LST-AI: A deep learning ensemble for accurate MS lesion segmentation.

Automated segmentation of brain white matter lesions is crucial for both clinical assessment and scientific research in multiple sclerosis (MS). Over a decade ago, we introduced an engineered lesion segmentation tool, LST. While recent lesion segmentation approaches have leveraged artificial intelligence (AI), they often remain proprietary and difficult to adopt. As an open-source tool, we present LST-AI, an advanced deep learning-based extension of LST that consists of an ensemble of three 3D U-Nets. LST-AI explicitly addresses the imbalance between white matter (WM) lesions and non-lesioned WM. It employs a composite loss function incorporating binary cross-entropy and Tversky loss to improve segmentation of the highly heterogeneous MS lesions. We train the network ensemble on 491 MS pairs of T1-weighted and FLAIR images, collected in-house from a 3T MRI scanner, and expert neuroradiologists manually segmented the utilized lesion maps for training. LST-AI also includes a lesion location annotation tool, labeling lesions as periventricular, infratentorial, and juxtacortical according to the 2017 McDonald criteria, and, additionally, as subcortical. We conduct evaluations on 103 test cases consisting of publicly available data using the Anima segmentation validation tools and compare LST-AI with several publicly available lesion segmentation models. Our empirical analysis shows that LST-AI achieves superior performance compared to existing methods. Its Dice and F1 scores exceeded 0.62, outperforming LST, SAMSEG (Sequence Adaptive Multimodal SEGmentation), and the popular nnUNet framework, which all scored below 0.56. Notably, LST-AI demonstrated exceptional performance on the MSSEG-1 challenge dataset, an international WM lesion segmentation challenge, with a Dice score of 0.65 and an F1 score of 0.63-surpassing all other competing models at the time of the challenge. With increasing lesion volume, the lesion detection rate rapidly increased with a detection rate of >75% for lesions with a volume between 10 mm3 and 100 mm3. Given its higher segmentation performance, we recommend that research groups currently using LST transition to LST-AI. To facilitate broad adoption, we are releasing LST-AI as an open-source model, available as a command-line tool, dockerized container, or Python script, enabling diverse applications across multiple platforms.

Quantitative susceptibility mapping in multiple sclerosis: A systematic review and meta-analysis.

BACKGROUND: Quantitative susceptibility mapping (QSM) is a quantitative measure based on magnetic resonance imaging sensitive to iron and myelin content. This makes QSM a promising non-invasive tool for multiple sclerosis (MS) in research and clinical practice. OBJECTIVE: We performed a systematic review and meta-analysis on the use of QSM in MS. METHODS: Our review was prospectively registered on PROSPERO (CRD42022309563). We searched five databases for studies published between inception and 30th April 2023. We identified 83 English peer-reviewed studies that applied QSM images on MS cohorts. Fifty-five included studies had at least one of the following outcome measures: deep grey matter QSM values in MS, either compared to healthy controls (HC) (k = 13) or correlated with the score on the Expanded Disability Status Scale (EDSS) (k = 7), QSM lesion characteristics (k = 22) and their clinical correlates (k = 17), longitudinal correlates (k = 11), histological correlates (k = 7), or correlates with other imaging techniques (k = 12). Two meta-analyses on deep grey matter (DGM) susceptibility data were performed, while the remaining findings could only be analyzed descriptively. RESULTS: After outlier removal, meta-analyses demonstrated a significant increase in the basal ganglia susceptibility (QSM values) in MS compared to HC, caudate (k = 9, standardized mean difference (SDM) = 0.54, 95 % CI = 0.39-0.70, I2 = 46 %), putamen (k = 9, SDM = 0.38, 95 % CI = 0.19-0.57, I2 = 59 %), and globus pallidus (k = 9, SDM = 0.48, 95 % CI = 0.28-0.67, I2 = 60 %), whereas thalamic QSM values exhibited a significant reduction (k = 12, SDM = -0.39, 95 % CI = -0.66--0.12, I2 = 84 %); these susceptibility differences in MS were independent of age. Further, putamen QSM values positively correlated with EDSS (k = 4, r = 0.36, 95 % CI = 0.16-0.53, I2 = 0 %). Regarding rim lesions, four out of seven studies, representing 73 % of all patients, reported rim lesions to be associated with more severe disability. Moreover, lesion development from initial detection to the inactive stage is paralleled by increasing, plateauing (after about two years), and gradually decreasing QSM values, respectively. Only one longitudinal study provided clinical outcome measures and found no association. Histological data suggest iron content to be the primary source of QSM values in DGM and at the edges of rim lesions; further, when also considering data from myelin water imaging, the decrease of myelin is likely to drive the increase of QSM values within WM lesions. CONCLUSIONS: We could provide meta-analytic evidence for DGM susceptibility changes in MS compared to HC; basal ganglia susceptibility is increased and, in the putamen, associated with disability, while thalamic susceptibility is decreased. Beyond these findings, further investigations are necessary to establish the role of QSM in MS for research or even clinical routine.

