Applied Magnetic Resonance最新文献

Enhancing Generalization and Mitigating Overfitting in Deep Learning for Brain Cancer Diagnosis from MRI

IF 1.1 4区物理与天体物理

Applied Magnetic Resonance Pub Date : 2025-01-03 DOI: 10.1007/s00723-024-01743-y

Mohamad Abou Ali, Jinan Charafeddine, Fadi Dornaika, Ignacio Arganda-Carreras

{"title":"Enhancing Generalization and Mitigating Overfitting in Deep Learning for Brain Cancer Diagnosis from MRI","authors":"Mohamad Abou Ali, Jinan Charafeddine, Fadi Dornaika, Ignacio Arganda-Carreras","doi":"10.1007/s00723-024-01743-y","DOIUrl":"10.1007/s00723-024-01743-y","url":null,"abstract":"<div><p>Brain cancer represents a significant global health challenge with increasing incidence and mortality rates. Magnetic Resonance Imaging (MRI) plays a pivotal role in early detection and treatment planning. This study adopts a systematic approach across four phases: (1) Optimal Model Selection using the Adam optimizer, emphasizing accuracy metrics, weight computation, early stopping, and ReduceLROnPlateau techniques. (2) Real-world Scenario Simulation through synthetic perturbed datasets created by applying noise, blur (to simulate various magnetic field strengths: 1T, 1.5T, 3T), and patient motion artifacts (mimicking MRI scanning motion effects) to the testing data from the BT-MRI dataset, an online published brain tumor MRI dataset. (3) Optimization involving a range of optimizers (Adam, Adagrad, Nadam, RMSprop, SGD) and online augmentation techniques (AutoMix, CutMix, LGCOAMix, PatchUp). (4) Solution Exploration integrating Gaussian Noise and Blur as augmentation strategies during training to enhance model generalization under diverse conditions. Initial evaluations achieved strong performance, consistently reaching 99.45% accuracy on the BT-MRI dataset. However, testing against synthetic perturbed datasets mimicking real-world conditions revealed challenges in maintaining robust model performance. Despite employing diverse optimization methods and advanced augmentation techniques, this study identifies persistent challenges in ensuring model robustness with synthetic perturbed datasets. Notably, the integration of Gaussian Noise and Blur during training significantly improved model resilience. This research underscores the critical role of methodological rigor and innovative augmentation strategies in advancing deep learning applications for precise brain cancer diagnosis using MRI.</p></div>","PeriodicalId":469,"journal":{"name":"Applied Magnetic Resonance","volume":"56 3","pages":"359 - 394"},"PeriodicalIF":1.1,"publicationDate":"2025-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143471857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Derivation of Multi-Exponential Magnetic Resonance Relaxation Equations in Simple Pore Geometries

IF 1.1 4区物理与天体物理

Applied Magnetic Resonance Pub Date : 2024-12-24 DOI: 10.1007/s00723-024-01729-w

Armin Afrough

{"title":"Derivation of Multi-Exponential Magnetic Resonance Relaxation Equations in Simple Pore Geometries","authors":"Armin Afrough","doi":"10.1007/s00723-024-01729-w","DOIUrl":"10.1007/s00723-024-01729-w","url":null,"abstract":"<div><p>The common interpretation of magnetic resonance relaxation time distribution of liquids in porous media assumes a one-to-one relationship between the pore size and the relaxation time constants. This common conviction may not be correct in many microporous materials. Each pore size may be associated with more than one peak in the NMR relaxation time distributions: a single dominant peak and also possibly one or a few minor peaks. The appearance of minor peaks is due to the non-vanishing nonground eigenvalues of the diffusion–relaxation equation. Brownstein and Tarr (Phys Rev A 19:2446, 1979) described these features, but their solutions at conditions beyond the fast-diffusion regime are not widely adopted. We provide the derivation of Brownstein–Tarr equations for multi-exponential magnetic resonance relaxation decay for liquids in simple pore geometries. General solutions are presented for planar, cylindrical, and spherical pores—as well as two limiting cases of fast and slow diffusion for each geometry. Similar solutions are also relevant to first-order dilute reactions in porous media in heterogeneous reaction–diffusion systems. We hope that the availability of these derivations helps wider adoption of more realistic interpretation of magnetic resonance relaxation in porous media in the light of the multi-exponential Brownstein–Tarr model.</p></div>","PeriodicalId":469,"journal":{"name":"Applied Magnetic Resonance","volume":"56 3","pages":"313 - 358"},"PeriodicalIF":1.1,"publicationDate":"2024-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s00723-024-01729-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143471856","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cryogenic W-band Electron Spin Resonance Probehead with an Integral Cryogenic Low Noise Amplifier

