Streak artefact removal in x-ray dark-field computed tomography using a convolutional neural network.

BACKGROUND: Computed tomography (CT) relies on the attenuation of x-rays, and is, hence, of limited use for weakly attenuating organs of the body, such as the lung. X-ray dark-field (DF) imaging is a recently developed technology that utilizes x-ray optical gratings to enable small-angle scattering as an alternative contrast mechanism. The DF signal provides structural information about the micromorphology of an object, complementary to the conventional attenuation signal. A first human-scale x-ray DF CT has been developed by our group. Despite specialized processing algorithms, reconstructed images remain affected by streaking artifacts, which often hinder image interpretation. In recent years, convolutional neural networks have gained popularity in the field of CT reconstruction, amongst others for streak artefact removal. PURPOSE: Reducing streak artifacts is essential for the optimization of image quality in DF CT, and artefact free images are a prerequisite for potential future clinical application. The purpose of this paper is to demonstrate the feasibility of CNN post-processing for artefact reduction in x-ray DF CT and how multi-rotation scans can serve as a pathway for training data. METHODS: We employed a supervised deep-learning approach using a three-dimensional dual-frame UNet in order to remove streak artifacts. Required training data were obtained from the experimental x-ray DF CT prototype at our institute. Two different operating modes were used to generate input and corresponding ground truth data sets. Clinically relevant scans at dose-compatible radiation levels were used as input data, and extended scans with substantially fewer artifacts were used as ground truth data. The latter is neither dose-, nor time-compatible and, therefore, unfeasible for clinical imaging of patients. RESULTS: The trained CNN was able to greatly reduce streak artifacts in DF CT images. The network was tested against images with entirely different, previously unseen image characteristics. In all cases, CNN processing substantially increased the image quality, which was quantitatively confirmed by increased image quality metrics. Fine details are preserved during processing, despite the output images appearing smoother than the ground truth images. CONCLUSIONS: Our results showcase the potential of a neural network to reduce streak artifacts in x-ray DF CT. The image quality is successfully enhanced in dose-compatible x-ray DF CT, which plays an essential role for the adoption of x-ray DF CT into modern clinical radiology.

Improving image quality of sparse-view lung tumor CT images with U-Net.

BACKGROUND: We aimed to improve the image quality (IQ) of sparse-view computed tomography (CT) images using a U-Net for lung metastasis detection and determine the best tradeoff between number of views, IQ, and diagnostic confidence. METHODS: CT images from 41 subjects aged 62.8 ± 10.6 years (mean ± standard deviation, 23 men), 34 with lung metastasis, 7 healthy, were retrospectively selected (2016-2018) and forward projected onto 2,048-view sinograms. Six corresponding sparse-view CT data subsets at varying levels of undersampling were reconstructed from sinograms using filtered backprojection with 16, 32, 64, 128, 256, and 512 views. A dual-frame U-Net was trained and evaluated for each subsampling level on 8,658 images from 22 diseased subjects. A representative image per scan was selected from 19 subjects (12 diseased, 7 healthy) for a single-blinded multireader study. These slices, for all levels of subsampling, with and without U-Net postprocessing, were presented to three readers. IQ and diagnostic confidence were ranked using predefined scales. Subjective nodule segmentation was evaluated using sensitivity and Dice similarity coefficient (DSC); clustered Wilcoxon signed-rank test was used. RESULTS: The 64-projection sparse-view images resulted in 0.89 sensitivity and 0.81 DSC, while their counterparts, postprocessed with the U-Net, had improved metrics (0.94 sensitivity and 0.85 DSC) (p = 0.400). Fewer views led to insufficient IQ for diagnosis. For increased views, no substantial discrepancies were noted between sparse-view and postprocessed images. CONCLUSIONS: Projection views can be reduced from 2,048 to 64 while maintaining IQ and the confidence of the radiologists on a satisfactory level. RELEVANCE STATEMENT: Our reader study demonstrates the benefit of U-Net postprocessing for regular CT screenings of patients with lung metastasis to increase the IQ and diagnostic confidence while reducing the dose. KEY POINTS: • Sparse-projection-view streak artifacts reduce the quality and usability of sparse-view CT images. • U-Net-based postprocessing removes sparse-view artifacts while maintaining diagnostically accurate IQ. • Postprocessed sparse-view CTs drastically increase radiologists' confidence in diagnosing lung metastasis.

