Geometry-based vs. intensity-based medical image registration: A comparative study on 3D CT data

A Novel Approach for Estimating Ovarian Cancer Tissue Heterogeneity through the Application of Image Processing Techniques and Artificial Intelligence

PURPOSE

Tumor heterogeneity may be responsible for poor response to treatment and adverse prognosis in women with HGOEC. The purpose of this study is to propose an automated classification system that allows medical experts to automatically identify intratumoral areas of different cellularity indicative of tumor heterogeneity.

METHODS

Twenty-two patients underwent dedicated pelvic MRI, and a database of 11,095 images was created. After image processing techniques were applied to align and assess the cancerous regions, two specific imaging series were used to extract quantitative features (radiomics). These features were employed to create, through artificial intelligence, an estimator of the highly cellular intratumoral area as defined by arbitrarily selected apparent diffusion coefficient (ADC) cut-off values (ADC < 0.85 × 10-3 mm2/s).

RESULTS

The average recorded accuracy of the proposed automated classification system was equal to 0.86.

CONCLUSION

The proposed classification system for assessing highly cellular intratumoral areas, based on radiomics, may be used as a tool for assessing tumor heterogeneity.

Image Registration for Visualizing Magnetic Flux Leakage Testing under Different Orientations of Magnetization

The Magnetic Flux Leakage (MFL) visualization technique is widely used in the surface defect inspection of ferromagnetic materials. However, the information of the images detected through the MFL method is incomplete when the defect (especially for the cracks) is complex, and some information would be lost when magnetized unidirectionally. Then, the multidirectional magnetization method is proposed to fuse the images detected under different magnetization orientations. It causes a critical problem: the existing image registration methods cannot be applied to align the images because the images are different when detected under different magnetization orientations. This study presents a novel image registration method for MFL visualization to solve this problem. In order to evaluate the registration, and to fuse the information detected in different directions, the mutual information between the reference image and the MFL image calculated by the forward model is designed as a measure. Furthermore, Particle Swarm Optimization (PSO) is used to optimize the registration process. The comparative experimental results demonstrate that this method has a higher registration accuracy for the MFL images of complex cracks than the existing methods.

Reattachable fiducial skin marker for automatic multimodality registration

Purpose

Fusing image information has become increasingly important for optimal diagnosis and treatment of the patient. Despite intensive research towards markerless registration approaches, fiducial marker-based methods remain the default choice for a wide range of applications in clinical practice. However, as especially non-invasive markers cannot be positioned reproducibly in the same pose on the patient, pre-interventional imaging has to be performed immediately before the intervention for fiducial marker-based registrations.

Methods

We propose a new non-invasive, reattachable fiducial skin marker concept for multi-modal registration approaches including the use of electromagnetic or optical tracking technologies. We furthermore describe a robust, automatic fiducial marker localization algorithm for computed tomography (CT) and magnetic resonance imaging (MRI) images. Localization of the new fiducial marker has been assessed for different marker configurations using both CT and MRI. Furthermore, we applied the marker in an abdominal phantom study. For this, we attached the marker at three poses to the phantom, registered ten segmented targets of the phantom’s CT image to live ultrasound images and determined the target registration error (TRE) for each target and each marker pose.

Results

Reattachment of the marker was possible with a mean precision of 0.02 mm ± 0.01 mm. Our algorithm successfully localized the marker automatically in all ($$n=201$$ n = 201 ) evaluated CT/MRI images. Depending on the marker pose, the mean ($$n=10$$ n = 10 ) TRE of the abdominal phantom study ranged from 1.51 ± 0.75 mm to 4.65 ± 1.22 mm.

Conclusions

The non-invasive, reattachable skin marker concept allows reproducible positioning of the marker and automatic localization in different imaging modalities. The low TREs indicate the potential applicability of the marker concept for clinical interventions, such as the puncture of abdominal lesions, where current image-based registration approaches still lack robustness and existing marker-based methods are often impractical.

Probabilistic Learning Coherent Point Drift for 3D Ultrasound Fetal Head Registration

Quantification of brain growth is crucial for the assessment of fetal well being, for which ultrasound (US) images are the chosen clinical modality. However, they present artefacts, such as acoustic occlusion, especially after the 18^th gestational week, when cranial calcification appears. Fetal US volume registration is useful in one or all of the following cases: to monitor the evolution of fetometry indicators, to segment different structures using a fetal brain atlas, and to align and combine multiple fetal brain acquisitions. This paper presents a new approach for automatic registration of real 3D US fetal brain volumes, volumes that contain a considerable degree of occlusion artefacts, noise, and missing data. To achieve this, a novel variant of the coherent point drift method is proposed. This work employs supervised learning to segment and conform a point cloud automatically and to estimate their subsequent weight factors. These factors are obtained by a random forest-based classification and are used to appropriately assign nonuniform membership probability values of a Gaussian mixture model. These characteristics allow for the automatic registration of 3D US fetal brain volumes with occlusions and multiplicative noise, without needing an initial point cloud. Compared to other intensity and geometry-based algorithms, the proposed method achieves an error reduction of 7.4% to 60.7%, with a target registration error of only 6.38 ± 3.24 mm. This makes the herein proposed approach highly suitable for 3D automatic registration of fetal head US volumes, an approach which can be useful to monitor fetal growth, segment several brain structures, or even compound multiple acquisitions taken from different projections.