[Expert recommendations for magnetic resonance imaging of muscle disorders]. / Expertenempfehlung zur Magnetresonanztomographie bei Muskelerkrankungen.

BACKGROUND: Magnetic resonance (MRI) imaging of the skeletal muscles (muscle MRI for short) is increasingly being used in clinical routine for diagnosis and longitudinal assessment of muscle disorders. However, cross-centre standards for measurement protocol and radiological assessment are still lacking. OBJECTIVES: The aim of this expert recommendation is to present standards for the application and interpretation of muscle MRI in hereditary and inflammatory muscle disorders. METHODS: This work was developed in collaboration between neurologists, neuroradiologists, radiologists, neuropaediatricians, neuroscientists and MR physicists from different university hospitals in Germany. The recommendations are based on expert knowledge and a focused literature search. RESULTS: The indications for muscle MRI are explained, including the detection and monitoring of structural tissue changes and oedema in the muscle, as well as the identification of a suitable biopsy site. Recommendations for the examination procedure and selection of appropriate MRI sequences are given. Finally, steps for a structured radiological assessment are presented. CONCLUSIONS: The present work provides concrete recommendations for the indication, implementation and interpretation of muscle MRI in muscle disorders. Furthermore, it provides a possible basis for the standardisation of the measurement protocols at all clinical centres in Germany.

[Expert recommendations for magnetic resonance imaging of muscle disorders]. / Expertenempfehlung zur Magnetresonanztomographie bei Muskelerkrankungen.

BACKGROUND: Magnetic resonance (MRI) imaging of the skeletal muscles (muscle MRI for short) is increasingly being used in clinical routine for diagnosis and longitudinal assessment of muscle disorders. However, cross-centre standards for measurement protocol and radiological assessment are still lacking. OBJECTIVES: The aim of this expert recommendation is to present standards for the application and interpretation of muscle MRI in hereditary and inflammatory muscle disorders. METHODS: This work was developed in collaboration between neurologists, neuroradiologists, radiologists, neuropaediatricians, neuroscientists and MR physicists from different university hospitals in Germany. The recommendations are based on expert knowledge and a focused literature search. RESULTS: The indications for muscle MRI are explained, including the detection and monitoring of structural tissue changes and oedema in the muscle, as well as the identification of a suitable biopsy site. Recommendations for the examination procedure and selection of appropriate MRI sequences are given. Finally, steps for a structured radiological assessment are presented. CONCLUSIONS: The present work provides concrete recommendations for the indication, implementation and interpretation of muscle MRI in muscle disorders. Furthermore, it provides a possible basis for the standardisation of the measurement protocols at all clinical centres in Germany.

LST-AI: a Deep Learning Ensemble for Accurate MS Lesion Segmentation.

Automated segmentation of brain white matter lesions is crucial for both clinical assessment and scientific research in multiple sclerosis (MS). Over a decade ago, we introduced an engineered lesion segmentation tool, LST. While recent lesion segmentation approaches have leveraged artificial intelligence (AI), they often remain proprietary and difficult to adopt. As an open-source tool, we present LST-AI, an advanced deep learning-based extension of LST that consists of an ensemble of three 3D-UNets. LST-AI explicitly addresses the imbalance between white matter (WM) lesions and non-lesioned WM. It employs a composite loss function incorporating binary cross-entropy and Tversky loss to improve segmentation of the highly heterogeneous MS lesions. We train the network ensemble on 491 MS pairs of T1w and FLAIR images, collected in-house from a 3T MRI scanner, and expert neuroradiologists manually segmented the utilized lesion maps for training. LST-AI additionally includes a lesion location annotation tool, labeling lesion location according to the 2017 McDonald criteria (periventricular, infratentorial, juxtacortical, subcortical). We conduct evaluations on 103 test cases consisting of publicly available data using the Anima segmentation validation tools and compare LST-AI with several publicly available lesion segmentation models. Our empirical analysis shows that LST-AI achieves superior performance compared to existing methods. Its Dice and F1 scores exceeded 0.62, outperforming LST, SAMSEG (Sequence Adaptive Multimodal SEGmentation), and the popular nnUNet framework, which all scored below 0.56. Notably, LST-AI demonstrated exceptional performance on the MSSEG-1 challenge dataset, an international WM lesion segmentation challenge, with a Dice score of 0.65 and an F1 score of 0.63-surpassing all other competing models at the time of the challenge. With increasing lesion volume, the lesion detection rate rapidly increased with a detection rate of >75% for lesions with a volume between 10mm3 and 100mm3. Given its higher segmentation performance, we recommend that research groups currently using LST transition to LST-AI. To facilitate broad adoption, we are releasing LST-AI as an open-source model, available as a command-line tool, dockerized container, or Python script, enabling diverse applications across multiple platforms.

