO were estimated with channels derived from the singular value decomposition (SVD) of the system operator, principal component analysis (PCA), and PLS. The huge variety of regularization strategies proposed by CT system vendors for statistical image reconstruction (SIR) make the generalization capability of an observer a key point to consider upfront of implementation in clinical practice. To evaluate the generalization properties of the observers, we adopted a 2-step testing process: (1) achieved with the same regularization strategy (as in the training phase) and (2) performed using different reconstruction properties. We generated simulated- signal-known-exactly/background-known-exactly (SKE/BKE) tasks in which different noise structures were generated using Markov random field (MRF) regularizations using either a Green or a quadratic, function.

RESULTS

The CSVM outperformed the CHO for all types of channels and regularization strategies. Furthermore, even though both COs generalized well to images reconstructed with the same regularization strategy as the images considered in the training phase, the CHO failed to generalize to images reconstructed differently whereas the CSVM managed to successfully generalize. Lastly, the proposed CSVM observer used with PCA channels outperformed the CHO with PLS channels while using a smaller training data set.

CONCLUSION

These results argue for introducing the supervised-learning paradigm in the detector function rather than in the operator of the channels when designing a CO to provide an accurate estimate of HO performance. The CSVM with PCA channels proposed here could be used as a surrogate for HO in image quality assessment.

Investigating the limited performance of a deep-learning-based SPECT denoising approach: An observer-study-based characterization

Multiple objective assessment of image-quality (OAIQ)-based studies have reported that several deep-learning (DL)-based denoising methods show limited performance on signal-detection tasks. Our goal was to investigate the reasons for this limited performance. To achieve this goal, we conducted a task-based characterization of a DL-based denoising approach for individual signal properties. We conducted this study in the context of evaluating a DL-based approach for denoising single photon-emission computed tomography (SPECT) images. The training data consisted of signals of different sizes and shapes within a clustered-lumpy background, imaged with a 2D parallel-hole-collimator SPECT system. The projections were generated at normal and 20% low-count level, both of which were reconstructed using an ordered-subset-expectation-maximization (OSEM) algorithm. A convolutional neural network (CNN)-based denoiser was trained to process the low-count images. The performance of this CNN was characterized for five different signal sizes and four different signal-to-background ratio (SBRs) by designing each evaluation as a signal-known-exactly/background-known-statistically (SKE/BKS) signal-detection task. Performance on this task was evaluated using an anthropomorphic channelized Hotelling observer (CHO). As in previous studies, we observed that the DL-based denoising method did not improve performance on signal-detection tasks. Evaluation using the idea of observer-study-based characterization demonstrated that the DL-based denoising approach did not improve performance on the signal-detection task for any of the signal types. Overall, these results provide new insights on the performance of the DL-based denoising approach as a function of signal size and contrast. More generally, the observer study-based characterization provides a mechanism to evaluate the sensitivity of the method to specific object properties, and may be explored as analogous to characterizations such as modulation transfer function for linear systems. Finally, this work underscores the need for objective task-based evaluation of DL-based denoising approaches.

Design Study of an Ultrahigh Resolution Brain SPECT System Using a Synthetic Compound-Eye Camera Design With Micro-Slit and Micro-Ring Apertures

In this paper, we discuss the design study for a brain SPECT imaging system, referred to as the HelmetSPECT system, based on a spherical synthetic compound-eye (SCE) gamma camera design. The design utilizes a large number (  ∼ 500 ) of semiconductor detector modules, each coupled to an aperture with a very narrow opening for high-resolution SPECT imaging applications. In this study, we demonstrate that this novel system design could provide an excellent spatial resolution, a very high sensitivity, and a rich angular sampling without scanning motion over a clinically relevant field-of-view (FOV). These properties make the proposed HelmetSPECT system attractive for dynamic imaging of epileptic patients during seizures. In ictal SPECT, there is typically no prior information on where the seizures would happen, and both the imaging resolution and quantitative accuracy of the dynamic SPECT images would provide critical information for staging the seizures outbreak and refining the plans for subsequent surgical intervention.We report the performance evaluation and comparison among similar system geometries using non-conventional apertures, such as micro-ring and micro-slit, and traditional lofthole apertures. We demonstrate that the combination of ultrahigh-resolution imaging detectors, the SCE gamma camera design, and the micro-ring and micro-slit apertures would offer an interesting approach for the future ultrahigh-resolution clinical SPECT imaging systems without sacrificing system sensitivity and FOV.

Objective Task-Based Evaluation of Artificial Intelligence-Based Medical Imaging Methods:: Framework, Strategies, and Role of the Physician

Artificial intelligence-based methods are showing promise in medical imaging applications. There is substantial interest in clinical translation of these methods, requiring that they be evaluated rigorously. We lay out a framework for objective task-based evaluation of artificial intelligence methods. We provide a list of available tools to conduct this evaluation. We outline the important role of physicians in conducting these evaluation studies. The examples in this article are proposed in the context of PET scans with a focus on evaluating neural network-based methods. However, the framework is also applicable to evaluate other medical imaging modalities and other types of artificial intelligence methods.

Impact of iterative reconstruction algorithms on the applicability of Fourier-based detectability index for x-ray CT imaging

PURPOSE

The increasing application of iterative reconstruction algorithms in clinical computed tomography to improve image quality and reduce radiation dose, elicits strong interest, and needs model observers to optimize CT scanning protocols objectively and efficiently. The current paradigm for evaluating imaging system performance relies on Fourier methods, which presuppose a linear, wide-sense stationary system. Long-range correlations introduced by iterative reconstruction algorithms may narrow the applicability of Fourier techniques. Differences in the implementation of reconstruction algorithms between manufacturers add further complexity. The present work set out to quantify the errors entailed by the use of Fourier methods, which can lead to design decisions that do not correlate with detectability.

METHODS

To address this question, we evaluated the noise properties and the detectability index of the ideal linear observer using the spatial approach and the Fourier-based approach. For this purpose, a homogeneous phantom was imaged on two scanners: the Revolution CT (GE Healthcare) and the Somatom Definition AS+ (Siemens Healthcare) at different exposure levels. Images were reconstructed using different strength levels of IR algorithms available on the systems considered: Adaptative Statistical Iterative Reconstruction (ASIR-V) and Sinogram Affirmed Iterative Reconstruction (SAFIRE).

