Uncertain prediction of deformable image registration on lung CT using multi-category features and supervised learning

The assessment of deformable registration uncertainty is an important task for the safety and reliability of registration methods in clinical applications. However, it is typically done by a manual and time-consuming procedure. We propose a novel automatic method to predict registration uncertainty based on multi-category features and supervised learning. Three types of features, including deformation field statistical features, deformation field physiologically realistic features, and image similarity features, are introduced and calculated to train the random forest regressor for local registration uncertain prediction. Deformation field statistical features represent the numerical stability of registration optimization, which are correlated to the uncertainty of deformation fields; deformation field physiologically realistic features represent the biomechanical properties of organ motions, which mathematically reflect the physiological reality of deformation; image similarity features reflect the similarity between the warped image and fixed image. The multi-category features comprehensively reflect the registration uncertainty. The strategy of spatial adaptive random perturbations is also introduced to accurately simulate spatial distribution of registration uncertainty, which makes deformation field statistical features more discriminative to the uncertainty of deformation fields. Experiments were conducted on three publicly available thoracic CT image datasets. Seventeen randomly selected image pairs are used to train the random forest model, and 9 image pairs are used to evaluate the prediction model. The quantitative experiments on lung CT images show that the proposed method outperforms the baseline method for uncertain prediction of classical iterative optimization-based registration and deep learning-based registration with different registration qualities. The proposed method achieves good performance for registration uncertain prediction, which has great potential in improving the accuracy of registration uncertain prediction.

A review of fake news detection approaches: A critical analysis of relevant studies and highlighting key challenges associated with the dataset, feature representation, and data fusion

Currently, social networks have become the main source to acquire news about current global affairs. However, fake news appears and spreads on social media daily. This disinformation has a negative influence on several domains, such as politics, the economy, and health. In addition, it further generates detriments to societal stability. Several studies have provided effective models for detecting fake news in social networks through a variety of methods; however, there are limitations. Furthermore, since it is a critical field, the accuracy of the detection models was found to be notably insufficient. Although many review articles have addressed the repercussions of fake news, most have focused on specific and recurring aspects of fake news detection models. For example, the majority of reviews have primarily focused on dividing datasets, features, and classifiers used in this field by type. The limitations of the datasets, their features, how these features are fused, and the impact of all these factors on detection models were not investigated, especially since most detection models were based on a supervised learning approach. This review article analyzes relevant studies for the few last years and highlights the challenges faced by fake news detection models and their impact on their performance. The investigation of fake news detection studies relied on the following aspects and their impact on detection accuracy, namely datasets, overfitting/underfitting, image-based features, feature vector representation, machine learning models, and data fusion. Based on the analysis of relevant studies, the review showed that these issues significantly affect the performance and accuracy of detection models. This review aims to provide room for other researchers in the future to improve fake news detection models.

The perception of translucency from surface gloss

Translucent objects (like fruit and wax) reflect and transmit incident light to generate complex retinal image structure. Understanding how we visually perceive translucency from these images is challenging, but previous studies have demonstrated that perceived shape and shading is important for perceiving translucency. We considered the possibility that perceived translucency might also depend on 3D shape inferred from surface gloss (i.e., shape from specular highlights). Here, we performed experiments to test whether interactions between specular and non-specular image properties generated by different 3D shape information influences perceived translucency. Results revealed that perceived translucency could be explained by incongruence in 3D shape used to generate specular and non-specular image components. We proposed a new computational model based on measurable image features informative of shading relative to specular highlights that accounted for 59% of the variability in judgments of perceived translucency from the result of 10-fold cross validation. This model was found to outperform other models based on explicit subjective measures of perceived surface shape, suggesting it implicitly taps much of the relevant geometric information necessary for predicting observer judgments of translucency for glossy materials. These results provide new insight into how the visual system might infer translucency from the structure of specular and non-specular shading generated by glossy semi-opaque materials.