Denoising diffusion-based MRI to CT image translation enables automated spinal segmentation.

BACKGROUND: Automated segmentation of spinal magnetic resonance imaging (MRI) plays a vital role both scientifically and clinically. However, accurately delineating posterior spine structures is challenging. METHODS: This retrospective study, approved by the ethical committee, involved translating T1-weighted and T2-weighted images into computed tomography (CT) images in a total of 263 pairs of CT/MR series. Landmark-based registration was performed to align image pairs. We compared two-dimensional (2D) paired - Pix2Pix, denoising diffusion implicit models (DDIM) image mode, DDIM noise mode - and unpaired (SynDiff, contrastive unpaired translation) image-to-image translation using "peak signal-to-noise ratio" as quality measure. A publicly available segmentation network segmented the synthesized CT datasets, and Dice similarity coefficients (DSC) were evaluated on in-house test sets and the "MRSpineSeg Challenge" volumes. The 2D findings were extended to three-dimensional (3D) Pix2Pix and DDIM. RESULTS: 2D paired methods and SynDiff exhibited similar translation performance and DCS on paired data. DDIM image mode achieved the highest image quality. SynDiff, Pix2Pix, and DDIM image mode demonstrated similar DSC (0.77). For craniocaudal axis rotations, at least two landmarks per vertebra were required for registration. The 3D translation outperformed the 2D approach, resulting in improved DSC (0.80) and anatomically accurate segmentations with higher spatial resolution than that of the original MRI series. CONCLUSIONS: Two landmarks per vertebra registration enabled paired image-to-image translation from MRI to CT and outperformed all unpaired approaches. The 3D techniques provided anatomically correct segmentations, avoiding underprediction of small structures like the spinous process. RELEVANCE STATEMENT: This study addresses the unresolved issue of translating spinal MRI to CT, making CT-based tools usable for MRI data. It generates whole spine segmentation, previously unavailable in MRI, a prerequisite for biomechanical modeling and feature extraction for clinical applications. KEY POINTS: • Unpaired image translation lacks in converting spine MRI to CT effectively. • Paired translation needs registration with two landmarks per vertebra at least. • Paired image-to-image enables segmentation transfer to other domains. • 3D translation enables super resolution from MRI to CT. • 3D translation prevents underprediction of small structures.

Compositional and Functional MRI of Skeletal Muscle: A Review.

Due to its exceptional sensitivity to soft tissues, MRI has been extensively utilized to assess anatomical muscle parameters such as muscle volume and cross-sectional area. Quantitative Magnetic Resonance Imaging (qMRI) adds to the capabilities of MRI, by providing information on muscle composition such as fat content, water content, microstructure, hypertrophy, atrophy, as well as muscle architecture. In addition to compositional changes, qMRI can also be used to assess function for example by measuring muscle quality or through characterization of muscle deformation during passive lengthening/shortening and active contractions. The overall aim of this review is to provide an updated overview of qMRI techniques that can quantitatively evaluate muscle structure and composition, provide insights into the underlying biological basis of the qMRI signal, and illustrate how qMRI biomarkers of muscle health relate to function in healthy and diseased/injured muscles. While some applications still require systematic clinical validation, qMRI is now established as a comprehensive technique, that can be used to characterize a wide variety of structural and compositional changes in healthy and diseased skeletal muscle. Taken together, multiparametric muscle MRI holds great potential in the diagnosis and monitoring of muscle conditions in research and clinical applications. EVIDENCE LEVEL: 5 TECHNICAL EFFICACY: Stage 2.

AI-based detection of contrast-enhancing MRI lesions in patients with multiple sclerosis.

BACKGROUND: Contrast-enhancing (CE) lesions are an important finding on brain magnetic resonance imaging (MRI) in patients with multiple sclerosis (MS) but can be missed easily. Automated solutions for reliable CE lesion detection are emerging; however, independent validation of artificial intelligence (AI) tools in the clinical routine is still rare. METHODS: A three-dimensional convolutional neural network for CE lesion segmentation was trained externally on 1488 datasets of 934 MS patients from 81 scanners using concatenated information from FLAIR and T1-weighted post-contrast imaging. This externally trained model was tested on an independent dataset comprising 504 T1-weighted post-contrast and FLAIR image datasets of MS patients from clinical routine. Two neuroradiologists (R1, R2) labeled CE lesions for gold standard definition in the clinical test dataset. The algorithmic output was evaluated on both patient- and lesion-level. RESULTS: On a patient-level, recall, specificity, precision, and accuracy of the AI tool to predict patients with CE lesions were 0.75, 0.99, 0.91, and 0.96. The agreement between the AI tool and both readers was within the range of inter-rater agreement (Cohen's kappa; AI vs. R1: 0.69; AI vs. R2: 0.76; R1 vs. R2: 0.76). On a lesion-level, false negative lesions were predominately found in infratentorial location, significantly smaller, and at lower contrast than true positive lesions (p < 0.05). CONCLUSIONS: AI-based identification of CE lesions on brain MRI is feasible, approaching human reader performance in independent clinical data and might be of help as a second reader in the neuroradiological assessment of active inflammation in MS patients. CRITICAL RELEVANCE STATEMENT: Al-based detection of contrast-enhancing multiple sclerosis lesions approaches human reader performance, but careful visual inspection is still needed, especially for infratentorial, small and low-contrast lesions.

