National Magnetic Resonance Research Center (UMRAM)

Permanent URI for this collection

https://hdl.handle.net/11693/48237

Browse

Now showing 1 - 20 of 352

Open Access
Demonstration of chewing-related areas in the brain via functional magnetic resonance imaging
(International Scientific Literature, Inc., 2023-01-31) Algın, Oktay; Kocak, O. M.; Gökçekuyu, Yasemin; Turker, K. S.
Purpose: To localize and identify chewing-related areas and their connections with other centres in the human brain using functional magnetic resonance imaging (fMRI). Material and methods: The paradigm of the present study was block designed. Spontaneous and controlled chewing with sugar-free gum was used as the main task in a 3-Tesla fMRI unit with a 32-channel birdcage coil. Our study popu lation comprised 32 healthy volunteers. To determine possible intersections, we also put the rosary pulling (silent tell one’s beads) movement in the fMRI protocol. The data analyses were performed with the Statistical Parametric Mapping (SPM) toolbox integrated into the Matlab platform. Results: The superomedial part of the right cerebellum was activated during either pulling rosary beads or spontaneous chewing. This region, however, was not activated during controlled chewing. We did not find statistically significant activation or connection related to the brain stem. Conclusion: We have confirmed that the cerebellum plays an important role in chewing. However, we could not find a definite central pattern generator (CPG) in the brain stem, which has been hypothesized to underlie spontaneous chewing.
Embargo
Neural correlates of distorted body images in adolescent girls with anorexia nervosa: how is it different from major depressive disorder?
(John Wiley and Sons Ltd, 2023-06-28) Karakuş Aydos, Y.; Dövencioğlu, D.; Karlı Oğuz, Kader; Özdemir, P.; Pehlivantürk Kızılkan, M.; Kanbur, N.; Ünal, D.; Nalbant, K.; Çetin Çuhadaroğlu, F.; Akdemir, D.
Body image disturbance is closely linked to eating disorders including anorexia nervosa (AN). Distorted body image perception, dissatisfaction and preoccupation with weight and shape are often key factors in the development and maintenance of these disorders. Although the pathophysiological mechanism of body image disorder is not yet fully understood, aberrant biological processes may interfere with perceptive, cognitive and emotional aspects of body image. This study focuses on the neurobiological aspects of body image disturbance. The sample consisted of 12 adolescent girls diagnosed with AN, nine girls with major depressive disorder (MDD) and 10 without psychiatric diagnoses (HC, the healthy control group). We applied a block-design task in functional magnetic resonance imaging using participants' original and distorted overweight and underweight images. After imaging, the participants scored the images for resemblance, satisfaction and anxiety levels. The findings of this study demonstrate that overweight images elicited dissatisfaction and increased occipitotemporal activations across all participants. However, no difference was found between the groups. Furthermore, the MDD and HC groups showed increased activations in the prefrontal cortex and insula in response to underweight images compared to their original counterparts, whereas the AN group exhibited increased activations in the parietal cortex, cingulate gyrus and parahippocampal cortex in response to the same stimuli.
Open Access
A transformer-based prior legal case retrieval method
(IEEE - Institute of Electrical and Electronics Engineers, 2023-08-28) Öztürk, Ceyhun Emre; Özçelik, Şemsi Barış; Koç, Aykut
In this work, BERTurk-Legal, a transformer-based language model, is introduced to retrieve prior legal cases. BERTurk-Legal is pre-trained on a dataset from the Turkish legal domain. This dataset does not contain any labels related to the prior court case retrieval task. Masked language modeling is used to train BERTurk-Legal in a self-supervised manner. With zero-shot classification, BERTurk-Legal provides state-of-the-art results on the dataset consisting of legal cases of the Court of Cassation of Turkey. The results of the experiments show the necessity of developing language models specific to the Turkish law domain.