IF 1.1 4区物理与天体物理

Applied Magnetic Resonance Pub Date : 2024-12-05 DOI: 10.1007/s00723-024-01732-1

Moamen Jbara, Oleg Zgadzai, Wolfgang Harneit, Aharon Blank

{"title":"Cryogenic W-band Electron Spin Resonance Probehead with an Integral Cryogenic Low Noise Amplifier","authors":"Moamen Jbara, Oleg Zgadzai, Wolfgang Harneit, Aharon Blank","doi":"10.1007/s00723-024-01732-1","DOIUrl":"10.1007/s00723-024-01732-1","url":null,"abstract":"<div><p>The quest to enhance the sensitivity of electron spin resonance (ESR) is an ongoing challenge. One potential strategy involves increasing the frequency, for instance, moving from Q-band (approximately 35 GHz) to W-band (approximately 94 GHz). However, this shift typically results in higher transmission and switching losses, as well as increased noise in signal amplifiers. In this work, we address these shortcomings by employing a W-band probehead integrated with a cryogenic low-noise amplifier (LNA) and a microresonator. This configuration allows us to position the LNA close to the resonator, thereby amplifying the acquired ESR signal with minimal losses. Furthermore, when operated at cryogenic temperatures, the LNA exhibits unparalleled noise levels that are significantly lower than those of conventional room temperature LNAs. We detail the novel probehead design and provide some experimental results at room temperature as well as cryogenic temperatures for representative paramagnetic samples. We find, for example, that spin sensitivity of ~ 3 × 10<sup>5</sup> spins/√Hz is achieved for a sample of phosphorus doped <sup>28</sup>Si, even for sub-optimal sample geometry with potential improvement to < 10<sup>3</sup> spins/√Hz in more optimal scenarios.</p></div>","PeriodicalId":469,"journal":{"name":"Applied Magnetic Resonance","volume":"56 1-2","pages":"265 - 284"},"PeriodicalIF":1.1,"publicationDate":"2024-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s00723-024-01732-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143108399","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Influence of Second-Order Effects due to Hyperfine Interaction on the Magnitude of the Larmor Frequency 14N

IF 1.1 4区物理与天体物理

Applied Magnetic Resonance Pub Date : 2024-12-05 DOI: 10.1007/s00723-024-01733-0

G. V. Mamin, F. F. Murzakhanov, I. N. Gracheva, M. R. Gafurov, V. A. Soltamov

{"title":"Influence of Second-Order Effects due to Hyperfine Interaction on the Magnitude of the Larmor Frequency 14N","authors":"G. V. Mamin, F. F. Murzakhanov, I. N. Gracheva, M. R. Gafurov, V. A. Soltamov","doi":"10.1007/s00723-024-01733-0","DOIUrl":"10.1007/s00723-024-01733-0","url":null,"abstract":"<div><p>The negatively charged boron vacancy (<span>({text{V}}_{text{B}}^{-})</span>) in hexagonal boron nitride (hBN) is currently considered an intriguing quantum object for testing and developing quantum technologies on two-dimensional van der Waals materials. This article presents results from photoinduced electron spin echo (ESE)-detected electron spin resonance (ESR) and electron–nuclear double-resonance (ENDOR) spectroscopy at the W-band (ν = 94 GHz), focusing on the interactions of the <span>({text{V}}_{text{B}}^{-})</span> electron spin with the three nearest nitrogen nuclei (<sup>14</sup>N, <i>I</i> = 1). The lines in the ENDOR spectrum are due to both hyperfine and quadrupole interactions for <i>M</i><sub>S</sub> = ± 1 levels and only quadrupole interactions for <i>M</i><sub>S</sub> = 0 levels. We show that significant hyperfine interaction with the three nearest nitrogen atoms, despite the high magnetic field, results in a mixing of the hyperfine sublevels for <i>M</i><sub>S</sub> = 0. We show that significant hyperfine interaction with the three nearest nitrogen atoms, despite the high magnetic field, results in mixing of the hyperfine sublevels. This mixing shifts the <sup>14</sup>N Larmor frequency from its nominal value defined as <span>({{varvec{nu}}}_{{varvec{L}}}=boldsymbol{ }{{varvec{g}}}_{{varvec{N}}}{{varvec{mu}}}_{{varvec{N}}}{varvec{B}}/{varvec{h}})</span>. This shift observed through ENDOR experiments can be understood using spin-Hamiltonian formalism within the second-order perturbation theory. These findings enhance an understanding of electron–nuclear interactions in hBN.</p></div>","PeriodicalId":469,"journal":{"name":"Applied Magnetic Resonance","volume":"56 1-2","pages":"285 - 295"},"PeriodicalIF":1.1,"publicationDate":"2024-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143108400","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Preface to Special Issue Celebration of 80 Years of EPR Part 2

IF 1.1 4区物理与天体物理

Applied Magnetic Resonance Pub Date : 2024-12-03 DOI: 10.1007/s00723-024-01738-9

Kev M. Salikhov, Christiane R. Timmel, Eric J. L. McInnes, Gareth R. Eaton

引用次数: 0

Weighted Majority Voting Ensemble for MRI-Based Brain Tumor Classification Using Capsule Networks and XGBoost