Implementation of GAN-Based, Synthetic T2-Weighted Fat Saturated Images in the Routine Radiological Workflow Improves Spinal Pathology Detection.

(1) Background and Purpose: In magnetic resonance imaging (MRI) of the spine, T2-weighted (T2-w) fat-saturated (fs) images improve the diagnostic assessment of pathologies. However, in the daily clinical setting, additional T2-w fs images are frequently missing due to time constraints or motion artifacts. Generative adversarial networks (GANs) can generate synthetic T2-w fs images in a clinically feasible time. Therefore, by simulating the radiological workflow with a heterogenous dataset, this study's purpose was to evaluate the diagnostic value of additional synthetic, GAN-based T2-w fs images in the clinical routine. (2) Methods: 174 patients with MRI of the spine were retrospectively identified. A GAN was trained to synthesize T2-w fs images from T1-w, and non-fs T2-w images of 73 patients scanned in our institution. Subsequently, the GAN was used to create synthetic T2-w fs images for the previously unseen 101 patients from multiple institutions. In this test dataset, the additional diagnostic value of synthetic T2-w fs images was assessed in six pathologies by two neuroradiologists. Pathologies were first graded on T1-w and non-fs T2-w images only, then synthetic T2-w fs images were added, and pathologies were graded again. Evaluation of the additional diagnostic value of the synthetic protocol was performed by calculation of Cohen's ĸ and accuracy in comparison to a ground truth (GT) grading based on real T2-w fs images, pre- or follow-up scans, other imaging modalities, and clinical information. (3) Results: The addition of the synthetic T2-w fs to the imaging protocol led to a more precise grading of abnormalities than when grading was based on T1-w and non-fs T2-w images only (mean ĸ GT versus synthetic protocol = 0.65; mean ĸ GT versus T1/T2 = 0.56; p = 0.043). (4) Conclusions: The implementation of synthetic T2-w fs images in the radiological workflow significantly improves the overall assessment of spine pathologies. Thereby, high-quality, synthetic T2-w fs images can be virtually generated by a GAN from heterogeneous, multicenter T1-w and non-fs T2-w contrasts in a clinically feasible time, which underlines the reproducibility and generalizability of our approach.

Synthetic T2-weighted fat sat based on a generative adversarial network shows potential for scan time reduction in spine imaging in a multicenter test dataset.

OBJECTIVES: T2-weighted (w) fat sat (fs) sequences, which are important in spine MRI, require a significant amount of scan time. Generative adversarial networks (GANs) can generate synthetic T2-w fs images. We evaluated the potential of synthetic T2-w fs images by comparing them to their true counterpart regarding image and fat saturation quality, and diagnostic agreement in a heterogenous, multicenter dataset. METHODS: A GAN was used to synthesize T2-w fs from T1- and non-fs T2-w. The training dataset comprised scans of 73 patients from two scanners, and the test dataset, scans of 101 patients from 38 multicenter scanners. Apparent signal- and contrast-to-noise ratios (aSNR/aCNR) were measured in true and synthetic T2-w fs. Two neuroradiologists graded image (5-point scale) and fat saturation quality (3-point scale). To evaluate whether the T2-w fs images are indistinguishable, a Turing test was performed by eleven neuroradiologists. Six pathologies were graded on the synthetic protocol (with synthetic T2-w fs) and the original protocol (with true T2-w fs) by the two neuroradiologists. RESULTS: aSNR and aCNR were not significantly different between the synthetic and true T2-w fs images. Subjective image quality was graded higher for synthetic T2-w fs (p = 0.023). In the Turing test, synthetic and true T2-w fs could not be distinguished from each other. The intermethod agreement between synthetic and original protocol ranged from substantial to almost perfect agreement for the evaluated pathologies. DISCUSSION: The synthetic T2-w fs might replace a physical T2-w fs. Our approach validated on a challenging, multicenter dataset is highly generalizable and allows for shorter scan protocols. KEY POINTS: • Generative adversarial networks can be used to generate synthetic T2-weighted fat sat images from T1- and non-fat sat T2-weighted images of the spine. • The synthetic T2-weighted fat sat images might replace a physically acquired T2-weighted fat sat showing a better image quality and excellent diagnostic agreement with the true T2-weighted fat images. • The present approach validated on a challenging, multicenter dataset is highly generalizable and allows for significantly shorter scan protocols.