Open Access
The brainstem connections of the supplementary motor area and its relations to the corticospinal tract: experimental rat and human 3-tesla tractography study
(Elsevier Ireland Ltd, 2023-01-28) Çavdar, Safiye; Köse, Büşra; Altınöz, Damlasu; Özkan, Mazhar; Güneş, Yasin Celal; Algın, Oktay
Although the supplementary motor area (SMA) is a large region on the medial surface of the frontal lobe of the brain, little is known about its function. The current study uses 3-tesla high-resolution diffusion tensor tractography (DTI) in healthy individuals and biotinylated dextran amine (BDA) and fluoro-gold (FG) tracer in rats to demonstrate the afferent and efferent connections of the SMA with brainstem structures. It also aims to clarify how SMA fibers relate to the corticospinal tract (CST). The BDA (n = 6) and FG (n = 8) tracers were pressure-injected into the SMA of 14 Wistar albino rats. Light and fluorescence microscopy was used to capture images of the FG and BDA-labeled cells and axons. High-resolution 3-tesla DTI data were acquired from the Human Connectome Project database. Tracts between the SMA and brainstem structures were analyzed using diffusion spectrum imaging (DSI) studio software. The FG injections into the SMA showed afferent projections from mesencephalic (periaqueductal gray matter, substantia nigra pars reticulata, ventral tegmental area, inferior colliculus, mesencephalic reticular, tegmental, and raphe nuclei), pontine (locus coeruleus, pontine reticular and vestibular nuclei), and medullary (area postrema, parabrachial, and medullary reticular nuclei) structures. The anterograde tracer BDA injections into the SMA showed efferent connections with mesencephalic (periaqueductal gray, substantia nigra pars compacta, dorsal raphe, trigeminal motor mesencephalic, and mesencephalic reticular nuclei), pontine (locus coeruleus, nucleus of the lateral lemniscus, vestibular, cochlear, and pontine reticular nuclei), and medullary (area postrema, medullary reticular, olivary, and parabrachial nuclei) structures. The SMA had efferent but no afferent connections with the cerebellar nuclei. The DTI results in healthy human subjects highly corresponded with the experimental results. Further, the DTI results showed a distinct bundle that descended to spinal levels closely related to the CST. Understanding SMA's afferent and efferent connections will enrich our knowledge of its contribution to various brainstem networks and may provide new perspectives for understanding its motor and non-motor functions. © 2023 Elsevier B.V.
Open Access
Denoising diffusion adversarial models for unconditional medical image generation
(IEEE - Institute of Electrical and Electronics Engineers, 2023-08-28) Dalmaz, Onat; Sağlam, Baturay; Elmas, Gökberk; Mirza, Muhammad Usama; Çukur, Tolga
Unconditional medical image synthesis is the task of generating realistic and diverse medical images from random noise without any prior information or constraints. Synthesizing realistic medical images can enrich the quality and diversity of medical imaging datasets, which in turn, enhance the performance and generalization of deep learning models for medical imaging. Prevalent approach for synthesizing medical images involves generative adversarial networks (GAN) or denoising diffusion probabilistic models (DDPM). However, GAN models that implicitly learn the image distribution are prone to limited sample fidelity and diversity. On the other hand, diffusion models suffer from slow sampling speed due to small diffusion steps. In this paper, we propose a novel diffusion-based method for unconditional medical image synthesis, Diff-Med-Synth, that generates realistic and diverse medical images from random noise. Diff-Med-Synth combines the advantages of denoising diffusion probabilistic models and GANs to achieve fast and efficient image sampling. We evaluate our method on two multi-contrast MRI datasets and show that it outperforms state-of-the-art methods in terms of quality, diversity, and fidelity of the synthesized images.