Feasibility of Dark-Field Radiography to Enhance Detection of Nondisplaced Fractures.

Background Many clinically relevant fractures are occult on conventional radiographs and therefore challenging to diagnose reliably. X-ray dark-field radiography is a developing method that uses x-ray scattering as an additional signal source. Purpose To investigate whether x-ray dark-field radiography enhances the depiction of radiographically occult fractures in an experimental model compared with attenuation-based radiography alone and whether the directional dependence of dark-field signal impacts observer ratings. Materials and Methods Four porcine loin ribs had nondisplaced fractures experimentally introduced. Microstructural changes were visually verified using high-spatial-resolution three-dimensional micro-CT. X-ray dark-field radiographs were obtained before and after fracture, with the before-fracture scans serving as control images. The presence of a fracture was scored by three observers using a six-point scale (6, surely; 5, very likely; 4, likely; 3, unlikely; 2, very unlikely; and 1, certainly not). Differences between scores based on attenuation radiographs alone (n = 96) and based on combined attenuation and dark-field radiographs (n = 96) were evaluated by using the DeLong method to compare areas under the receiver operating characteristic curve. The impact of the dark-field signal directional sensitivity on observer ratings was evaluated using the Wilcoxon test. The dark-field data were split into four groups (24 images per group) according to their sensitivity orientation and tested against each other. Musculoskeletal dark-field radiography was further demonstrated on human finger and foot specimens. Results The addition of dark-field radiographs was found to increase the area under the receiver operating characteristic curve to 1 compared with an area under the receiver operating characteristic curve of 0.87 (95% CI: 0.80, 0.94) using attenuation-based radiographs alone (P < .001). There were similar observer ratings for the four different dark-field sensitivity orientations (P = .16-.65 between the groups). Conclusion These results suggested that the inclusion of dark-field radiography has the potential to help enhance the detection of nondisplaced fractures compared with attenuation-based radiography alone. © RSNA, 2024 See also the editorial by Rubin in this issue.

Improving Automated Hemorrhage Detection at Sparse-View CT via U-Net-based Artifact Reduction.

Purpose To explore the potential benefits of deep learning-based artifact reduction in sparse-view cranial CT scans and its impact on automated hemorrhage detection. Materials and Methods In this retrospective study, a U-Net was trained for artifact reduction on simulated sparse-view cranial CT scans in 3000 patients, obtained from a public dataset and reconstructed with varying sparse-view levels. Additionally, EfficientNet-B2 was trained on full-view CT data from 17 545 patients for automated hemorrhage detection. Detection performance was evaluated using the area under the receiver operating characteristic curve (AUC), with differences assessed using the DeLong test, along with confusion matrices. A total variation (TV) postprocessing approach, commonly applied to sparse-view CT, served as the basis for comparison. A Bonferroni-corrected significance level of .001/6 = .00017 was used to accommodate for multiple hypotheses testing. Results Images with U-Net postprocessing were better than unprocessed and TV-processed images with respect to image quality and automated hemorrhage detection. With U-Net postprocessing, the number of views could be reduced from 4096 (AUC: 0.97 [95% CI: 0.97, 0.98]) to 512 (0.97 [95% CI: 0.97, 0.98], P < .00017) and to 256 views (0.97 [95% CI: 0.96, 0.97], P < .00017) with a minimal decrease in hemorrhage detection performance. This was accompanied by mean structural similarity index measure increases of 0.0210 (95% CI: 0.0210, 0.0211) and 0.0560 (95% CI: 0.0559, 0.0560) relative to unprocessed images. Conclusion U-Net-based artifact reduction substantially enhanced automated hemorrhage detection in sparse-view cranial CT scans. Keywords: CT, Head/Neck, Hemorrhage, Diagnosis, Supervised Learning Supplemental material is available for this article. © RSNA, 2024.

Spectral propagation-based x-ray phase-contrast computed tomography.