Longitudinal Assessment of Multiple Sclerosis Lesion Load With Synthetic Magnetic Resonance Imaging-A Multicenter Validation Study.

INTRODUCTION: Double inversion recovery (DIR) has been validated as a sensitive magnetic resonance imaging (MRI) contrast in multiple sclerosis (MS). Deep learning techniques can use basic input data to generate synthetic DIR (synthDIR) images that are on par with their acquired counterparts. As assessment of longitudinal MRI data is paramount in MS diagnostics, our study's purpose is to evaluate the utility of synthDIR longitudinal subtraction imaging for detection of disease progression in a multicenter data set of MS patients. METHODS: We implemented a previously established generative adversarial network to synthesize DIR from input T1-weighted and fluid-attenuated inversion recovery (FLAIR) sequences for 214 MRI data sets from 74 patients and 5 different centers. One hundred and forty longitudinal subtraction maps of consecutive scans (follow-up scan-preceding scan) were generated for both acquired FLAIR and synthDIR. Two readers, blinded to the image origin, independently quantified newly formed lesions on the FLAIR and synthDIR subtraction maps, grouped into specific locations as outlined in the McDonald criteria. RESULTS: Both readers detected significantly more newly formed MS-specific lesions in the longitudinal subtractions of synthDIR compared with acquired FLAIR (R1: 3.27 ± 0.60 vs 2.50 ± 0.69 [ P = 0.0016]; R2: 3.31 ± 0.81 vs 2.53 ± 0.72 [ P < 0.0001]). Relative gains in detectability were most pronounced in juxtacortical lesions (36% relative gain in lesion counts-pooled for both readers). In 5% of the scans, synthDIR subtraction maps helped to identify a disease progression missed on FLAIR subtraction maps. CONCLUSIONS: Generative adversarial networks can generate high-contrast DIR images that may improve the longitudinal follow-up assessment in MS patients compared with standard sequences. By detecting more newly formed MS lesions and increasing the rates of detected disease activity, our methodology promises to improve clinical decision-making.

On the water-fat in-phase assumption for quantitative susceptibility mapping.

PURPOSE: To (a) define multi-peak fat model-based effective in-phase echo times for quantitative susceptibility mapping (QSM) in water-fat regions, (b) analyze the relationship between fat fraction, field map quantification bias and susceptibility bias, and (c) evaluate the susceptibility mapping performance of the proposed effective in-phase echoes in comparison to single-peak in-phase echoes and water-fat separation for regions where both water and fat are present. METHODS: Effective multipeak in-phase echo times for a bone marrow and a liver fat spectral model were derived from a single voxel simulation. A Monte Carlo simulation was performed to assess the field map estimation error as a function of fat fraction for the different in-phase echoes. Additionally, a phantom scan and in vivo scans in the liver, spine, and breast were performed and evaluated with respect to quantification accuracy. RESULTS: The use of single-peak in-phase echoes can introduce a worst-case susceptibility bias of 0.43 $$ 0.43 $$  ppm. The use of effective multipeak in-phase echoes shows a similar quantitative performance in the numerical simulation, the phantom and in all in vivo anatomies when compared to water-fat separation-based QSM. CONCLUSION: QSM based on the proposed effective multipeak in-phase echoes can alleviate the quantification bias present in QSM based on single-peak in-phase echoes. When compared to water-fat separation-based QSM the proposed effective in-phase echo times achieve a similar quantitative performance while drastically reducing the computational expense for field map estimation.

Comparing myelin-sensitive magnetic resonance imaging measures and resulting g-ratios in healthy and multiple sclerosis brains.

The myelin concentration and the degree of myelination of nerve fibers can provide valuable information on the integrity of human brain tissue. Magnetic resonance imaging (MRI) of myelin-sensitive parameters can help to non-invasively evaluate demyelinating diseases such as multiple sclerosis (MS). Several different myelin-sensitive MRI methods have been proposed to determine measures of the degree of myelination, in particular the g-ratio. However, variability in underlying physical principles and different biological models influence measured myelin concentrations, and consequently g-ratio values. We therefore investigated similarities and differences between five different myelin-sensitive MRI measures and their effects on g-ratio mapping in the brains of both MS patients and healthy volunteers. We compared two different estimates of the myelin water fraction (MWF) as well as the inhomogeneous magnetization transfer ratio (ihMTR), magnetization transfer saturation (MTsat), and macromolecular tissue volume (MTV) in 13 patients with MS and 14 healthy controls. In combination with diffusion-weighted imaging, we derived g-ratio parameter maps for each of the five different myelin measures. The g-ratio values calculated from different myelin measures varied strongly, especially in MS lesions. While, compared to normal-appearing white matter, MTsat and one estimate of the MWF resulted in higher g-ratio values within lesions, ihMTR, MTV, and the second MWF estimate resulted in lower lesion g-ratio values. As myelin-sensitive measures provide rough estimates of myelin content rather than absolute myelin concentrations, resulting g-ratio values strongly depend on the utilized myelin measure and model used for g-ratio mapping. When comparing g-ratio values, it is, thus, important to utilize the same MRI methods and models or to consider methodological differences. Particular caution is necessary in pathological tissue such as MS lesions.