Open Access
A diffusion-based reconstruction technique for single pixel camera
(IEEE - Institute of Electrical and Electronics Engineers, 2023-08-28) Güven, Baturalp; Güngör, A.; Bahçeci, M. U.; Çukur, Tolga
Single-pixel imaging enables high-resolution imaging through multiple coded measurements based on lowresolution snapshots. To reconstruct a high-resolution image from these coded measurements, an ill-posed inverse problem is solved. Despite the recent popularity of deep learning-based methods for single-pixel imaging reconstruction, they are insufficient in preserving spatial details and achieving a stable reconstruction. Diffusion-based methods, which have gained attention in recent years, provide a solution to this problem. In this study, to the best of our knowledge, the single-pixel image reconstruction is performed for the first time using a denoising diffusion probabilistic model. The proposed method reconstructs the image by conditioning it towards the least squares solution while preserving data consistency after unconditional training of the model. The proposed method is compared against existing singlepixel imaging methods, and ablation studies are conducted to demonstrate the individual model components. The proposed method outperforms competing methods in both quantitative measurements and visual quality.
Open Access
Focal modulation based end-to-end multi-label classification for chest X-ray image classification
(IEEE - Institute of Electrical and Electronics Engineers, 2023-08-28) Öztürk, Şaban; Çukur, Tolga
Chest X-ray imaging is of critical importance in order to effectively diagnose chest diseases, which are increasing today due to various environmental and hereditary factors. Although chest X-ray is the most commonly used device for detecting pathological abnormalities, it can be quite challenging for specialists due to misleading locations and sizes of pathological abnormalities, visual similarities, and complex backgrounds. Traditional deep learning (DL) architectures fall short due to relatively small areas of pathological abnormalities and similarities between diseased and healthy areas. In addition, DL structures with standard classification approaches are not ideal for dealing with problems involving multiple diseases. In order to overcome the aforementioned problems, firstly, background-independent feature maps were created using a conventional convolutional neural network (CNN). Then, the relationships between objects in the feature maps are made suitable for multi-label classification tasks using the focal modulation network (FMA), an innovative attention module that is more effective than the self-attention approach. Experiments using a Chest x-ray dataset containing both single and multiple labels for a total of 14 different diseases show that the proposed approach can provide superior performance for multi-label datasets.
Open Access
Transformer-based bug/feature classification
(IEEE - Institute of Electrical and Electronics Engineers, 2023-08-28) Öztürk, Ceyhun Emre; Yılmaz, E. H.; Köksal, Ö.
Automatic classification of a software bug report as a 'bug' or 'feature' is essential to accelerate closed-source software development. In this work, we focus on automating the bug/feature classification task with artificial intelligence using a newly constructed dataset of Turkish software bug reports collected from a commercial project. We train and test support vector machine (SVM), k-nearest neighbors (KNN), convolutional neural network (CNN), transformer-based models, and similar artificial intelligence models on the collected reports. Results of the experiments show that transformer-based BERTurk is the best-performing model for the bug/feature classification task.
Open Access
Use of the Woven EndoBridge device for sidewall aneurysms: systematic review and meta-analysis
(American Society of Neuroradiology, 2023-02) Rodriguez-Calienes, A.; Vivanco-Suarez, J.; Galecio-Castillo, M.; Zevallos, C. B.; Farooqui, M.; Malaga, M.; Moran-Mariños, C.; Fanning, N. F.; Algın, Oktay; Samaniego, E. A.; Pabon, B.; Mouchtouris, N.; Altschul, D. J.; Jabbour, P.; Ortega-Gutierrez, S.