Purpose: Propagation-based x-ray imaging (PBI) is a phase-contrast technique that is employed in high-resolution imaging by introducing some distance between sample and detector. PBI causes characteristic intensity fringes that have to be processed with appropriate phase-retrieval algorithms, which has historically been a difficult task for objects composed of several different materials. Spectral x-ray imaging has been introduced to PBI to overcome this issue and to potentially utilize the spectral nature of the data for material-specific imaging. We aim to explore the potential of spectral PBI in three-dimensional computed tomography (CT) imaging in this work. Approach: We demonstrate phase-retrieval for experimental high-resolution spectral propagation-based CT data of a simple two-component sample, as well as a multimaterial capacitor test sample. Phase-retrieval was performed using an algorithm based on the Alvarez-Macovski model. Virtual monochromatic (VMI) and effective atomic number images were calculated after phase-retrieval. Results: Phase-retrieval results from the spectral data set show a distinct gray-level for each material with no residual phase-contrast fringes. Several representations of the phase-retrieved data are provided. The VMI is used to display an attenuation-equivalent image at a chosen display energy of 80 keV, to provide good separation of materials with minimal noise. The effective atomic number image shows the material composition of the sample. Conclusions: Spectral photon-counting detector technology has already been shown to be compatible with spectral PBI, and there is a foreseeable need for robust phase-retrieval in high-resolution, spectral x-ray CT in the future. Our results demonstrate the feasibility of phase-retrieval for spectral PBI CT.

Full Field X-Ray Scatter Tomography.

In X-ray imaging, photons are transmitted through and absorbed by the target object, but are also scattered in significant quantities. Previous attempts to use scattered X-ray photons for imaging applications used pencil or fan beam illumination. Here we present 3D X-ray Scatter Tomography using full-field illumination for small-animal imaging. Synchrotron imaging experiments were performed on a phantom and the chest of a juvenile rat. Transmitted and scattered photons were simultaneously imaged with separate cameras; a scientific camera directly downstream of the sample stage, and a pixelated detector with a pinhole imaging system placed at 45° to the beam axis. We obtained scatter tomogram feature fidelity sufficient for segmentation of the lungs and major airways in the rat. The image contrast in the scatter tomogram slices approached that of transmission imaging, indicating robustness to the amount of multiple scattering present in our case. This opens the possibility of augmenting full-field 2D imaging systems with additional scatter detectors to obtain complementary modes or to improve the fidelity of existing images without additional dose, potentially leading to single-shot or reduced-angle tomography or overall dose reduction for live animal studies.

Quantitative material decomposition using linear iterative near-field phase retrieval dual-energy x-ray imaging.

This paper expands the linear iterative near-field phase retrieval (LIPR) formalism to achieve quantitative material thickness decomposition. Propagation-based phase contrast x-ray imaging with subsequent phase retrieval has been shown to improve the signal-to-noise ratio (SNR) by factors of up to hundreds compared to conventional x-ray imaging. This is a key step in biomedical imaging, where radiation exposure must be kept low without compromising the SNR. However, for a satisfactory phase retrieval from a single measurement, assumptions must be made about the object investigated. To avoid such assumptions, we use two measurements collected at the same propagation distance but at different x-ray energies. Phase retrieval is then performed by incorporating the Alvarez-Macovski (AM) model, which models the x-ray interactions as being comprised of distinct photoelectric and Compton scattering components. We present the first application of dual-energy phase retrieval with the AM model to monochromatic experimental x-ray projections at two different energies for obtaining split x-ray interactions. Our phase retrieval method allows us to separate the object investigated into the projected thicknesses of two known materials. Our phase retrieval output leads to no visible loss in spatial resolution while the SNR improves by factors of 2 to 10. This corresponds to a possible x-ray dose reduction by a factor of 4 to 100, under the Poisson noise assumption.

Material Decomposition Using Spectral Propagation-Based Phase-Contrast X-Ray Imaging.

Material decomposition in X-ray imaging uses the energy-dependence of attenuation to digitally decompose an object into specific constituent materials, generally at the cost of enhanced image noise. Propagation-based X-ray phase-contrast imaging is a developing technique that can be used to reduce image noise, in particular from weakly attenuating objects. In this paper, we combine spectral phase-contrast imaging with material decomposition to both better visualize weakly attenuating features and separate them from overlying objects in radiography. We derive an algorithm that performs both tasks simultaneously and verify it against numerical simulations and experimental measurements of ideal two-component samples composed of pure aluminum and poly(methyl methacrylate). Additionally, we showcase first imaging results of a rabbit kitten's lung. The attenuation signal of a thorax, in particular, is dominated by the strongly attenuating bones of the ribcage. Combined with the weak soft tissue signal, this makes it difficult to visualize the fine anatomical structures across the whole lung. In all cases, clean material decomposition was achieved, without residual phase-contrast effects, from which we generate an un-obstructed image of the lung, free of bones. Spectral propagation-based phase-contrast imaging has the potential to be a valuable tool, not only in future lung research, but also in other systems for which phase-contrast imaging in combination with material decomposition proves to be advantageous.