RESULTS

Our findings highlight that the spatial domain estimate of the detectability index is higher than the Fourier domain estimate. This trend is found to be dependent on the specific regularization used by IR algorithms as well as the signal to be detected. The eigenanalysis of the noise covariance matrix and of its circulant approximation yields explanation about the evoked trends. In particular, this analysis suggests that the predictive power of the Fourier-based ideal linear observer depends on the ability of each basis analyzed to be relevant to the signal to be detected.

CONCLUSION

The applicability of Fourier techniques is dependent on the specific regularization used by IR algorithms. These results argue for verifying the assumptions made when using Fourier methods since Fourier-task-based detectability index does not always correlate with signal detectability.

DeepAMO: a multi-slice, multi-view anthropomorphic model observer for visual detection tasks performed on volume images

Purpose: We propose a deep learning-based anthropomorphic model observer (DeepAMO) for image quality evaluation of multi-orientation, multi-slice image sets with respect to a clinically realistic 3D defect detection task. Approach: The DeepAMO is developed based on a hypothetical model of the decision process of a human reader performing a detection task using a 3D volume. The DeepAMO is comprised of three sequential stages: defect segmentation, defect confirmation (DC), and rating value inference. The input to the DeepAMO is a composite image, typical of that used to view 3D volumes in clinical practice. The output is a rating value designed to reproduce a human observer's defect detection performance. In stages 2 and 3, we propose: (1) a projection-based DC block that confirms defect presence in two 2D orthogonal orientations and (2) a calibration method that "learns" the mapping from the features of stage 2 to the distribution of observer ratings from the human observer rating data (thus modeling inter- or intraobserver variability) using a mixture density network. We implemented and evaluated the DeepAMO in the context of   Tc 99 m -DMSA SPECT imaging. A human observer study was conducted, with two medical imaging physics graduate students serving as observers. A 5 × 2 -fold cross-validation experiment was conducted to test the statistical equivalence in defect detection performance between the DeepAMO and the human observer. We also compared the performance of the DeepAMO to an unoptimized implementation of a scanning linear discriminant observer (SLDO). Results: The results show that the DeepAMO's and human observer's performances on unseen images were statistically equivalent with a margin of difference ( Δ AUC ) of 0.0426 at p < 0.05 , using 288 training images. A limited implementation of an SLDO had a substantially higher AUC (0.99) compared to the DeepAMO and human observer. Conclusion: The results show that the DeepAMO has the potential to reproduce the absolute performance, and not just the relative ranking of human observers on a clinically realistic defect detection task, and that building conceptual components of the human reading process into deep learning-based models can allow training of these models in settings where limited training images are available.

Optimizing illumination for precise multi-parameter estimations in coherent diffractive imaging

Coherent diffractive imaging (CDI) is widely used to characterize structured samples from measurements of diffracting intensity patterns. We introduce a numerical framework to quantify the precision that can be achieved when estimating any given set of parameters characterizing the sample from measured data. The approach, based on the calculation of the Fisher information matrix, provides a clear benchmark to assess the performance of CDI methods. Moreover, by optimizing the Fisher information metric using deep learning optimization libraries, we demonstrate how to identify the optimal illumination scheme that minimizes the estimation error under specified experimental constraints. This work paves the way for an efficient characterization of structured samples at the sub-wavelength scale.

Interobserver variability in quality assessment of magnetic resonance images

Background

The perceptual quality of magnetic resonance (MR) images influences diagnosis and may compromise the treatment. The purpose of this study was to evaluate how the image quality changes influence the interobserver variability of their assessment.

Methods

For the variability evaluation, a dataset containing distorted MRI images was prepared and then assessed by 31 experienced medical professionals (radiologists). Differences between observers were analyzed using the Fleiss’ kappa. However, since the kappa evaluates the agreement among radiologists taking into account aggregated decisions, a typically employed criterion of the image quality assessment (IQA) performance was used to provide a more thorough analysis. The IQA performance of radiologists was evaluated by comparing the Spearman correlation coefficients, ρ, between individual scores with the mean opinion scores (MOS) composed of the subjective opinions of the remaining professionals.

Results

The experiments show that there is a significant agreement among radiologists (κ=0.12; 95% confidence interval [CI]: 0.118, 0.121; P<0.001) on the quality of the assessed images. The resulted κ is strongly affected by the subjectivity of the assigned scores, separately presenting close scores. Therefore, the ρ was used to identify poor performance cases and to confirm the consistency of the majority of collected scores (ρ_mean = 0.5706). The results for interns (ρ_mean = 0.6868) supports the finding that the quality assessment of MR images can be successfully taught.

Conclusions

The agreement observed among radiologists from different imaging centers confirms the subjectivity of the perception of MR images. It was shown that the image content and severity of distortions affect the IQA. Furthermore, the study highlights the importance of the psychosomatic condition of the observers and their attitude.

In-plane image quality and NPWE detectability index in digital breast tomosynthesis

A rigorous 2D analysis of signal and noise transfer applied to reconstructed planes in digital breast tomosynthesis (DBT) is necessary for system characterization and optimization. This work proposes a method for assessing technical image quality and system detective quantum efficiency (DQE_sys) for reconstructed planes in DBT. Measurements of 2D in-plane modulation transfer function (MTF) and noise power spectrum (NPS) were made on five DBT systems using different acquisition parameters, reconstruction algorithms and plane spacing. This work develops the noise equivalent quanta (NEQ), DQE_sys and detectability index (d') calculated using a non-prewhitening model observer with eye filter (NPWE) for reconstructed DBT planes. The images required for this implementation were acquired using a homogeneous test object of thickness 40 mm poly(methyl) methacrylate plus 0.5 mm Al; 2D MTF was calculated from an Al disc of thickness 0.2 mm and diameter 50 mm positioned within the phantom. The radiant contrast of the MTF disc and the air kerma at the system input were used as normalization factors. The NPWE detectability index was then compared to the in-plane contrast-detail (c-d) threshold measured using the CDMAM phantom. The MTF and NPS measured on the different systems showed a strong anisotropy, consistent with the cascaded models developed in the literature for DBT. Detectability indices calculated from the measured MTF and NPS successfully predicted changes in c-d detectability for details between 0.1 mm and 2.0 mm, for DBT plane spacings between 0.5 mm and 10 mm, and for air kerma values at the system input between 157 µGy and 1170 μGy. The linear Pearson correlation between the detectability index and threshold gold thickness of the CDMAM phantom was -0.996. The method implements a parametric means of assessing the technical image quality of reconstructed DBT planes, providing valuable information for optimization of DBT systems.