Background: The Woven EndoBridge device was originally approved to treat intracranial wide-neck saccular bifurcation aneurysms. Recent studies have suggested its use for the treatment of sidewall intracranial aneurysms with variable success. Purpose: Our aim was to evaluate the safety and efficacy of the Woven EndoBridge device for sidewall aneurysms using a meta-analysis of the literature. Data sources: We performed a systematic review of all studies including patients treated with the Woven EndoBridge device for sidewall aneurysms from inception until May 2022 on Scopus, EMBASE, MEDLINE, the Web of Science, and the Cochrane Central Register of Controlled Trials. Study selection: Ten studies were selected, and 285 patients with 288 sidewall aneurysms were included. Data analysis: A random-effects meta-analysis of proportions using a generalized linear mixed model was performed as appropriate. Statistical heterogeneity across studies was assessed with I2 statistics. Data synthesis: The adequate occlusion rate at last follow-up was 89% (95% CI, 81%-94%; I2, = 0%), the composite safety outcome was 8% (95% CI, 3%-17%; I2 = 34%), and the mortality rate was 2% (95% CI, 1%-7%; I2 = 0%). Aneurysm width (OR = 0.5; P = .03) was the only significant predictor of complete occlusion. Limitations: Given the level of evidence, our results should be interpreted cautiously until confirmation from larger prospective studies is obtained. Conclusions: The initial evidence evaluating the use of the Woven EndoBridge device for the treatment of wide-neck sidewall intracranial aneurysms has demonstrated high rates of adequate occlusion with low procedural complications. Our findings favor the consideration of the Woven EndoBridge device as an option for the treatment of sidewall aneurysms.
Embargo
Zebrafish optomotor response to second-order motion illustrates that age-related changes in motion detection depend on the activated motion system
(Elsevier Inc., 2023-06-10) Karaduman, Ayşenur; Karoğlu-Eravşar, Elif Tuğçe; Kaya, Utku; Aydın, Alaz; Adams, Michelle Marie; Kafalıgönül, Hulusi
Various aspects of visual functioning, including motion perception, change with age. Yet, there is a lack of comprehensive understanding of age-related alterations at different stages of motion processing and in each motion system. To understand the effects of aging on second-order motion processing, we investigated optomotor responses (OMR) in younger and older wild-type (AB-strain) and acetylcholinesterase (achesb55/+) mutant zebrafish. The mutant fish with decreased levels of acetylcholinesterase have been shown to have delayed age-related cognitive decline. Compared to previous results on first-order motion, we found distinct changes in OMR to second-order motion. The polarity of OMR was dependent on age, such that second-order stimulation led to mainly negative OMR in the younger group while older zebrafish had positive responses. Hence, these findings revealed an overall aging effect on the detection of second-order motion. Moreover, neither the genotype of zebrafish nor the spatial frequency of motion significantly changed the response magnitude. Our findings support the view that age-related changes in motion detection depend on the activated motion system. © 2023 Elsevier Inc.
Open Access
Content-based medical image retrieval with opponent class adaptive margin loss
(Elsevier Inc., 2023-04-13) Öztürk, Şaban; Çelik, Emin; Çukur, Tolga
The increasing utilization of medical imaging technology with digital storage capabilities has facilitated the compilation of large-scale data repositories. Fast access to image samples with similar appearance to suspected cases in these repositories can help establish a consulting system for healthcare professionals, and improve diagnostic procedures while minimizing processing delays. However, manual querying of large repositories is labor intensive. Content-based image retrieval (CBIR) offers an automated solution based on quantitative assessment of image similarity based on image features in a latent space. Since conventional methods based on hand-crafted features typically show poor generalization performance, learning-based CBIR methods have received attention recently. A common framework in this domain involves classifier-guided models that are trained to detect different image classes. Similarity assessments are then performed on the features captured by the intermediate stages of the trained models. While classifier-guided methods are powerful in inter-class discrimination, they are suboptimally sensitive to within-class differences in image features. An alternative framework instead performs task-agnostic training to learn an embedding space that enforces the representational discriminability of images. Within this representational-learning framework, a powerful method is triplet-wise learning that addresses the deficiencies of point-wise and pair-wise learning in characterizing the similarity relationships between image classes. However, the traditional triplet loss enforces separation between only a subset of image samples within the triplet via a manually-set constant margin value, so it can lead to suboptimal segregation of opponent classes and limited generalization performance. To address these limitations, we introduce a triplet-learning method for automated querying of medical image repositories based on a novel Opponent Class Adaptive Margin (OCAM) loss. To maintain optimally discriminative representations, OCAM considers relationships among all image pairs within the triplet and utilizes an adaptive margin value that is automatically selected per dataset and during the course of training iterations. CBIR performance of OCAM is compared against state-of-the-art loss functions for representational learning on three public databases (gastrointestinal disease, skin lesion, lung disease). On average, OCAM shows an mAP performance of 86.30% in the KVASIR dataset, 70.30% in the ISIC 2019 dataset, and 85.57% in the X-RAY dataset. Comprehensive experiments in each application domain demonstrate the superior performance of OCAM against competing triplet-wise methods at 1.52%, classifier-guided methods at 2.29%, and non-triplet representational-learning methods at 4.56%.