Spectral x-ray imaging: Conditions under which propagation-based phase-contrast is beneficial.

Energy-resolved attenuation data in spectral x-ray imaging enables material decomposition, in which the different materials inside an object can be identified and separated virtually. Material decomposition has the drawback of increased noise in the resulting material images relative to the measured images. Recently, spectral x-ray imaging was combined with propagation-based x-ray phase-contrast imaging, an x-ray technique that has the potential to greatly reduce image noise by utilizing wave-optical effects. The net combined effects on image noise of performing spectral material decomposition with phase-contrast are not yet well understood, and we provide a detailed theoretical investigation of this topic here. In particular, we investigate how the addition of phase-contrast in spectral imaging affects material decomposition compared to using conventional spectral attenuation data. We show how the underlying equations can be rearranged into parts that resemble low- and high-pass filters on the input images, from which we are able to identify different energy-dependent cases where phase-contrast is or is not advantageous. Our results suggest that the benefits of phase-contrast in the context of material decomposition are primarily restricted to x-ray energies under a certain threshold, where that threshold depends on the given material combination, and sits in a region where photoelectric absorption dominates x-ray attenuation. Additionally, we show that decomposition of the electron density using an image basis spanned by functions of the Alvarez-Macovski model benefits from phase-contrast, regardless of the x-ray energies. All our findings are based purely on theoretical considerations, and can, therefore, be used to determine the feasibility and utility of propagation-based phase-contrast in spectral x-ray imaging ahead of any data collection.

Material decomposition from a single x-ray projection via single-grid phase contrast imaging.

This study describes a new approach for material decomposition in x-ray imaging, utilizing phase contrast both to increase sensitivity to weakly attenuating samples and to act as a complementary measurement to attenuation, therefore allowing two overlaid materials to be separated. The measurements are captured using the single-exposure, single-grid x-ray phase contrast imaging technique, with a novel correction that aims to remove propagation-based phase effects seen at sharp edges in the attenuation image. The use of a single-exposure technique means that images can be collected in a high-speed sequence. Results are shown for both a known two-material sample and for a biological specimen.

Photon-counting, energy-resolving and super-resolution phase contrast X-ray imaging using an integrating detector.

This work demonstrates the use of a scientific-CMOS (sCMOS) energy-integrating detector as a photon-counting detector, thereby eliminating dark current and read-out noise issues, that simultaneously provides both energy resolution and sub-pixel spatial resolution for X-ray imaging. These capabilities are obtained by analyzing visible light photon clouds that result when X-ray photons produce fluorescence from a scintillator in front of the visible light sensor. Using low-fluence monochromatic X-ray projections to avoid overlapping photon clouds, the centroid of individual X-ray photon interactions was identified. This enabled a tripling of the spatial resolution of the detector to 6.71 ± 0.04 µm. By calculating the total charge deposited by this interaction, an energy resolution of 61.2 ± 0.1% at 17 keV was obtained. When combined with propagation-based phase contrast imaging and phase retrieval, a signal-to-noise ratio of up to 15 ± 3 was achieved for an X-ray fluence of less than 3 photons/mm2.

Brain Connectivity Exposed by Anisotropic X-ray Dark-field Tomography.

To understand the interaction of different parts of the human brain it is essential to know how they are connected. Such connections are predominantly related to the brain's white matter, which forms the neuronal pathways, the axons. These axons, also referred to as nerve fibers, have a size on the micrometer scale and are therefore too small to be imaged by standard X-ray systems. In this paper, we use a grating interferometer and a method based on Anisotropic X-ray Dark-field Tomography (AXDT) with the goal to generate a three-dimensional tomographic reconstruction of these functional structures. A first preclinical survey shows that we successfully reconstruct the orientations of the brain fibers connectivity with our approach.

Simultaneous wood and metal particle detection on dark-field radiography.