Is there a role for image science in the brave new world of artificial intelligence?

Knowledge of the principles of image science is essential to the successful application of artificial intelligence in medical imaging.

Task-based image quality assessment in radiation therapy: initial characterization and demonstration with computer-simulation study

In the majority of current radiation therapy (RT) applications, image quality is still assessed subjectively or by utilizing physical measures. A novel theory that applies objective task-based image quality assessment in radiation therapy (IQA-in-RT) was recently proposed, in which the area under the therapeutic operating characteristic curve (AUTOC) was employed as the figure-of-merit (FOM) for evaluating RT effectiveness. Although theoretically more appealing than conventional subjective or physical measures, a comprehensive implementation and evaluation of this novel task-based IQA-in-RT theory is required for its further application in improving clinical RT. In this work, a practical and modular IQA-in-RT framework is presented for implementing this theory for the assessment of imaging components on the basis of RT treatment outcomes. Computer-simulation studies are conducted to demonstrate the feasibility and utility of the proposed IQA-in-RT framework in optimizing x-ray computed tomography (CT) pre-treatment imaging, including the optimization of CT imaging dose and image reconstruction parameters. The potential advantages of optimizing imaging components in the RT workflow by use of the AUTOC as the FOM are also compared against those of other physical measures. The results demonstrate that optimization using the AUTOC leads to selecting different parameters from those indicated by physical measures, potentially improving RT performance. The sources of systemic randomness and bias that affect the determination of the AUTOC are also analyzed. The presented work provides a practical solution for the further investigation and analysis of the task-based IQA-in-RT theory and advances its applications in improving RT clinical practice and cancer patient care.

Predicting Detection Performance on Security X-Ray Images as a Function of Image Quality

Developing methods to predict how image quality affects the task performance is a topic of great interest in many applications. While such studies have been performed in the medical imaging community, little work has been reported in the security X-ray imaging literature. In this paper, we develop models that predict the effect of image quality on the detection of the improvised explosive device components by bomb technicians in images taken using portable X-ray systems. Using a newly developed NIST-LIVE X-Ray Task Performance Database, we created a set of objective algorithms that predict bomb technician detection performance based on the measures of image quality. Our basic measures are traditional image quality indicators (IQIs) and perceptually relevant natural scene statistics (NSS)-based measures that have been extensively used in visible light image quality prediction algorithms. We show that these measures are able to quantify the perceptual severity of degradations and can predict the performance of expert bomb technicians in identifying threats. Combining NSS- and IQI-based measures yields even better task performance prediction than either of these methods independently. We also developed a new suite of statistical task prediction models that we refer to as quality inspectors of X-ray images (QUIX); we believe this is the first NSS-based model for security X-ray images. We also show that QUIX can be used to reliably predict conventional IQI metric values on the distorted X-ray images.

Experimentally measuring a detectability index of a computational imaging system

Computational imaging (CI) systems are an enabling technology for multifunctional cameras capable of performing a wide variety of imaging tasks. However, given the complexity of CI systems, it is often difficult to characterize their performance. In this research, a novel measurement technique is proposed and tested to evaluate the performance of complex non-shift invariant linear CI systems performing a detection task at the system level. The performance is characterized using detectability indexes such as an average Hotelling's statistic (t²). The proposed measurement technique relies on a previously developed general CI system framework. The detectability predicts the upper-bounded signal-to-noise ratio of a linear algorithm through evaluation of a matched filter. The experimental results are compared with theoretical expected values through the Night Vision Integrated Performance Model (NV-IPM) and Monte Carlo simulations. We demonstrate the experimental results for a variety of target sizes, colors, and brightnesses on different colored flat backgrounds. Our results demonstrate how the detectability indexes can provide valuable insight into the final system performance. Finally, the measurement technique is used to compare the detection performance of two different cameras.

Theoretical characterization of annular array as a volumetric optoacoustic ultrasound handheld probe

Optoacoustic ultrasound (OPUS) is a promising hybridized technique for simultaneous acquisition of functional and morphological data. The optical specificity of optoacoustic leverages the diagnostic aptitude of ultrasonography beyond anatomy. However, this integration has been rarely practiced for volumetric imaging. The challenge lies in the effective imaging probes that preserve the functionality of both modalities. The potentials of a sparse annular array for volumetric OPUS imaging are theoretically investigated. In order to evaluate and optimize the performance characteristics of the probe, series of analysis in the framework of system model matrix was carried out. The two criteria of voxel crosstalk and eigenanalysis have been employed to unveil information about the spatial sensitivity, aliasing, and number of definable spatial frequency components. Based on these benchmarks, the optimal parameters for volumetric handheld probe are determined. In particular, the number, size, and the arrangement of the elements and overall aperture dimension were investigated. The result of the numerical simulation suggests that the segmented-annular array of 128 negatively focused elements with 1λ  ×  20λ size, operating at 5-MHz central frequency showcases a good agreement with the physical requirement of both imaging systems. We hypothesize that these features enable a high-throughput volumetric passive/active ultrasonic imaging system with great potential for clinical applications.

On the Complexity of Human Neuroanatomy at the Millimeter Morphome Scale: Developing Codes and Characterizing Entropy Indexed to Spatial Scale

In this work we devise a strategy for discrete coding of anatomical form as described by a Bayesian prior model, quantifying the entropy of this representation as a function of code rate (number of bits), and its relationship geometric accuracy at clinically relevant scales. We study the shape of subcortical gray matter structures in the human brain through diffeomorphic transformations that relate them to a template, using data from the Alzheimer's Disease Neuroimaging Initiative to train a multivariate Gaussian prior model. We find that the at 1 mm accuracy all subcortical structures can be described with less than 35 bits, and at 1.5 mm error all structures can be described with less than 12 bits. This work represents a first step towards quantifying the amount of information ordering a neuroimaging study can provide about disease status.