Open Access
Cisternography of arachnoid cysts
(Springer Link, 2023-01-01) Güneş, Y. C.; Algın, Oktay; Turgut, Mehmet; Akhaddar, Ali; Turgut, Ahmet T.; Hall, Walter A.
Conventional magnetic resonance (MR) sequences are useful for the diagnosis of arachnoid cysts (ACs), but they are not sufficient to show the communication between ACs and adjacent cerebrospinal fluid (CSF)-containing areas. Contrast material-enhanced computed tomography (CT) and MR cisternography play a crucial role in demonstrating this relationship. These tests create a contrast difference in the CSF in the ventricular system and in the cisternal spaces. Compared with radionuclide cisternography and CT cisternography (CTC), the main advantages of contrast material-enhanced MR cisternography (CE-MRC) are its high contrast-noise ratio, multiplanar analysis capacity, images with thin section thickness, and not causing radiation exposure. ACs that were filled with intrathecal contrast on early-phase postcontrast (after intrathecal administration) images were included in the full communicating group that did not fill at the 24th hour later or showed minimal filling was included in the noncommunication group. In addition, 3D-SPACE with variant flip-angle mode (VFAM) and PC-MRI techniques have special capabilities for evaluating CSF flow. It was successful in showing the communication between ACs and the adjacent CSF-containing areas. In cases with suspicious findings on phase-contrast MRI images, the final decision can be made with CE-MRC. CE-MRC is considered the gold standard in demonstrating the relationship between ACs and CSF-containing spaces. Isotope cisternography has obvious disadvantages, and it is no longer used today. In conclusion, demonstrating the relationship of ACs with adjacent CSF-filled structures is important in making surgical decisions.
Embargo
Quantitative radial force measurements of Woven EndoBridge devices
(Sage Publications Ltd., 2023-10-31) Kutbay, Uğurhan; Algın, Oktay
Background Lateral/radial forces and the mechanical properties of Woven EndoBridge (WEB) devices have significant importance for therapeutic success. In other words, adequate apposition of the lateral wall of a cerebral aneurysm is critical for preventing recurrence or re-rupture risk. Objective This study aimed to investigate the pressure values applied by different WEB devices to the lateral walls of aneurysms and the relationships between these pressure measurements and the diameters of WEB devices. Methods By placing four WEB devices of different sizes and types between two rigid metal plates, the lateral forces applied by these WEB devices to plates of different apertures were measured quantitatively. We tested a single device of each size over multiple periods. The total number of examined WEB devices is four. Results There was a significant negative relationship between plate distances and pressure values (correlation coefficient:–0.956, p = 0.000). The lateral wall apposition pressure of a 4- or 5-mm aperture size was higher than a 6-mm aperture size for SL-type WEB devices with a 7-mm diameter. Similarly, the lateral wall apposition pressure detected for a 3- or 3.5-mm aperture size was higher than a 4-mm aperture size for W5-4.5-3 and W5-5-3.6. It was observed that maximum lateral wall pressure was detected in plate measurements of SLS-type devices compared to SL-type devices. The diameter and height values of 3 of the 4 unconstrained WEB devices analyzed differed from the catalog values. Conclusion It seems that SLS-type devices apply more pressure on the aneurysm's lateral borders than SL-type devices.
