Charidimos Tsagkas
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Charidimos Tsagkas has a total of 124 co-authors. Here are the top co-authors by number of shared publications:
- Cristina Granziera
Translational Imaging in Neurology (ThINk) Basel, Department of Biomedical Engineering, Faculty of Medicine, University Hospital Basel and University of Basel, Switzerland
9 shared publications - Alessandro Cagol
Translational Imaging in Neurology (ThINk) Basel, Department of Biomedical Engineering, Faculty of Medicine, University Hospital Basel and University of Basel, Switzerland
6 shared publications - Ludwig Kappos
University Hospital Basel, Switzerland
6 shared publications Katrin Parmar
Neurologic Clinic and Policlinic, Departments of Head, Spine and Neuromedicine, Clinical Research and Biomedical Engineering, University Hospital Basel and University of Basel, Basel, Switzerland; Translational Imaging in Neurology (ThINk) Basel, Departments of Head, Spine and Neuromedicine and Biomedical Engineering, University Hospital Basel and University of Basel, Basel, Switzerland; Reha Rheinfelden, Rheinfelden, Switzerland
5 shared publicationsLudwig Kappos
Neurologic Clinic and Policlinic, Departments of Head, Spine and Neuromedicine, Clinical Research and Biomedical Engineering, University Hospital Basel and University of Basel, Basel, Switzerland; Research Center for Clinical Neuroimmunology and Neuroscience Basel (RC2NB), Departments of Head, Spine and Neuromedicine, Clinical Research and Biomedical Engineering, University Hospital Basel and University of Basel, Basel, Switzerland
5 shared publications- Matthias Weigel
Translational Imaging in Neurology (ThINK) Basel, Department of Biomedical Engineering, Faculty of Medicine, University Hospital Basel and University of Basel, Basel, Switzerland
5 shared publications
Latest Publications
- Open Access
- 9 October 2026
- Published by SAGE Publications in Multiple Sclerosis Journal – Experimental, Translational and Clinical
- Vol. 12 (4)
Macro- and microstructural changes in the upper cervical spinal cord and their clinical correlates in multiple sclerosis
Background: Spinal cord (SC) damage is central to multiple sclerosis (MS) progression but remains difficult to quantify with conventional magnetic resonance imaging (MRI). Objectives: To characterize upper cervical SC pathology using quantitative MRI (qMRI) and assess longitudinal changes and predictors of clinical progression and fluid biomarker trajectories. Methods: We conducted a prospective study including people with MS (PwMS) and healthy controls (HCs). Participants underwent brain 3T-MRI, clinical assessments, and serum biomarker analyses at baseline and 2 years, with clinical follow-up to a median of 3.0 years. SC lesion burden, morphometry, quantitative T1 (qT1), and magnetization transfer saturation (MTsat) metrics were derived. Results: We enrolled 183 PwMS and 100 HCs at baseline, with follow-up for a subset. PwMS showed widespread SC macro/microstructural damage compared with HCs, with qT1 at C1-C4 and MTsat at C4 providing the highest discrimination (model-area under the curve = 0.81). Disability was most strongly associated with cross-sectional area at C1–C2, and qT1 at C1, C2, and C4 ( R 2 = 0.49). Baseline SC metrics predicted disability worsening and progression independent of relapse activity, with lower transverse diameters at C3–C4 as key predictors (model-HRs = 0.59–0.69), and worsening serum glial fibrillary acidic protein but not serum neurofilament light chain. Conclusion: Upper cervical SC morphometry and qMRI capture clinically relevant pathology and are linked to disease evolution, supporting their potential prognostic value in MS.
- 9 October 2026
- Published by SAGE Publications in Multiple Sclerosis Journal – Experimental, Translational and Clinical
- Vol. 12 (4)
Open AccessMacro- and microstructural changes in the upper cervical spinal cord and their clinical correlates in multiple sclerosis
Background: Spinal cord (SC) damage is central to multiple sclerosis (MS) progression but remains difficult to quantify with conventional magnetic resonance imaging (MRI). Objectives: To characterize upper cervical SC pathology using quantitative MRI (qMRI) and assess longitudinal changes and predictors of clinical progression and fluid biomarker trajectories. Methods: We conducted a prospective study including people with MS (PwMS) and healthy controls (HCs). Participants underwent brain 3T-MRI, clinical assessments, and serum biomarker analyses at baseline and 2 years, with clinical follow-up to a median of 3.0 years. SC lesion burden, morphometry, quantitative T1 (qT1), and magnetization transfer saturation (MTsat) metrics were derived. Results: We enrolled 183 PwMS and 100 HCs at baseline, with follow-up for a subset. PwMS showed widespread SC macro/microstructural damage compared with HCs, with qT1 at C1-C4 and MTsat at C4 providing the highest discrimination (model-area under the curve = 0.81). Disability was most strongly associated with cross-sectional area at C1–C2, and qT1 at C1, C2, and C4 ( R 2 = 0.49). Baseline SC metrics predicted disability worsening and progression independent of relapse activity, with lower transverse diameters at C3–C4 as key predictors (model-HRs = 0.59–0.69), and worsening serum glial fibrillary acidic protein but not serum neurofilament light chain. Conclusion: Upper cervical SC morphometry and qMRI capture clinically relevant pathology and are linked to disease evolution, supporting their potential prognostic value in MS.