Effects of Image Quality on the Fundamental Limits of Image Registration Accuracy

For image-guided procedures, the imaging task is often tied to the registration of intraoperative and preoperative images to a common coordinate system. While the accuracy of this registration is a vital factor in system performance, there is a relatively little work that relates registration accuracy to image quality factors, such as dose, noise, and spatial resolution. To create a theoretical model for such a relationship, we present a Fisher information approach to analyze registration performance in explicit dependence on the underlying image quality factors of image noise, spatial resolution, and signal power spectrum. The model yields analysis of the Cramer-Rao lower bound (CRLB), in registration accuracy as a function of factors governing image quality. Experiments were performed in simulation of computed tomography low-contrast soft tissue images and high-contrast bone (head and neck) images to compare the measured accuracy [root mean squared error (RMSE) of the estimated transformations] with the theoretical lower bound. Analysis of the CRLB reveals that registration performance is closely related to the signal-to-noise ratio of the cross-correlation space. While the lower bound is optimistic, it exhibits consistent trends with experimental findings and yields a method for comparing the performance of various registration methods and similarity metrics. Further analysis validated a method for determining optimal post-processing (image filtering) for registration. Two figures of merit (CRLB and RMSE) are presented that unify models of image quality with registration performance, providing an important guide to optimizing intraoperative imaging with respect to the task of registration.

Standardized target-specific detectivity metric for computational imaging systems

Multifunctional cameras capable of performing a wide variety of nearly simultaneous imaging tasks are expected to play a major role in the near future. Computational imaging (CI) systems will serve as one of the main enabling technologies for multifunctional cameras, especially due to the abundance of low-cost, high-speed computational processing available today. An important aspect of these systems is to be able to quantify their performance with respect to specific imaging tasks. However, the non-traditional design of CI systems, both available and proposed, presents a considerable challenge to modeling, comparing, specifying, and measuring their performance. To solve this problem, this paper presents a standardized detection signal-to-noise ratio, referred to as a detectivity metric, along with a general CI system framework. This metric has the flexibility to handle various types of CI systems and specific targets while minimizing the complexity and assumptions needed. The detectivity metric is designed to assess the performance of a CI system searching for a specific known target or signal of interest. An analytical version of the detectivity metric is also presented for a compressive sensing CI system. Special considerations for standardization are also discussed.

Bayesian framework inspired no-reference region-of-interest quality measure for brain MRI images

We describe a postacquisition, attribute-based quality assessment method for brain magnetic resonance imaging (MRI) images. It is based on the application of Bayes theory to the relationship between entropy and image quality attributes. The entropy feature image of a slice is segmented into low- and high-entropy regions. For each entropy region, there are three separate observations of contrast, standard deviation, and sharpness quality attributes. A quality index for a quality attribute is the posterior probability of an entropy region given any corresponding region in a feature image where quality attribute is observed. Prior belief in each entropy region is determined from normalized total clique potential (TCP) energy of the slice. For TCP below the predefined threshold, the prior probability for a region is determined by deviation of its percentage composition in the slice from a standard normal distribution built from 250 MRI volume data provided by Alzheimer's Disease Neuroimaging Initiative. For TCP above the threshold, the prior is computed using a mathematical model that describes the TCP-noise level relationship in brain MRI images. Our proposed method assesses the image quality of each entropy region and the global image. Experimental results demonstrate good correlation with subjective opinions of radiologists for different types and levels of quality distortions.

Classification and evaluation strategies of auto-segmentation approaches for PET: Report of AAPM task group No. 211

PURPOSE

The purpose of this educational report is to provide an overview of the present state-of-the-art PET auto-segmentation (PET-AS) algorithms and their respective validation, with an emphasis on providing the user with help in understanding the challenges and pitfalls associated with selecting and implementing a PET-AS algorithm for a particular application.

APPROACH

A brief description of the different types of PET-AS algorithms is provided using a classification based on method complexity and type. The advantages and the limitations of the current PET-AS algorithms are highlighted based on current publications and existing comparison studies. A review of the available image datasets and contour evaluation metrics in terms of their applicability for establishing a standardized evaluation of PET-AS algorithms is provided. The performance requirements for the algorithms and their dependence on the application, the radiotracer used and the evaluation criteria are described and discussed. Finally, a procedure for algorithm acceptance and implementation, as well as the complementary role of manual and auto-segmentation are addressed.

FINDINGS

A large number of PET-AS algorithms have been developed within the last 20 years. Many of the proposed algorithms are based on either fixed or adaptively selected thresholds. More recently, numerous papers have proposed the use of more advanced image analysis paradigms to perform semi-automated delineation of the PET images. However, the level of algorithm validation is variable and for most published algorithms is either insufficient or inconsistent which prevents recommending a single algorithm. This is compounded by the fact that realistic image configurations with low signal-to-noise ratios (SNR) and heterogeneous tracer distributions have rarely been used. Large variations in the evaluation methods used in the literature point to the need for a standardized evaluation protocol.

CONCLUSIONS

Available comparison studies suggest that PET-AS algorithms relying on advanced image analysis paradigms provide generally more accurate segmentation than approaches based on PET activity thresholds, particularly for realistic configurations. However, this may not be the case for simple shape lesions in situations with a narrower range of parameters, where simpler methods may also perform well. Recent algorithms which employ some type of consensus or automatic selection between several PET-AS methods have potential to overcome the limitations of the individual methods when appropriately trained. In either case, accuracy evaluation is required for each different PET scanner and scanning and image reconstruction protocol. For the simpler, less robust approaches, adaptation to scanning conditions, tumor type, and tumor location by optimization of parameters is necessary. The results from the method evaluation stage can be used to estimate the contouring uncertainty. All PET-AS contours should be critically verified by a physician. A standard test, i.e., a benchmark dedicated to evaluating both existing and future PET-AS algorithms needs to be designed, to aid clinicians in evaluating and selecting PET-AS algorithms and to establish performance limits for their acceptance for clinical use. The initial steps toward designing and building such a standard are undertaken by the task group members.