BACKGROUND: Currently, the detection of retained wood is a frequent but challenging task in emergency care. The purpose of this study is to demonstrate improved foreign-body detection with the novel approach of preclinical X-ray dark-field radiography. METHODS: At a preclinical dark-field x-ray radiography, setup resolution and sensitivity for simultaneous detection of wooden and metallic particles have been evaluated in a phantom study. A clinical setting has been simulated with a formalin fixated human hand where different typical foreign-body materials have been inserted. Signal-to-noise ratios (SNR) have been determined for all test objects. RESULTS: On the phantom, the SNR value for wood in the dark-field channel was strongly improved by a factor 6 compared to conventional radiography and even compared to the SNR of an aluminium structure of the same size in conventional radiography. Splinters of wood < 300 µm in diameter were clearly detected on the dark-field radiography. Dark-field radiography of the formalin-fixated human hand showed a clear signal for wooden particles that could not be identified on conventional radiography. CONCLUSIONS: x-ray dark-field radiography enables the simultaneous detection of wooden and metallic particles in the extremities. It has the potential to improve and simplify the current state-of-the-art foreign-body detection.

The compositional and nano-structural basis of fracture healing in healthy and osteoporotic bone.

Osteoporosis, a prevalent metabolic bone disorder, predisposes individuals to increased susceptibility to fractures. It is also, somewhat controversially, thought to delay or impair the regenerative response. Using high-resolution Fourier-transform infrared spectroscopy and small/wide-angle X-ray scattering we sought to answer the following questions: Does the molecular composition and the nano-structure in the newly regenerated bone differ between healthy and osteoporotic environments? And how do pharmacological treatments, such as bone morphogenetic protein 7 (BMP-7) alone or synergistically combined with zoledronate (ZA), alter callus composition and nano-structure in such environments? Cumulatively, on the basis of compositional and nano-structural characterizations of newly formed bone in an open-osteotomy rat model, the healing response in untreated healthy and ovariectomy-induced osteoporotic environments was fundamentally the same. However, the BMP-7 induced osteogenic response resulted in greater heterogeneity in the nano-structural crystal dimensions and this effect was more pronounced with osteoporosis. ZA mitigated the effects of the upregulated catabolism induced by both BMP-7 and an osteoporotic bone environment. The findings contribute to our understanding of how the repair processes in healthy and osteoporotic bone differ in both untreated and treated contexts and the data presented represents the most comprehensive study of fracture healing at the nanoscale undertaken to date.

Non-iterative Directional Dark-field Tomography.

Dark-field imaging is a scattering-based X-ray imaging method that can be performed with laboratory X-ray tubes. The possibility to obtain information about unresolvable structures has already seen a lot of interest for both medical and material science applications. Unlike conventional X-ray attenuation, orientation dependent changes of the dark-field signal can be used to reveal microscopic structural orientation. To date, reconstruction of the three-dimensional dark-field signal requires dedicated, highly complex algorithms and specialized acquisition hardware. This severely hinders the possible application of orientation-dependent dark-field tomography. In this paper, we show that it is possible to perform this kind of dark-field tomography with common Talbot-Lau interferometer setups by reducing the reconstruction to several smaller independent problems. This allows for the reconstruction to be performed with commercially available software and our findings will therefore help pave the way for a straightforward implementation of orientation-dependent dark-field tomography.

Bone mineral crystal size and organization vary across mature rat bone cortex.

The macro- and micro-features of bone can be assessed by using imaging methods. However, nano- and molecular features require more detailed characterization, such as use of e.g., vibrational spectroscopy and X-ray scattering. Nano- and molecular features also affect the mechanical competence of bone tissue. The aim of the present study was to reveal the effects of mineralization and its alterations on the mineral crystal scale, by investigating the spatial variation of molecular composition and mineral crystal structure across the cross-section of femur diaphyses in young rats, and healthy and osteoporotic mature rats (N=5). Fourier transform infrared spectroscopy and scanning small- and wide-angle X-ray scattering (SAXS/WAXS) techniques with high spatial resolution were used at identical locations over the whole cross-section. This allowed quantification of point-by-point information about the spatial distribution of mineral crystal volume. All measured parameters (crystal dimensions, degree of orientation and predominant orientation) varied across the cortex. Specifically, the crystal dimensions were lower in the central cortex than in the endosteal and periosteal regions. Mineral crystal orientation followed the cortical circumference in the periosteal and endosteal regions, but was less well-oriented in the central regions. Central cortex is formed rapidly during development through endochondral ossification. Since rats possess no osteonal remodeling, this bone remains (until old age). Significant linear correlations were observed between the dimensional and organizational parameters, e.g., between crystal length and degree of orientation (R(2)=0.83, p<0.001). Application of SAXS/WAXS provides valuable information on bone nanostructure and its constituents, effects of diseases and, prospectively, mechanical competence.