Open Access
Multi-label sentiment analysis on 100 languages with dynamic weighting for label imbalance
(Institute of Electrical and Electronics Engineers Inc., 2023-01-01) Yılmaz, Selim Fırat; Kaynak, Ergün Batuhan; Koç, Aykut; Dibeklioğlu, Hamdi; Kozat, Süleyman Serdar
We investigate cross-lingual sentiment analysis, which has attracted significant attention due to its applications in various areas including market research, politics, and social sciences. In particular, we introduce a sentiment analysis framework in multi-label setting as it obeys Plutchik’s wheel of emotions. We introduce a novel dynamic weighting method that balances the contribution from each class during training, unlike previous static weighting methods that assign non-changing weights based on their class frequency. Moreover, we adapt the focal loss that favors harder instances from single-label object recognition literature to our multi-label setting. Furthermore, we derive a method to choose optimal class-specific thresholds that maximize the macro-f1 score in linear time complexity. Through an extensive set of experiments, we show that our method obtains the state-of-the-art performance in seven of nine metrics in three different languages using a single model compared with the common baselines and the best performing methods in the SemEval competition. We publicly share our code for our model, which can perform sentiment analysis in 100 languages, to facilitate further research.
Open Access
Unsupervised medical image translation with adversarial diffusion models
(Institute of Electrical and Electronics Engineers , 2023-11-30) Özbey, Muzaffer; Dalmaz, Onat; Dar, Salman Ul Hassan; Bedel, Hasan Atakan; Özturk, Şaban; Güngör, Alper; Çukur, Tolga
Imputation of missing images via source-to-target modality translation can improve diversity in medical imaging protocols. A pervasive approach for synthesizing target images involves one-shot mapping through generative adversarial networks (GAN). Yet, GAN models that implicitly characterize the image distribution can suffer from limited sample fidelity. Here, we propose a novel method based on adversarial diffusion modeling, SynDiff, for improved performance in medical image translation. To capture a direct correlate of the image distribution, SynDiff leverages a conditional diffusion process that progressively maps noise and source images onto the target image. For fast and accurate image sampling during inference, large diffusion steps are taken with adversarial projections in the reverse diffusion direction. To enable training on unpaired datasets, a cycle-consistent architecture is devised with coupled diffusive and non-diffusive modules that bilaterally translate between two modalities. Extensive assessments are reported on the utility of SynDiff against competing GAN and diffusion models in multi-contrast MRI and MRI-CT translation. Our demonstrations indicate that SynDiff offers quantitatively and qualitatively superior performance against competing baselines.
Open Access
Segmentation of spinal subarachnoid lumen with 3d attention u-net
(World Scientific Publishing, 2023-05-01) Keleş, A.; Algın, Oktay; Özışık Akdemir, P.; Şen, B.; Çelebi, F. V.
Phase Contrast Magnetic Resonance Image (PC-MRI) is an emerging noninvasive technique that contains pulsatile information by measuring the parameters of cerebrospinal fluid (CSF) flow. As CSF flow quantities are measured from the selected region on the images, the accuracy in the identification of the interested region is the most essential, and the examination requires a lot of time and experience to analyze and for accurate CSF flow assessment. In this study, a three-dimensional (3D)-Unet architecture, including pulsatile flow data as the third dimension, is proposed to address the issue. The dataset contains 2176 phase and rephase images from 57 slabs of 39 3-tesla PC-MRI subjects collected from the lower thoracic levels of control and Idiopathic Scoliosis (IS) patients. The procedure starts with labeling the CSF containing spaces in the spinal canal. In the preprocessing step, unequal cardiac cycle images (i.e., frame) and the numbers of MRIs in cases are adjusted by interpolation to align the temporal dimension of the dataset to an equal size. The five-fold cross-validation procedure is used to evaluate the 3D Attention-U-Net model after training and achieved an average weighted performance of 97% precision, 95% recall, 98% F1 score, and 95% area under curve. The success of the model is also measured using the CSF flow waveform quantities as well. The mean flow rates through the labeled and predicted CSF lumens have a significant correlation coefficient of 0.96, and the peak CSF flow rates have a coefficient of 0.65. To our knowledge, this is the first fully automatic 3D deep learning architecture implementation to segment spinal CSF-containing spaces that utilizes both spatial and pulsatile information in PC-MRI data. We expect that our work will attract future research on the use of PC-MRI temporal information for training deep models.