Linear System Models for Ultrasonic Imaging: Intensity Signal Statistics

Despite a great deal of work characterizing the statistical properties of radio frequency backscattered ultrasound signals, less is known about the statistical properties of demodulated intensity signals. Analysis of intensity is made more difficult by a strong nonlinearity that arises in the process of demodulation. This limits our ability to characterize the spatial resolution and noise properties of B-mode ultrasound images. In this paper, we generalize earlier results on two-point intensity covariance using a multivariate systems approach. We derive the mean and autocovariance function of the intensity signal under Gaussian assumptions on both the object scattering function and acquisition noise, and with the assumption of a locally shift-invariant pulse-echo system function. We investigate the limiting cases of point statistics and a uniform scattering field with a stationary distribution. Results from validation studies using simulation and data from a real system applied to a uniform scattering phantom are presented. In the simulation studies, we find errors less than 10% between the theoretical mean and variance, and sample estimates of these quantities. Prediction of the intensity power spectrum (PS) in the real system exhibits good qualitative agreement (errors less than 3.5 dB for frequencies between 0.1 and 10 cyc/mm, but with somewhat higher error outside this range that may be due to the use of a window in the PS estimation procedure). We also replicate the common finding that the intensity mean is equal to its standard deviation (i.e., signal-to-noise ratio = 1) for fully developed speckle. We show how the derived statistical properties can be used to characterize the quality of an ultrasound linear array for low-contrast patterns using generalized noise-equivalent quanta directly on the intensity signal.

Fundamental limits of image registration performance: Effects of image noise and resolution in CT-guided interventions

PURPOSE

In image-guided procedures, image acquisition is often performed primarily for the task of geometrically registering information from another image dataset, rather than detection / visualization of a particular feature. While the ability to detect a particular feature in an image has been studied extensively with respect to image quality characteristics (noise, resolution) and is an ongoing, active area of research, comparatively little has been accomplished to relate such image quality characteristics to registration performance.

METHODS

To establish such a framework, we derived Cramer-Rao lower bounds (CRLB) for registration accuracy, revealing the underlying dependencies on image variance and gradient strength. The CRLB was analyzed as a function of image quality factors (in particular, dose) for various similarity metrics and compared to registration accuracy using CT images of an anthropomorphic head phantom at various simulated dose levels. Performance was evaluated in terms of root mean square error (RMSE) of the registration parameters.

RESULTS

Analysis of the CRLB shows two primary dependencies: 1) noise variance (related to dose); and 2) sum of squared image gradients (related to spatial resolution and image content). Comparison of the measured RMSE to the CRLB showed that the best registration method, RMSE achieved the CRLB to within an efficiency factor of 0.21, and optimal estimators followed the predicted inverse proportionality between registration performance and radiation dose.

CONCLUSIONS

Analysis of the CRLB for image registration is an important step toward understanding and evaluating an intraoperative imaging system with respect to a registration task. While the CRLB is optimistic in absolute performance, it reveals a basis for relating the performance of registration estimators as a function of noise content and may be used to guide acquisition parameter selection (e.g., dose) for purposes of intraoperative registration.

Motion compensation using origin ensembles in awake small animal positron emission tomography

In emission tomographic imaging, the stochastic origin ensembles algorithm provides unique information regarding the detected counts given the measured data. Precision in both voxel and region-wise parameters may be determined for a single data set based on the posterior distribution of the count density allowing uncertainty estimates to be allocated to quantitative measures. Uncertainty estimates are of particular importance in awake animal neurological and behavioral studies for which head motion, unique for each acquired data set, perturbs the measured data. Motion compensation can be conducted when rigid head pose is measured during the scan. However, errors in pose measurements used for compensation can degrade the data and hence quantitative outcomes. In this investigation motion compensation and detector resolution models were incorporated into the basic origin ensembles algorithm and an efficient approach to computation was developed. The approach was validated against maximum liklihood-expectation maximisation and tested using simulated data. The resultant algorithm was then used to analyse quantitative uncertainty in regional activity estimates arising from changes in pose measurement precision. Finally, the posterior covariance acquired from a single data set was used to describe correlations between regions of interest providing information about pose measurement precision that may be useful in system analysis and design. The investigation demonstrates the use of origin ensembles as a powerful framework for evaluating statistical uncertainty of voxel and regional estimates. While in this investigation rigid motion was considered in the context of awake animal PET, the extension to arbitrary motion may provide clinical utility where respiratory or cardiac motion perturb the measured data.

Optimal channels for channelized quadratic estimators

We present a new method for computing optimized channels for estimation tasks that is feasible for high-dimensional image data. Maximum-likelihood (ML) parameter estimates are challenging to compute from high-dimensional likelihoods. The dimensionality reduction from M measurements to L channels is a critical advantage of channelized quadratic estimators (CQEs), since estimating likelihood moments from channelized data requires smaller sample sizes and inverting a smaller covariance matrix is easier. The channelized likelihood is then used to form ML estimates of the parameter(s). In this work we choose an imaging example in which the second-order statistics of the image data depend upon the parameter of interest: the correlation length. Correlation lengths are used to approximate background textures in many imaging applications, and in these cases an estimate of the correlation length is useful for pre-whitening. In a simulation study we compare the estimation performance, as measured by the root-mean-squared error (RMSE), of correlation length estimates from CQE and power spectral density (PSD) distribution fitting. To abide by the assumptions of the PSD method we simulate an ergodic, isotropic, stationary, and zero-mean random process. These assumptions are not part of the CQE formalism. The CQE method assumes a Gaussian channelized likelihood that can be a valid for non-Gaussian image data, since the channel outputs are formed from weighted sums of the image elements. We have shown that, for three or more channels, the RMSE of CQE estimates of correlation length is lower than conventional PSD estimates. We also show that computing CQE by using a standard nonlinear optimization method produces channels that yield RMSE within 2% of the analytic optimum. CQE estimates of anisotropic correlation length estimation are reported to demonstrate this technique on a two-parameter estimation problem.

Effective noise-suppressed and artifact-reduced reconstruction of SPECT data using a preconditioned alternating projection algorithm

PURPOSE

The authors have recently developed a preconditioned alternating projection algorithm (PAPA) with total variation (TV) regularizer for solving the penalized-likelihood optimization model for single-photon emission computed tomography (SPECT) reconstruction. This algorithm belongs to a novel class of fixed-point proximity methods. The goal of this work is to investigate how PAPA performs while dealing with realistic noisy SPECT data, to compare its performance with more conventional methods, and to address issues with TV artifacts by proposing a novel form of the algorithm invoking high-order TV regularization, denoted as HOTV-PAPA, which has been explored and studied extensively in the present work.