IF 1.1 4区物理与天体物理

Applied Magnetic Resonance Pub Date : 2024-11-29 DOI: 10.1007/s00723-024-01731-2

D. Saravanan, G. Arunkumar, T. Ragupathi, P. B. V. Raja Rao

{"title":"Weighted Majority Voting Ensemble for MRI-Based Brain Tumor Classification Using Capsule Networks and XGBoost","authors":"D. Saravanan, G. Arunkumar, T. Ragupathi, P. B. V. Raja Rao","doi":"10.1007/s00723-024-01731-2","DOIUrl":"10.1007/s00723-024-01731-2","url":null,"abstract":"<div><p>Brain Tumor (BT) is the most serious illness affecting humans, and its diagnosis is a complex process. Tumor location and type significantly affect treatment decisions, and survival rates improve with accurate identification and classification in the early stages. Magnetic Resonance Imaging (MRI) is mainly used for brain tumor analysis, but manual detection and classification by clinicians is challenging, often leading to high error rates, inaccurate diagnoses, and prolonged time requirements. To overcome these challenges, this paper introduces a novel hybrid classification approach that combines Capsule Networks (CapsNet) and XGBoost (XGB) to classify brain tumors from MRI images. The preprocessing step includes normalization, image blurring, resizing, contrast enhancement, and noise elimination, which are used to improve image quality. The classification process employs CapsNet to capture hierarchical features and spatial relationships in the images, while XGB utilizes extracted features, such as texture, intensity, and shape, to classify tumors effectively. To improve diagnostic accuracy, a Meta Ensemble Model combines the predictions of both algorithms using a Weighted Majority Voting approach, adjusting contributions based on each model’s confidence. Additionally, the Mantis Search Algorithm (MSA) is utilized for hyperparameter tuning, optimizing model performance by exploring the hyperparameter space effectively. The experiment assessed using the Brain Tumor MRI Dataset and Figshare Brain Tumor Dataset demonstrates the effectiveness of the proposed method, achieving an accuracy of 99.34% and a precision of 98.82%. These results indicate that the hybrid method is highly effective in accurately classifying various brain tumor types, which provides the best solution for clinical diagnostics.</p></div>","PeriodicalId":469,"journal":{"name":"Applied Magnetic Resonance","volume":"56 3","pages":"395 - 425"},"PeriodicalIF":1.1,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143471858","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Evolution of Bruker EPR Spectrometers as well as Prospects for Present and Future EPR Applications

IF 1.1 4区物理与天体物理

Applied Magnetic Resonance Pub Date : 2024-11-27 DOI: 10.1007/s00723-024-01734-z

Ralph T. Weber

引用次数: 0

Autobiography 自传

IF 1.1 4区物理与天体物理

Applied Magnetic Resonance Pub Date : 2024-11-16 DOI: 10.1007/s00723-024-01705-4

Carlo Corvaja

引用次数: 0

Performance Comparison of Different Rapid Freeze–Quench Strategies for Electron Paramagnetic Resonance

IF 1.1 4区物理与天体物理

Applied Magnetic Resonance Pub Date : 2024-11-15 DOI: 10.1007/s00723-024-01725-0

Maruan Bracci, Ilenia Serra, Inés García-Rubio, Sabine Van Doorslaer

{"title":"Performance Comparison of Different Rapid Freeze–Quench Strategies for Electron Paramagnetic Resonance","authors":"Maruan Bracci, Ilenia Serra, Inés García-Rubio, Sabine Van Doorslaer","doi":"10.1007/s00723-024-01725-0","DOIUrl":"10.1007/s00723-024-01725-0","url":null,"abstract":"<div><p>This work addresses the development of a custom-made home-built rapid freeze–quench (RFQ) device and the comparison of its performance to the one of a commercial RFQ setup that was in-house custom adapted. Both systems consist of two syringes that push the reactants into a mixing chamber and the products to a subsequent freezing setup. Using the binding of azide to myoglobin as a calibration reaction, the quenching times of the different setups were compared, evaluating different instrumental parameters, such as software-controlled variation of the aging time, variations of the flow rate and variations of the distance travelled by the mixed sample before freezing. In addition to minimal sample consumption, the home-built RFQ device was found to lead to the shorter reaction times which could be controlled in a time range from 10 to 25 ms. The commercial RFQ system yielded optimal reaction control in a time range from 50 to 200 ms, although a larger volume of reactants needed to be used due to the significant dead volume of the system. Three different freezing methods were also evaluated, among which, in our hands, freezing the jet directly in a deep bath of cold isopentane yielded shorter and reproducible freezing times.</p></div>","PeriodicalId":469,"journal":{"name":"Applied Magnetic Resonance","volume":"56 1-2","pages":"229 - 252"},"PeriodicalIF":1.1,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143108733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Carlo Corvaja: On the Occasion of His 85th Birthday 卡洛-科瓦哈在他 85 岁生日之际

IF 1.1 4区物理与天体物理

Applied Magnetic Resonance Pub Date : 2024-11-15 DOI: 10.1007/s00723-024-01728-x

Lorenzo Franco, Marco Ruzzi

引用次数: 0