Quality of Service aware Medical CT Image Transmission Anti-collision Mechanism Based on Big Data Autonomous Anti-collision Control

Background:

At present, due to the limitation of hardware, software and network transmission performance, the medical diagnosis of medical CT image equipment is easy to be carried out based on the wrong image. In addition, due to the complex structure of human organs and unpredictable lesion location, it is difficult to judge the reliability of medical CT images, spatial localization of the lesion, two-dimensional slice images and shape based on stereotypes. Therefore, how to improve the efficiency of medical CT terminal and the image quality has become the key technology to improve the satisfaction of medical diagnosis and treatment.

Objective:

To improve the work efficiency of medical CT terminal and medical image transmission quality, with the medical CT terminal state and service quality.

Methods:

Firstly, from the view of throughput, packet loss rate, delay and so on, a QoS aware model for medical CT image transmission has been established. Then, with throughput, packet length, path loss, service area size, access point location, and the number of medical CT terminals, the performance change regulation of the medical CT image transmission is completed and the optimal quality of service guarantee parameters sequence is obtained. Next, the medical CT image big data autonomous collision control scheme is proposed.

Results:

The experimental and mathematical results verify the real-time performance, reliability, effectiveness and feasibility of the proposed medical CT image transmission anti-collision mechanism.

Conclusion:

The proposed scheme can satisfy the high-quality high demand for data transmission at the same time, according to a variety of user experience demand and real-time adjustment of medical CT terminal working state, which provides effective data quality assurance and optimization of the network source distribution, and also enhances the quality of medical image data transmission service.

Intensity-Assisted ICP for Fast Registration of 2D-LIDAR

Iterative closest point (ICP) is a method commonly used to perform scan-matching and registration. To be a simple and robust algorithm, it is still computationally expensive, and it has been regarded as having a crucial challenge especially in a real-time application as used for the simultaneous localization and mapping (SLAM) problem. For these reasons, this paper presents a new method for the acceleration of ICP with an assisted intensity. Unlike the conventional ICP, this method is proposed to reduce the computational cost and avoid divergences. An initial transformation guess is computed with an assisted intensity for their relative rigid-body transformation. Moreover, a target function is proposed to determine the best initial transformation guess based on the statistic of their spatial distances and intensity residuals. Additionally, this method is also proposed to reduce the iteration number. The Anderson acceleration is utilized for increasing the iteration speed which has better ability than the Picard iteration procedure. The proposed algorithm is operated in real time with a single core central processing unit (CPU) thread. Hence, it is suitable for the robot which has limited computation resources. To validate the novelty, this proposed method is evaluated on the SEMANTIC3D.NET benchmark dataset. According to comparative results, the proposed method is declared as having better accuracy and robustness than the conventional ICP methods.

Spotting L3 slice in CT scans using deep convolutional network and transfer learning

In this article, we present a complete automated system for spotting a particular slice in a complete 3D Computed Tomography exam (CT scan). Our approach does not require any assumptions on which part of the patient's body is covered by the scan. It relies on an original machine learning regression approach. Our models are learned using the transfer learning trick by exploiting deep architectures that have been pre-trained on imageNet database, and therefore it requires very little annotation for its training. The whole pipeline consists of three steps: i) conversion of the CT scans into Maximum Intensity Projection (MIP) images, ii) prediction from a Convolutional Neural Network (CNN) applied in a sliding window fashion over the MIP image, and iii) robust analysis of the prediction sequence to predict the height of the desired slice within the whole CT scan. Our approach is applied to the detection of the third lumbar vertebra (L3) slice that has been found to be representative to the whole body composition. Our system is evaluated on a database collected in our clinical center, containing 642 CT scans from different patients. We obtained an average localization error of 1.91±2.69 slices (less than 5 mm) in an average time of less than 2.5 s/CT scan, allowing integration of the proposed system into daily clinical routines.

A Registration Method Based on Contour Point Cloud for 3D Whole-Body PET and CT Images

The PET and CT fusion image, combining the anatomical and functional information, has important clinical meaning. An effective registration of PET and CT images is the basis of image fusion. This paper presents a multithread registration method based on contour point cloud for 3D whole-body PET and CT images. Firstly, a geometric feature-based segmentation (GFS) method and a dynamic threshold denoising (DTD) method are creatively proposed to preprocess CT and PET images, respectively. Next, a new automated trunk slices extraction method is presented for extracting feature point clouds. Finally, the multithread Iterative Closet Point is adopted to drive an affine transform. We compare our method with a multiresolution registration method based on Mattes Mutual Information on 13 pairs (246~286 slices per pair) of 3D whole-body PET and CT data. Experimental results demonstrate the registration effectiveness of our method with lower negative normalization correlation (NC = −0.933) on feature images and less Euclidean distance error (ED = 2.826) on landmark points, outperforming the source data (NC = −0.496, ED = 25.847) and the compared method (NC = −0.614, ED = 16.085). Moreover, our method is about ten times faster than the compared one.