METHODS

Using Monte Carlo methods, the authors simulate noisy SPECT data from two water cylinders; one contains lumpy "warm" background and "hot" lesions of various sizes with Gaussian activity distribution, and the other is a reference cylinder without hot lesions. The authors study the performance of HOTV-PAPA and compare it with PAPA using first-order TV regularization (TV-PAPA), the Panin-Zeng-Gullberg one-step-late method with TV regularization (TV-OSL), and an expectation-maximization algorithm with Gaussian postfilter (GPF-EM). The authors select penalty-weights (hyperparameters) by qualitatively balancing the trade-off between resolution and image noise separately for TV-PAPA and TV-OSL. However, the authors arrived at the same penalty-weight value for both of them. The authors set the first penalty-weight in HOTV-PAPA equal to the optimal penalty-weight found for TV-PAPA. The second penalty-weight needed for HOTV-PAPA is tuned by balancing resolution and the severity of staircase artifacts. The authors adjust the Gaussian postfilter to approximately match the local point spread function of GPF-EM and HOTV-PAPA. The authors examine hot lesion detectability, study local spatial resolution, analyze background noise properties, estimate mean square errors (MSEs), and report the convergence speed and computation time.

RESULTS

HOTV-PAPA yields the best signal-to-noise ratio, followed by TV-PAPA and TV-OSL/GPF-EM. The local spatial resolution of HOTV-PAPA is somewhat worse than that of TV-PAPA and TV-OSL. Images reconstructed using HOTV-PAPA have the lowest local noise power spectrum (LNPS) amplitudes, followed by TV-PAPA, TV-OSL, and GPF-EM. The LNPS peak of GPF-EM is shifted toward higher spatial frequencies than those for the three other methods. The PAPA-type methods exhibit much lower ensemble noise, ensemble voxel variance, and image roughness. HOTV-PAPA performs best in these categories. Whereas images reconstructed using both TV-PAPA and TV-OSL are degraded by severe staircase artifacts; HOTV-PAPA substantially reduces such artifacts. It also converges faster than the other three methods and exhibits the lowest overall reconstruction error level, as measured by MSE.

CONCLUSIONS

For high-noise simulated SPECT data, HOTV-PAPA outperforms TV-PAPA, GPF-EM, and TV-OSL in terms of hot lesion detectability, noise suppression, MSE, and computational efficiency. Unlike TV-PAPA and TV-OSL, HOTV-PAPA does not create sizable staircase artifacts. Moreover, HOTV-PAPA effectively suppresses noise, with only limited loss of local spatial resolution. Of the four methods, HOTV-PAPA shows the best lesion detectability, thanks to its superior noise suppression. HOTV-PAPA shows promise for clinically useful reconstructions of low-dose SPECT data.

Hardware assessment using the multi-module, multi-resolution system (M3R): a signal-detection study

The multi-module, multi-resolution system (M3R) is used for hardware assessment in objective, task-based signal detection studies in projection data. A phantom capable of generating multiple realizations of a random textured background is introduced. Measured backgrounds from this phantom are used along with simulated lumpy and uniform backgrounds to investigate signal-to-noise ratio as a function of exposure time. Results are shown to agree with theoretical predictions, exhibiting a power-law like dependence previously seen for studies performed either in simulation or without an imaging system, and help validate the use of simulated lumpy backgrounds in observer studies. A second study looks at signal-detection performance, measured by AUC (area under the receiver operating characteristic curve), in lumpy backgrounds for 20 M3R aperture combinations as a function of lump size and signal size. Observer performance reveals an improvement in AUC for certain ranges of signal and lump combinations through the use of multiplexed, multiple-pinhole apertures, indicating a need for task-specific aperture optimization. The channelized Hotelling observer is used with Laguerre-Gauss channels for both observer studies. Methods for selection of number of channels and channel width are discussed.

Revisiting the comparison between the Shack-Hartmann and the pyramid wavefront sensors via the Fisher information matrix

Exoplanet direct imaging with large ground based telescopes requires eXtreme Adaptive Optics that couples high-order adaptive optics and coronagraphy. A key element of such systems is the high-order wavefront sensor. We study here several high-order wavefront sensing approaches, and more precisely compare their sensitivity to noise. Three techniques are considered: the classical Shack-Hartmann sensor, the pyramid sensor and the recently proposed LIFTed Shack-Hartmann sensor. They are compared in a unified framework based on precise diffractive models and on the Fisher information matrix, which conveys the information present in the data whatever the estimation method. The diagonal elements of the inverse of the Fisher information matrix, which we use as a figure of merit, are similar to noise propagation coefficients. With these diagonal elements, so called "Fisher coefficients", we show that the LIFTed Shack-Hartmann and pyramid sensors outperform the classical Shack-Hartmann sensor. In photon noise regime, the LIFTed Shack-Hartmann and modulated pyramid sensors obtain a similar overall noise propagation. The LIFTed Shack-Hartmann sensor however provides attractive noise properties on high orders.

Objective assessment and design improvement of a staring, sparse transducer array by the spatial crosstalk matrix for 3D photoacoustic tomography

Accurate reconstruction of 3D photoacoustic (PA) images requires detection of photoacoustic signals from many angles. Several groups have adopted staring ultrasound arrays, but assessment of array performance has been limited. We previously reported on a method to calibrate a 3D PA tomography (PAT) staring array system and analyze system performance using singular value decomposition (SVD). The developed SVD metric, however, was impractical for large system matrices, which are typical of 3D PAT problems. The present study consisted of two main objectives. The first objective aimed to introduce the crosstalk matrix concept to the field of PAT for system design. Figures-of-merit utilized in this study were root mean square error, peak signal-to-noise ratio, mean absolute error, and a three dimensional structural similarity index, which were derived between the normalized spatial crosstalk matrix and the identity matrix. The applicability of this approach for 3D PAT was validated by observing the response of the figures-of-merit in relation to well-understood PAT sampling characteristics (i.e. spatial and temporal sampling rate). The second objective aimed to utilize the figures-of-merit to characterize and improve the performance of a near-spherical staring array design. Transducer arrangement, array radius, and array angular coverage were the design parameters examined. We observed that the performance of a 129-element staring transducer array for 3D PAT could be improved by selection of optimal values of the design parameters. The results suggested that this formulation could be used to objectively characterize 3D PAT system performance and would enable the development of efficient strategies for system design optimization.