Unilateral absence of pulmonary artery analysis based on echocardiographic feature

The unilateral absence of pulmonary artery (UAPA) is a rare congenital cardiovascular malformation, which is asymptomatic and easy to be ignored in early stage. A large number of complications may occur in the later stage. Therefore, early diagnosis and treatment is of great significance. The imaging data of 49 patients with UAPA discovered and confirmed clinically by the echocardiography in our hospital are analyzed. The results show that left pulmonary artery absence is more common (55%) and most of them are associated with other cardiovascular malformations (92%). Atrial septal defect and patent foramen ovale were most common in 56%. In which the absence of isolated pulmonary artery was 8% (4/49), and the absence of right pulmonary artery was 75% (3/4). Especially in the patients with tetralogy of Fallot, 77% (5/9) of them miss the diagnosis of UAPA. This suggests that doctors and sonographers should pay attention to the development of pulmonary artery bifurcation and left and right branches in multi-section, and strengthen the scanning of short axis section of high large artery.

18F-FDG PET/CT imaging findings in anaplastic large cell lymphoma, a rare subtype of lymphoma

OBJECTIVE

To investigate the 18F-FDG PET/CT imaging manifestations for anaplastic large cell lymphoma (ALCL), a rare subtype of T/NK cell lymphoma.

METHODS

Fifty patients with ALCL, including 32 anaplastic lymphoma kinase (ALK)-positive patients and 18 ALK-negative patients, were enrolled. The positive detection, maximal standardized uptake value (SUVmax), and distribution of nodal and extranodal involvement were recorded and analysed. Fifty patients with diffuse large B cell lymphoma (DLBCL) were collected as a control group.

RESULTS

ALCL lesions were demonstrated to be 18F-FDG-avid tumours with a mean SUVmax of 19.4 ± 12.6. Most (76%) ALCL patients presented with stage III-IV disease, and nodal and extranodal involvement occurred in 74.0 and 72.0% of the patients, respectively. ALCL and DLBCL showed many similarities in tumour stage, 18F-FDG uptake and tumour involvement (P > 0.05), although the preferred extranodal organs of involvement (bone and the gastrointestinal tract, respectively) were different (P < 0.05). Compared to ALK-negative lesions, a higher uptake of 18F-FDG was found in the ALK-positive lesions (SUVmax: 22.1 ± 14.3 vs. 15.1 ± 6.6, t = 2.354, P = 0.023). ALK-positive ALCL was more likely to involve the lymph nodes than ALK-negative ALCL (84.3% vs. 55.5%, χ2 = 4.973, P = 0.043), while ALK-negative ALCL was more prone to involve the extranodal organs compared to ALK-positive ALCL (88.9% vs. 62.5%, χ2 = 3.979, P = 0.046).

CONCLUSION

The present study demonstrated that ALCL is a systemic 18F-FDG-avid lymphoma with many imaging manifestations similar to DLBCL on PET/CT. The present study also showed that ALK expression actually influenced tumour 18F-FDG uptake and lesion distribution. These findings may be useful to improve the understanding of the biological characteristics of ALCL.

An automatic restoration framework based on GPU-accelerated collateral filtering in brain MR images

BACKGROUND

Image restoration is one of the fundamental and essential tasks within image processing. In medical imaging, developing an effective algorithm that can automatically remove random noise in brain magnetic resonance (MR) images is challenging. The collateral filter has been shown a more powerful algorithm than many existing methods. However, the computation of the collateral filter is more time-consuming and the selection of the filter parameters is also laborious. This paper proposes an automatic noise removal system based on the accelerated collateral filter for brain MR images.

METHODS

To solve these problems, we first accelerated the collateral filter with parallel computing using the graphics processing unit (GPU) architecture. We adopted the compute unified device architecture (CUDA), an application programming interface for the GPU by NVIDIA, to hasten the computation. Subsequently, the optimal filter parameters were selected and the automation was achieved by artificial neural networks. Specifically, an artificial neural network system associated with image feature analysis was adopted to establish the automatic image restoration framework. The best feature combination was selected by the paired t-test and the sequential forward floating selection (SFFS) methods.