Impact of normative brain volume reports on the diagnosis of neurodegenerative dementia disorders in neuroradiology: A real-world, clinical practice study.

Background: Normative brain volume reports (NBVR) are becoming more available in the work-up of patients with suspected dementia disorders, potentially leveraging the value of structural MRI in clinical settings. The present study aims to investigate the impact of NBVRs on the diagnosis of neurodegenerative dementia disorders in real-world clinical practice. Methods: We retrospectively analyzed data of 112 memory clinic patients, who were consecutively referred for MRI and 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) during a 12-month period. Structural MRI was assessed by two residents with 2 and 3 years of neuroimaging experience. Statements and diagnostic confidence regarding the presence of a neurodegenerative disorder in general (first level) and Alzheimer's disease (AD) pattern in particular (second level) were recorded without and with NBVR information. FDG-PET served as the reference standard. Results: Overall, despite a trend towards increased accuracy, the impact of NBVRs on diagnostic accuracy was low and non-significant. We found a significant drop of sensitivity (0.75-0.58; p < 0.001) and increase of specificity (0.62-0.85; p < 0.001) for rater 1 at identifying patients with neurodegenerative dementia disorders. Diagnostic confidence increased for rater 2 (p < 0.001). Conclusions: Overall, NBVRs had a limited impact on diagnostic accuracy in real-world clinical practice. Potentially, NBVR might increase diagnostic specificity and confidence of neuroradiology residents. To this end, a well-defined framework for integration of NBVR in the diagnostic process and improved algorithms of NBVR generation are essential.

Beyond mean value analysis - a voxel-based analysis of the quantitative MR biomarker water T2 in the presence of fatty infiltration in skeletal muscle tissue of patients with neuromuscular diseases.

The main pathologies in the muscles of patients with neuromuscular diseases (NMD) are fatty infiltration and edema. Recently, quantitative magnetic resonance (MR) imaging for determination of the MR biomarkers proton density fat fraction (PDFF) and water T2 (T2w ) has been advanced. Biophysical effects or pathology can have different effects on MR biomarkers. Thus, for heterogeneously affected muscles, the routinely performed mean or median value analyses of MR biomarkers are questionable. Our work presents a voxel-based histogram analysis of PDFF and T2w images to point out potential quantification errors. In 12 patients with NMD, chemical-shift encoding-based water-fat imaging for PDFF and T2 mapping with spectral adiabatic inversion recovery (SPAIR) for T2w determination was performed. Segmentation of nine thigh muscles was performed bilaterally (n = 216). PDFF and T2 maps were coregistered. A voxel-based comparison of PDFF and T2w showed a decreased T2w with increasing PDFF. Mean T2w and mean T2w without fatty voxels (PDFF < 10%) show good agreement, whereas standard deviation (σ) T2w and σ T2w without fatty voxels show increasing difference with increasing values of σ. Thereby two subgroups can be observed, referring to muscles in which the exclusion of fatty voxels has a negligible influence versus muscles in which a strong dependency of the T2w value distribution on the exclusion of fatty voxels is present. Because of the two opposite effects that influence T2w in a voxel, namely, (i) a pathophysiologically increased water mobility leading to T2w elevation, and (ii) a dependency of T2w on the PDFF leading to decreased T2w , the T2w distribution within a muscle might be heterogenous and the routine mean or median analysis can lead to a misinterpretation of the muscle health. It was concluded that muscle T2w mean values can wrongly suggest healthy muscle tissue. A deeper analysis of the underlying value distribution is necessary. Therefore, a quantitative analysis of T2w histograms is a potential alternative.

Uncertainty-Aware and Lesion-Specific Image Synthesis in Multiple Sclerosis Magnetic Resonance Imaging: A Multicentric Validation Study.