Task-based image quality evaluation of iterative reconstruction methods for low dose CT using computer simulations

Iterative reconstruction (IR) methods for x-ray CT is a promising approach to improve image quality or reduce radiation dose to patients. The goal of this work was to use task based image quality measures and the channelized Hotelling observer (CHO) to evaluate both analytic and IR methods for clinical x-ray CT applications. We performed realistic computer simulations at five radiation dose levels, from a clinical reference low dose D0 to 25% D0. A fixed size and contrast lesion was inserted at different locations into the liver of the XCAT phantom to simulate a weak signal. The simulated data were reconstructed on a commercial CT scanner (SOMATOM Definition Flash; Siemens, Forchheim, Germany) using the vendor-provided analytic (WFBP) and IR (SAFIRE) methods. The reconstructed images were analyzed by CHOs with both rotationally symmetric (RS) and rotationally oriented (RO) channels, and with different numbers of lesion locations (5, 10, and 20) in a signal known exactly (SKE), background known exactly but variable (BKEV) detection task. The area under the receiver operating characteristic curve (AUC) was used as a summary measure to compare the IR and analytic methods; the AUC was also used as the equal performance criterion to derive the potential dose reduction factor of IR. In general, there was a good agreement in the relative AUC values of different reconstruction methods using CHOs with RS and RO channels, although the CHO with RO channels achieved higher AUCs than RS channels. The improvement of IR over analytic methods depends on the dose level. The reference dose level D0 was based on a clinical low dose protocol, lower than the standard dose due to the use of IR methods. At 75% D0, the performance improvement was statistically significant (p < 0.05). The potential dose reduction factor also depended on the detection task. For the SKE/BKEV task involving 10 lesion locations, a dose reduction of at least 25% from D0 was achieved.

Task-based measures of image quality and their relation to radiation dose and patient risk

The theory of task-based assessment of image quality is reviewed in the context of imaging with ionizing radiation, and objective figures of merit (FOMs) for image quality are summarized. The variation of the FOMs with the task, the observer and especially with the mean number of photons recorded in the image is discussed. Then various standard methods for specifying radiation dose are reviewed and related to the mean number of photons in the image and hence to image quality. Current knowledge of the relation between local radiation dose and the risk of various adverse effects is summarized, and some graphical depictions of the tradeoffs between image quality and risk are introduced. Then various dose-reduction strategies are discussed in terms of their effect on task-based measures of image quality.

Molecular Imaging in the College of Optical Sciences - An Overview of Two Decades of Instrumentation Development

During the past two decades, researchers at the University of Arizona's Center for Gamma-Ray Imaging (CGRI) have explored a variety of approaches to gamma-ray detection, including scintillation cameras, solid-state detectors, and hybrids such as the intensified Quantum Imaging Device (iQID) configuration where a scintillator is followed by optical gain and a fast CCD or CMOS camera. We have combined these detectors with a variety of collimation schemes, including single and multiple pinholes, parallel-hole collimators, synthetic apertures, and anamorphic crossed slits, to build a large number of preclinical molecular-imaging systems that perform Single-Photon Emission Computed Tomography (SPECT), Positron Emission Tomography (PET), and X-Ray Computed Tomography (CT). In this paper, we discuss the themes and methods we have developed over the years to record and fully use the information content carried by every detected gamma-ray photon.

Objective assessment of image quality VI: imaging in radiation therapy

Earlier work on objective assessment of image quality (OAIQ) focused largely on estimation or classification tasks in which the desired outcome of imaging is accurate diagnosis. This paper develops a general framework for assessing imaging quality on the basis of therapeutic outcomes rather than diagnostic performance. By analogy to receiver operating characteristic (ROC) curves and their variants as used in diagnostic OAIQ, the method proposed here utilizes the therapy operating characteristic or TOC curves, which are plots of the probability of tumor control versus the probability of normal-tissue complications as the overall dose level of a radiotherapy treatment is varied. The proposed figure of merit is the area under the TOC curve, denoted AUTOC. This paper reviews an earlier exposition of the theory of TOC and AUTOC, which was specific to the assessment of image-segmentation algorithms, and extends it to other applications of imaging in external-beam radiation treatment as well as in treatment with internal radioactive sources. For each application, a methodology for computing the TOC is presented. A key difference between ROC and TOC is that the latter can be defined for a single patient rather than a population of patients.

Numerical stability issues on channelized Hotelling observer under different background assumptions

This paper addresses the numerical stability issue on the channelized Hotelling observer (CHO). The CHO is a well-known approach in the medical image quality assessment domain. Many researchers have found that the detection performance of the CHO does not increase with the number of channels, contrary to expectation. And to our knowledge, nobody in this domain has found the reason. We illustrated that this is due to the ill-posed problem of the scatter matrix and proposed a solution based on Tikhonov regularization. Although Tikhonov regularization has been used in many other domains, we show in this paper another important application of Tikhonov regularization. This is very important for researchers to continue the CHO (and other channelized model observer) investigation with a reliable detection performance calculation.

Efficient determination of the uncertainty for the optimization of SPECT system design: a subsampled fisher information matrix

System designs in single photon emission tomography (SPECT) can be evaluated based on the fundamental trade-off between bias and variance that can be achieved in the reconstruction of emission tomograms. This trade off can be derived analytically using the Cramer-Rao type bounds, which imply the calculation and the inversion of the Fisher information matrix (FIM). The inverse of the FIM expresses the uncertainty associated to the tomogram, enabling the comparison of system designs. However, computing, storing and inverting the FIM is not practical with 3-D imaging systems. In order to tackle the problem of the computational load in calculating the inverse of the FIM, a method based on the calculation of the local impulse response and the variance, in a single point, from a single row of the FIM, has been previously proposed for system design. However this approximation (circulant approximation) does not capture the global interdependence between the variables in shift-variant systems such as SPECT, and cannot account e.g., for data truncation or missing data. Our new formulation relies on subsampling the FIM. The FIM is calculated over a subset of voxels arranged in a grid that covers the whole volume. Every element of the FIM at the grid points is calculated exactly, accounting for the acquisition geometry and for the object. This new formulation reduces the computational complexity in estimating the uncertainty, but nevertheless accounts for the global interdependence between the variables, enabling the exploration of design spaces hindered by the circulant approximation. The graphics processing unit accelerated implementation of the algorithm reduces further the computation times, making the algorithm a good candidate for real-time optimization of adaptive imaging systems. This paper describes the subsampled FIM formulation and implementation details. The advantages and limitations of the new approximation are explored, in comparison with the circulant approximation, in the context of design optimization of a parallel-hole collimator SPECT system and of an adaptive imaging system (similar to the commercially available D-SPECT).