RESULTS

Experimental results indicated that not only did the proposed automatic image restoration algorithm perform dramatically faster than the traditional collateral filter, but it also effectively removed the noise in a wide variety of brain MR images. A speed up gain of 34 was attained to process an MR image, which completed within 0.1 s. Representative illustrations of brain tumor images demonstrated the capability of identifying lesion boundaries, which outperformed many existing methods.

CONCLUSIONS

We believe that our accelerated and automated restoration framework is promising for achieving robust filtering in many brain MR image restoration applications.

CT-based radiomic model predicts high grade of clear cell renal cell carcinoma

PURPOSE

To compare the predictive models that can incorporate a set of CT image features for preoperatively differentiating the high grade (Fuhrman III-IV) from low grade (Fuhrman I-II) clear cell renal cell carcinoma (ccRCC).

MATERIAL AND METHODS

One hundred and fourteen patients with ccRCC treated with a partial or radical nephrectomy were enrolled in the training cohort. The six non-texture features, including Pseudocapsule, Round mass, maximal tumor diameter (Diameter_max), intratumoral artery (Artery_tumor), enhancement value of the tumor (TEV) and relative TEV (rTEV), were assessed for each tumor. The texture features were extracted from the CT images of the section with the largest area of renal mass at both corticomedullary and nephrographic phases. The least absolute shrinkage and selection operator (LASSO) was used to screen the most valuable texture features to calculate a texture score (Texture-score) for each patient. A logistic regression model was used in the training cohort to discriminate the high from low grade ccRCC at nephrectomy. The predictors would include all non-texture features in Model 1, all non-texture features and Texture-score in Model 2, and Texture-score in Model 3. The performance of the predictive models were tested and compared in an independent validation cohort composed of 92 cases with ccRCC.

RESULTS

Inter-rater agreement was good for each non-texture feature and Texture-score (the concordance correlation coefficient or Kappa coefficient > 0.70). The Texture-score was calculated via a linear combination of the 4 selected texture features. The three models shown good discrimination of the high from low grade ccRCC in the training cohort and the area under receiver operating characteristic curve (AUC) was 0.826 in Mode 1, 0.878 in Model 2 and 0.843 in Model 3, and a significant different AUC was found between Model 1 and Model 2. Application of the predictive models in the validation cohort still gave a discrimination (AUC > 0.670), and the Texture-score based models with or without the non-texture features (Model 2 and 3) showed a better discrimination of the high from low grade ccRCC (P < 0.05).

CONCLUSION

This study presented the Texture-score based models can facilitate the preoperative discrimination of the high from low grade ccRCC.

Structural and advanced imaging in predicting MGMT promoter methylation of primary glioblastoma: a region of interest based analysis

Background

The methylation status of oxygen 6-methylguanine-DNA methyltransferase (MGMT) promoter has been associated with treatment response in glioblastoma(GBM). Using pre-operative MRI techniques to predict MGMT promoter methylation status remains inconclusive. In this study, we investigated the value of features from structural and advanced imagings in predicting the methylation of MGMT promoter in primary glioblastoma patients.

Methods

Ninety-two pathologically confirmed primary glioblastoma patients underwent preoperative structural MR imagings and the efficacy of structural image features were qualitatively analyzed using Fisher’s exact test. In addition, 77 of the 92 patients underwent additional advanced MRI scans including diffusion-weighted (DWI) and 3-diminsional pseudo-continuous arterial spin labeling (3D pCASL) imaging. Apparent diffusion coefficient (ADC) and relative cerebral blood flow (rCBF) values within the manually drawn region-of-interest (ROI) were calculated and compared using independent sample t test for their efficacies in predicting MGMT promoter methylation. Receiver operating characteristic curve (ROC) analysis was used to investigate the predicting efficacy with the area under the curve (AUC) and cross validations. Multiple-variable logistic regression model was employed to evaluate the predicting performance of multiple variables.