Dentinal tubules revealed with X-ray tensor tomography.

Dentin is a mineralized material making up most of the tooth bulk. A system of microtubules, so called dentinal tubules, transverses it radially from the pulp chamber to the outside. This highly oriented structure leads to anisotropic mechanical properties directly connected to the tubules orientation and density: the ultimate tensile strength as well as the fracture toughness and the shear strength are largest perpendicular to dentinal tubules. Consequently, the fatigue strength depends on the direction of dentinal tubules, too. However, none of the existing techniques used to investigate teeth provide access to orientation and density of dentinal tubules for an entire specimen in a non-destructive way. In this paper, we measure a third molar human tooth both with conventional micro-CT and X-ray tensor tomography (XTT). While the achievable resolution in micro-CT is too low to directly resolve the dentinal tubules, we provide strong evidence that the direction and density of dentinal tubules can be indirectly measured by XTT, which exploits small-angle X-ray scattering to retrieve a 3D map of scattering tensors. We show that the mean directions of scattering structures correlate to the orientation of dentinal tubules and that the mean effective scattering strength provides an estimation of the relative density of dentinal tubules. Thus, this method could be applied to investigate the connection between tubule orientation and fatigue or tensile properties of teeth for a full sample without cutting one, non-representative peace of tooth out of the full sample.

Lens-term- and edge-effect in X-ray grating interferometry.

X-ray grating interferometry requires gratings with periods in the micrometer range and allows the acquisition of the dark-field contrast. The analyzer grating is designed to match the period of the interference pattern in order to translate it into a measurable intensity modulation. In this study, we explore the influence of a sample-induced mismatch between the interference pattern and the analyzer grating on the dark-field contrast. We propose a formula for the calculation of the signal due to a period mismatch and present estimations varying periods and detector pixel size. Furthermore, numerical simulations of the X-ray wave-front demonstrate that the wave-front curvature, described by the lens-term, e.g. behind a parabolic lens or edges of a sample can change the period of the interference pattern. Our results give a concrete explanation for the formation of a dark-field contrast from object edges and thus allow a better understanding of the dark-field signal obtained with a grating interferometer.

Six-dimensional real and reciprocal space small-angle X-ray scattering tomography.

When used in combination with raster scanning, small-angle X-ray scattering (SAXS) has proven to be a valuable imaging technique of the nanoscale, for example of bone, teeth and brain matter. Although two-dimensional projection imaging has been used to characterize various materials successfully, its three-dimensional extension, SAXS computed tomography, poses substantial challenges, which have yet to be overcome. Previous work using SAXS computed tomography was unable to preserve oriented SAXS signals during reconstruction. Here we present a solution to this problem and obtain a complete SAXS computed tomography, which preserves oriented scattering information. By introducing virtual tomography axes, we take advantage of the two-dimensional SAXS information recorded on an area detector and use it to reconstruct the full three-dimensional scattering distribution in reciprocal space for each voxel of the three-dimensional object in real space. The presented method could be of interest for a combined six-dimensional real and reciprocal space characterization of mesoscopic materials with hierarchically structured features with length scales ranging from a few nanometres to a few millimetres--for example, biomaterials such as bone or teeth, or functional materials such as fuel-cell or battery components.

Constrained X-ray tensor tomography reconstruction.

Quite recently, a method has been presented to reconstruct X-ray scattering tensors from projections obtained in a grating interferometry setup. The original publications present a rather specialised approach, for instance by suggesting a single SART-based solver. In this work, we propose a novel approach to solving the inverse problem, allowing the use of other algorithms than SART (like conjugate gradient), a faster tensor recovery, and an intuitive visualisation. Furthermore, we introduce constraint enforcement for X-ray tensor tomography (cXTT) and demonstrate that this yields visually smoother results in comparison to the state-of-art approach, similar to regularisation.