Generative adversarial networks (GANs) can synthesize high-contrast MRI from lower-contrast input. Targeted translation of parenchymal lesions in multiple sclerosis (MS), as well as visualization of model confidence further augment their utility, provided that the GAN generalizes reliably across different scanners. We here investigate the generalizability of a refined GAN for synthesizing high-contrast double inversion recovery (DIR) images and propose the use of uncertainty maps to further enhance its clinical utility and trustworthiness. A GAN was trained to synthesize DIR from input fluid-attenuated inversion recovery (FLAIR) and T1w of 50 MS patients (training data). In another 50 patients (test data), two blinded readers (R1 and R2) independently quantified lesions in synthetic DIR (synthDIR), acquired DIR (trueDIR) and FLAIR. Of the 50 test patients, 20 were acquired on the same scanner as training data (internal data), while 30 were scanned at different scanners with heterogeneous field strengths and protocols (external data). Lesion-to-Background ratios (LBR) for MS-lesions vs. normal appearing white matter, as well as image quality parameters were calculated. Uncertainty maps were generated to visualize model confidence. Significantly more MS-specific lesions were found in synthDIR compared to FLAIR (R1: 26.7 ± 2.6 vs. 22.5 ± 2.2 p < 0.0001; R2: 22.8 ± 2.2 vs. 19.9 ± 2.0, p = 0.0005). While trueDIR remained superior to synthDIR in R1 [28.6 ± 2.9 vs. 26.7 ± 2.6 (p = 0.0021)], both sequences showed comparable lesion conspicuity in R2 [23.3 ± 2.4 vs. 22.8 ± 2.2 (p = 0.98)]. Importantly, improvements in lesion counts were similar in internal and external data. Measurements of LBR confirmed that lesion-focused GAN training significantly improved lesion conspicuity. The use of uncertainty maps furthermore helped discriminate between MS lesions and artifacts. In conclusion, this multicentric study confirms the external validity of a lesion-focused Deep-Learning tool aimed at MS imaging. When implemented, uncertainty maps are promising to increase the trustworthiness of synthetic MRI.

Faster and Better: How Anomaly Detection Can Accelerate and Improve Reporting of Head Computed Tomography.

BACKGROUND: Most artificial intelligence (AI) systems are restricted to solving a pre-defined task, thus limiting their generalizability to unselected datasets. Anomaly detection relieves this shortfall by flagging all pathologies as deviations from a learned norm. Here, we investigate whether diagnostic accuracy and reporting times can be improved by an anomaly detection tool for head computed tomography (CT), tailored to provide patient-level triage and voxel-based highlighting of pathologies. METHODS: Four neuroradiologists with 1-10 years of experience each investigated a set of 80 routinely acquired head CTs containing 40 normal scans and 40 scans with common pathologies. In a random order, scans were investigated with and without AI-predictions. A 4-week wash-out period between runs was included to prevent a reminiscence effect. Performance metrics for identifying pathologies, reporting times, and subjectively assessed diagnostic confidence were determined for both runs. RESULTS: AI-support significantly increased the share of correctly classified scans (normal/pathological) from 309/320 scans to 317/320 scans (p = 0.0045), with a corresponding sensitivity, specificity, negative- and positive- predictive value of 100%, 98.1%, 98.2% and 100%, respectively. Further, reporting was significantly accelerated with AI-support, as evidenced by the 15.7% reduction in reporting times (65.1 ± 8.9 s vs. 54.9 ± 7.1 s; p < 0.0001). Diagnostic confidence was similar in both runs. CONCLUSION: Our study shows that AI-based triage of CTs can improve the diagnostic accuracy and accelerate reporting for experienced and inexperienced radiologists alike. Through ad hoc identification of normal CTs, anomaly detection promises to guide clinicians towards scans requiring urgent attention.

Quantitative evaluation of the influence of multiple MRI sequences and of pathological tissues on the registration of longitudinal data acquired during brain tumor treatment.

Registration methods facilitate the comparison of multiparametric magnetic resonance images acquired at different stages of brain tumor treatments. Image-based registration solutions are influenced by the sequences chosen to compute the distance measure, and the lack of image correspondences due to the resection cavities and pathological tissues. Nonetheless, an evaluation of the impact of these input parameters on the registration of longitudinal data is still missing. This work evaluates the influence of multiple sequences, namely T1-weighted (T1), T2-weighted (T2), contrast enhanced T1-weighted (T1-CE), and T2 Fluid Attenuated Inversion Recovery (FLAIR), and the exclusion of the pathological tissues on the non-rigid registration of pre- and post-operative images. We here investigate two types of registration methods, an iterative approach and a convolutional neural network solution based on a 3D U-Net. We employ two test sets to compute the mean target registration error (mTRE) based on corresponding landmarks. In the first set, markers are positioned exclusively in the surroundings of the pathology. The methods employing T1-CE achieves the lowest mTREs, with a improvement up to 0.8 mm for the iterative solution. The results are higher than the baseline when using the FLAIR sequence. When excluding the pathology, lower mTREs are observable for most of the methods. In the second test set, corresponding landmarks are located in the entire brain volumes. Both solutions employing T1-CE obtain the lowest mTREs, with a decrease up to 1.16 mm for the iterative method, whereas the results worsen using the FLAIR. When excluding the pathology, an improvement is observable for the CNN method using T1-CE. Both approaches utilizing the T1-CE sequence obtain the best mTREs, whereas the FLAIR is the least informative to guide the registration process. Besides, the exclusion of pathology from the distance measure computation improves the registration of the brain tissues surrounding the tumor. Thus, this work provides the first numerical evaluation of the influence of these parameters on the registration of longitudinal magnetic resonance images, and it can be helpful for developing future algorithms.