Results

MGMT promoter methylation was associated with tumor location and necrosis (P <  0.05). Significantly increased ADC value (P <  0.001) and decreased rCBF (P <  0.001) were associated with MGMT promoter methylation in primary glioblastoma. The ADC achieved the better predicting efficacy than rCBF (ADC: AUC, 0.860; sensitivity, 81.1%; specificity, 82.5%; vs rCBF: AUC, 0.835; sensitivity, 75.0%; specificity, 78.4%; P = 0.032). The combination of tumor location, necrosis, ADC and rCBF resulted in the highest AUC of 0.914.

Conclusion

ADC and rCBF are promising imaging biomarkers in clinical routine to predict the MGMT promoter methylation in primary glioblastoma patients.

Automatic brain MR image denoising based on texture feature-based artificial neural networks

Noise is one of the main sources of quality deterioration not only for visual inspection but also in computerized processing in brain magnetic resonance (MR) image analysis such as tissue classification, segmentation and registration. Accordingly, noise removal in brain MR images is important for a wide variety of subsequent processing applications. However, most existing denoising algorithms require laborious tuning of parameters that are often sensitive to specific image features and textures. Automation of these parameters through artificial intelligence techniques will be highly beneficial. In the present study, an artificial neural network associated with image texture feature analysis is proposed to establish a predictable parameter model and automate the denoising procedure. In the proposed approach, a total of 83 image attributes were extracted based on four categories: 1) Basic image statistics. 2) Gray-level co-occurrence matrix (GLCM). 3) Gray-level run-length matrix (GLRLM) and 4) Tamura texture features. To obtain the ranking of discrimination in these texture features, a paired-samples t-test was applied to each individual image feature computed in every image. Subsequently, the sequential forward selection (SFS) method was used to select the best texture features according to the ranking of discrimination. The selected optimal features were further incorporated into a back propagation neural network to establish a predictable parameter model. A wide variety of MR images with various scenarios were adopted to evaluate the performance of the proposed framework. Experimental results indicated that this new automation system accurately predicted the bilateral filtering parameters and effectively removed the noise in a number of MR images. Comparing to the manually tuned filtering process, our approach not only produced better denoised results but also saved significant processing time.

An evolution of image source camera attribution approaches

Camera attribution plays an important role in digital image forensics by providing the evidence and distinguishing characteristics of the origin of the digital image. It allows the forensic analyser to find the possible source camera which captured the image under investigation. However, in real-world applications, these approaches have faced many challenges due to the large set of multimedia data publicly available through photo sharing and social network sites, captured with uncontrolled conditions and undergone variety of hardware and software post-processing operations. Moreover, the legal system only accepts the forensic analysis of the digital image evidence if the applied camera attribution techniques are unbiased, reliable, nondestructive and widely accepted by the experts in the field. The aim of this paper is to investigate the evolutionary trend of image source camera attribution approaches from fundamental to practice, in particular, with the application of image processing and data mining techniques. Extracting implicit knowledge from images using intrinsic image artifacts for source camera attribution requires a structured image mining process. In this paper, we attempt to provide an introductory tutorial on the image processing pipeline, to determine the general classification of the features corresponding to different components for source camera attribution. The article also reviews techniques of the source camera attribution more comprehensively in the domain of the image forensics in conjunction with the presentation of classifying ongoing developments within the specified area. The classification of the existing source camera attribution approaches is presented based on the specific parameters, such as colour image processing pipeline, hardware- and software-related artifacts and the methods to extract such artifacts. The more recent source camera attribution approaches, which have not yet gained sufficient attention among image forensics researchers, are also critically analysed and further categorised into four different classes, namely, optical aberrations based, sensor camera fingerprints based, processing statistics based and processing regularities based, to present a classification. Furthermore, this paper aims to investigate the challenging problems, and the proposed strategies of such schemes based on the suggested taxonomy to plot an evolution of the source camera attribution approaches with respect to the subjective optimisation criteria over the last decade. The optimisation criteria were determined based on the strategies proposed to increase the detection accuracy, robustness and computational efficiency of source camera brand, model or device attribution.