Channelized model observer for the detection and estimation of signals with unknown amplitude, orientation, and size

As a task-based approach for medical image quality assessment, model observers (MOs) have been proposed as surrogates for human observers. While most MOs treat only signal-known-exactly tasks, there are few studies on signal-known-statistically (SKS) MOs, which are clinically more relevant. In this paper, we present a new SKS MO named channelized joint detection and estimation observer (CJO), capable of detecting and estimating signals with unknown amplitude, orientation, and size. We evaluate its estimation and detection performance using both synthesized (correlated Gaussian) backgrounds and real clinical (magnetic resonance) backgrounds. The results suggest that the CJO has good performance in the SKS detection-estimation task.

Limits of Spatial Resolution for Thermography and Other Non-destructive Imaging Methods Based on Diffusion Waves

In this work the measured variable, such as temperature, is a random variable showing fluctuations. The loss of information caused by diffusion waves in non-destructive testing can be described by stochastic processes. In non-destructive imaging, the information about the spatial pattern of a samples interior has to be transferred to the sample surface by certain waves, e.g., thermal waves. At the sample surface these waves can be detected and the interior structure is reconstructed from the measured signals. The amount of information about the interior of the sample, which can be gained from the detected waves on the sample surface, is essentially influenced by the propagation from its excitation to the surface. Diffusion causes entropy production and information loss for the propagating waves. Mandelis has developed a unifying framework for treating diverse diffusion-related periodic phenomena under the global mathematical label of diffusion-wave fields, such as thermal waves. Thermography uses the time-dependent diffusion of heat (either pulsed or modulated periodically) which goes along with entropy production and a loss of information. Several attempts have been made to compensate for this diffusive effect to get a higher resolution for the reconstructed images of the samples interior. In this work it is shown that fluctuations limit this compensation. Therefore, the spatial resolution for non-destructive imaging at a certain depth is also limited by theory.

Design and performance of a small-animal imaging system using synthetic collimation

This work outlines the design and construction of a single-photon emission computed tomography imaging system based on the concept of synthetic collimation. A focused multi-pinhole collimator is constructed using rapid-prototyping and casting techniques. The collimator projects the centre of the field of view (FOV) through 46 pinholes when the detector is adjacent to the collimator, with the number reducing towards the edge of the FOV. The detector is then moved further from the collimator to increase the magnification of the system. The object distance remains constant, and each new detector distance is a new system configuration. The level of overlap of the pinhole projections increases as the system magnification increases, but the number of projections subtended by the detector is reduced. There is no rotation in the system; a single tomographic angle is used in each system configuration. Image reconstruction is performed using maximum-likelihood expectation-maximization and an experimentally measured system matrix. The system matrix is measured for each configuration on a coarse grid, using a point source. The pinholes are individually identified and tracked, and a Gaussian fit is made to each projection. The coefficients of these fits are used to interpolate the system matrix. The system is validated experimentally with a hot-rod phantom. The Fourier crosstalk matrix is also measured to provide an estimate of the average spatial resolution along each axis over the entire FOV. The 3D synthetic-collimator image is formed by estimating the activity distribution within the FOV and summing the activities in the voxels along the axis perpendicular to the collimator face.

Objective assessment of sonographic quality I: task information

In this paper, we explore relationships between the performance of the ideal observer and information-based measures of class separability in the context of sonographic breast-lesion diagnosis. This investigation was motivated by a finding that, since the test statistic of the ideal observer in sonography is a quadratic function of the echo data, it is not generally normally distributed. We found for some types of boundary discrimination tasks often required for sonographic lesion diagnosis, the deviation of the test statistic from a normal distribution can be significant. Hence the usual relationships between performance and information metrics become uncertain. Using Monte Carlo studies involving five common sonographic lesion-discrimination tasks, we found in each case that the detectability index d(A)(2) from receiver operating characteristic analysis was well approximated by the Kullback-Leibler divergence J, a measure of clinical task information available from the recorded radio-frequency echo data. However, the lesion signal-to-noise ratio, SNR(I)(2), calculated from moments of the ideal observer test statistic, consistently underestimates d(A)(2) for high-contrast boundary discrimination tasks. Thus, in a companion paper, we established a relationship between image-quality properties of the imaging system and J in order to predict ideal performance. These relationships provide a rigorous basis for sonographic instrument evaluation and design.

Objective assessment of sonographic: quality II acquisition information spectrum

This paper describes a task-based, information-theoretic approach to the assessment of image quality in diagnostic sonography. We expand the Kullback-Leibler divergence metric J, which quantifies the diagnostic information contained within recorded radio-frequency echo signals, into a spatial-frequency integral comprised of two spectral components: one describes patient features for low-contrast diagnostic tasks and the other describes instrumentation properties. The latter quantity is the acquisition information spectrum (AIS), which measures the density of object information that an imaging system is able to transfer to the echo data at each spatial frequency. AIS is derived based on unique properties of acoustic scattering in tissues that generate object contrast. Predictions made by the J integral expression were validated through Monte Carlo studies using echo-signal data from simulated lesions. Our analysis predicts the diagnostic performance of any sonographic system at specific diagnostic tasks based on engineering properties of the instrument that constitute image quality.

Quantifying admissible undersampling for sparsity-exploiting iterative image reconstruction in X-ray CT

Iterative image reconstruction with sparsity-exploiting methods, such as total variation (TV) minimization, investigated in compressive sensing claim potentially large reductions in sampling requirements. Quantifying this claim for computed tomography (CT) is nontrivial, because both full sampling in the discrete-to-discrete imaging model and the reduction in sampling admitted by sparsity-exploiting methods are ill-defined. The present article proposes definitions of full sampling by introducing four sufficient-sampling conditions (SSCs). The SSCs are based on the condition number of the system matrix of a linear imaging model and address invertibility and stability. In the example application of breast CT, the SSCs are used as reference points of full sampling for quantifying the undersampling admitted by reconstruction through TV-minimization. In numerical simulations, factors affecting admissible undersampling are studied. Differences between few-view and few-detector bin reconstruction as well as a relation between object sparsity and admitted undersampling are quantified.