Quantitative Muscle MRI in Patients with Neuromuscular Diseases-Association of Muscle Proton Density Fat Fraction with Semi-Quantitative Grading of Fatty Infiltration and Muscle Strength at the Thigh Region.

(1) Background and Purpose: The skeletal muscles of patients suffering from neuromuscular diseases (NMD) are affected by atrophy, hypertrophy, fatty infiltration, and edematous changes. Magnetic resonance imaging (MRI) is an important tool for diagnosis and monitoring. Concerning fatty infiltration, T1-weighted or T2-weighted DIXON turbo spin echo (TSE) sequences enable a qualitative assessment of muscle involvement. To achieve higher comparability, semi-quantitative grading scales, such as the four-point Mercuri scale, are commonly applied. However, the evaluation remains investigator-dependent. Therefore, effort is being invested to develop quantitative MRI techniques for determination of imaging markers such as the proton density fat fraction (PDFF). The present work aims to assess the diagnostic value of PDFF in correlation to Mercuri grading and clinically determined muscle strength in patients with myotonic dystrophy type 2 (DM2), limb girdle muscular dystrophy type 2A (LGMD2A), and adult Pompe disease. (2) Methods: T2-weighted two-dimensional (2D) DIXON TSE and chemical shift encoding-based water-fat MRI were acquired in 13 patients (DM2: n = 5; LGMD2A: n = 5; Pompe disease: n = 3). Nine different thigh muscles were rated in all patients according to the Mercuri grading and segmented to extract PDFF values. Muscle strength was assessed according to the British Medical Research Council (BMRC) scale. For correlation analyses between Mercuri grading, muscle strength, and PDFF, the Spearman correlation coefficient (rs) was computed. (3) Results: Mean PDFF values ranged from 7% to 37% in adults with Pompe disease and DM2 and up to 79% in LGMD2A patients. In all three groups, a strong correlation of the Mercuri grading and PDFF values was observed for almost all muscles (rs > 0.70, p < 0.05). PDFF values correlated significantly to muscle strength for muscle groups responsible for knee flexion (rs = -0.80, p < 0.01). (4) Conclusion: In the small, investigated patient cohort, PDFF offers similar diagnostic precision as the clinically established Mercuri grading. Based on these preliminary data, PDFF could be further considered as an MRI-based biomarker in the assessment of fatty infiltration of muscle tissue in NMD. Further studies with larger patient cohorts are needed to advance PDFF as an MRI-based biomarker in NMD, with advantages such as its greater dynamic range, enabling the assessment of subtler changes, the amplified objectivity, and the potential of direct correlation to muscle function for selected muscles.

Regional variation of thigh muscle fat infiltration in patients with neuromuscular diseases compared to healthy controls.

BACKGROUND: Chemical shift encoding-based water-fat magnetic resonance imaging (CSE-MRI) measures a quantitative biomarker: the proton density fat fraction (PDFF). The aim was to assess regional and proximo-distal PDFF variations at the thigh in patients with myotonic dystrophy type 2 (DM2), limb-girdle muscular dystrophy type 2A (LGMD2A), and late-onset Pompe disease (LOPD) as compared to healthy controls. METHODS: Seven patients (n=2 DM2, n=2 LGMD2A, n=3 LOPD) and 20 controls were recruited. A 3D-spoiled gradient echo sequence was used to scan the thigh musculature. Muscles were manually segmented to generate mean muscle PDFF. RESULTS: In all three disease entities, there was an increase in muscle fat replacement compared to healthy controls. However, within each disease group, there were patients with a shorter time since symptom onset that only showed mild PDFF elevation (range, 10% to 20%) compared to controls (P≤0.05), whereas patients with a longer period since symptom onset showed a more severe grade of fat replacement with a range of 50% to 70% (P<0.01). Increased PDFF of around 5% was observed for vastus medialis, semimembranosus and gracilis muscles in advanced compared to early DM2. LGMD2A_1 showed an early disease stage with predominantly mild PDFF elevations over all muscles and levels (10.9%±7.1%) compared to controls. The quadriceps, gracilis and biceps femoris muscles showed the highest difference between LGMD2A_1 with 5 years since symptom onset (average PDFF 11.1%±6.9%) compared to LGMD2A_2 with 32 years since symptom onset (average PDFF 66.3%±6.3%). For LOPD patients, overall PDFF elevations were observed in all major hip flexors and extensors (range, 25.8% to 30.8%) compared to controls (range, 1.7% to 2.3%, P<0.05). Proximal-to-distal PDFF highly varied within and between diseases and within controls. The intra-reader reliability was high (reproducibility coefficient ≤2.19%). CONCLUSIONS: By quantitatively measuring muscle fat infiltration at the thigh, we identified candidate muscles for disease monitoring due to their gradual PDFF elevation with longer disease duration. Regional variation between proximal, central, and distal muscle PDFF was high and is important to consider when performing longitudinal MRI follow-ups in the clinical setting or in longitudinal studies.