Open Access
A simulation study for an open-sided hybrid MPI-MRI scanner
(Infinite Science Publishing, 2023-03-19) Karaca, Sefa; Alptekin Soydan, D.; Top, C. B.; Sarıtaş, Emine Ülkü
Magnetic particle imaging (MPI) provides images of magnetic nanoparticle distribution without any signal from the surrounding tissue. MPI would benefit from an additional imaging technique that reveals the anatomical background information, required in many applications. Here, we present a simulation study based on our in-house open-sided prototype MPI system, in which the coils can be utilized interchangeably for MPI and MRI data acquisitions. The system can provide a selection field gradient of 0.5 T m−1 for MPI in field free line topology, and a B0 field of up to 50 mT for MRI. We analyze the system-induced deviations on MRI images for different B0 valuesand pulse sequence parameters.
Open Access
X-space image reconstruction for lissajous trajectory using multidimensional image tensor
(Infinite Science Publishing, 2023-03-19) Ömeroglu, Osmanalp; Erol, Hasan Sabri Melihcan; Özaslan, Ali Alper; Sarıtaş, Emine Ülkü
The tensor-based theory of multidimensional x-space MPI provides useful insight into MPI image reconstruction. Using this theory, it was shown that x-space MPI images with isotropic resolution can be achieved by scanning in two orthogonal directions separately and combining the resulting images. In this work, we propose an x-space image reconstruction that resolves the multidimensional image tensor, allowing us to reconstruct the isotropic MPI image for the Lissajous trajectory. The proposed method takes advantage of the self-crossing property of the Lissajous trajectory.
Open Access
Single-pass relaxation mapping at multiple frequencies using an arbitrary waveform MPI scanner
(Infinite Science Publishing, 2023-03-19) Alyüz, Beril; Arslan, Musa Tunç; Utkur, Mustafa; Sarıtaş, Emine Ülkü
In Magnetic Particle Imaging (MPI), relaxation behavior of magnetic nanoparticles (MNPs) has enabled the infer-ence of information about different MNP types and their local environments, such as viscosity and temperature. Previously, we have proposed and demonstrated an arbitrary waveform (AW) MPI scanner that facilitates operation in a wide range of drive field (DF) frequencies by eliminating the need for impedance matching. In this work, we propose a technique for simultaneous relaxation mapping at multiple DF frequencies in a single pass using an AWMPI scanner.
Open Access
A denoiser scaling technique for plug-and-play MPI reconstruction
(Infinite Science Publishing, 2023-03-19) Güngör, Alper; Aşkın, Barış; Alptekin Soydan, D.; Sarıtaş, Emine Ülkü; Top, C. B.; Çukur, Tolga
Image reconstruction based on the system matrix in magnetic particle imaging (MPI) involves an ill-posed inverse problem, which is often solved using iterative optimization procedures that use regularization. Reconstruction performance is highly dependent on the quality of information captured by the regularization prior. Learning-based methods have been recently introduced that significantly improve prior information in MPI reconstruction. Yet, these methods can perform suboptimally under drifts in the image scale between the training and test sets. In this study, we assess the influence of scale drifts on the performance a recent plug-ang-play method (PP-MPI) that uses a pre-trained denoiser. We introduce a new denoiser scaling technique that improves reliability of PP-MPI against deviations in image scale. The proposed technique enables high quality reconstructions that are robust against scale drifts between training and testing sets.