- Open Access
- 22 April 2026
- Published by Springer Nature in Communications Medicine
- Vol. 6
A pipeline to characterize spinal cord pathology in neurological disorders combining magnetic resonance microscopy and histopathology
Background: The length, shape, and size of the spinal cord (SC) present unique challenges for MRI, including the need for high resolution to distinguish anatomical and pathological features along its span. Postmortem MRI offers an opportunity to map SC tissue abnormalities and investigate their histological correlates. Methods: We developed a pipeline combining postmortem microscopic resolution MRI (MR microscopy; MRM) of whole formalin-fixed SC specimens with targeted histopathological analysis. A gadolinium-based tissue preparation protocol was optimized using SC tissue from common marmosets with experimental autoimmune encephalomyelitis. A custom tissue holder and container were designed to enable postmortem MRI of the entire human SC. Human SC samples from individuals with multiple sclerosis, amyotrophic lateral sclerosis, and intracranial hemorrhage were scanned at 75 μm isotropic resolution on a 9.4 T Bruker system after gadolinium preparation. Results: MRM after gadolinium-based tissue preparation yields images with high signal- and contrast-to-noise ratio while minimizing acquisition times. MRI demonstrates fine anatomical detail and pathological features, including demyelination and neurodegeneration throughout the SC. A complementary custom-made cutting rack enables targeted histological sectioning of MRI-identified regions. This approach provides precise spatial correspondence between imaging and histological findings, demonstrating strong agreement across modalities. Conclusions: In summary, this pipeline facilitates comprehensive SC assessment by integrating MRM with histology. It enables accurate localization of both subtle and widespread SC pathology and enhances interpretation of MRI signals in the context of neurodegenerative and inflammatory diseases.
- 22 April 2026
- Published by Springer Nature in Communications Medicine
- Vol. 6
Open AccessA pipeline to characterize spinal cord pathology in neurological disorders combining magnetic resonance microscopy and histopathology
Background: The length, shape, and size of the spinal cord (SC) present unique challenges for MRI, including the need for high resolution to distinguish anatomical and pathological features along its span. Postmortem MRI offers an opportunity to map SC tissue abnormalities and investigate their histological correlates. Methods: We developed a pipeline combining postmortem microscopic resolution MRI (MR microscopy; MRM) of whole formalin-fixed SC specimens with targeted histopathological analysis. A gadolinium-based tissue preparation protocol was optimized using SC tissue from common marmosets with experimental autoimmune encephalomyelitis. A custom tissue holder and container were designed to enable postmortem MRI of the entire human SC. Human SC samples from individuals with multiple sclerosis, amyotrophic lateral sclerosis, and intracranial hemorrhage were scanned at 75 μm isotropic resolution on a 9.4 T Bruker system after gadolinium preparation. Results: MRM after gadolinium-based tissue preparation yields images with high signal- and contrast-to-noise ratio while minimizing acquisition times. MRI demonstrates fine anatomical detail and pathological features, including demyelination and neurodegeneration throughout the SC. A complementary custom-made cutting rack enables targeted histological sectioning of MRI-identified regions. This approach provides precise spatial correspondence between imaging and histological findings, demonstrating strong agreement across modalities. Conclusions: In summary, this pipeline facilitates comprehensive SC assessment by integrating MRM with histology. It enables accurate localization of both subtle and widespread SC pathology and enhances interpretation of MRI signals in the context of neurodegenerative and inflammatory diseases.
- Open Access
- 15 July 2025
- Published by Elsevier in NeuroImage: Clinical
- Vol. 48, 103818
Fluid and White Matter Suppression contrasts MRI improves Deep Learning detection of Multiple Sclerosis Cortical Lesions
- 15 July 2025
- Published by Elsevier in NeuroImage: Clinical
- Vol. 48, 103818
Open AccessFluid and White Matter Suppression contrasts MRI improves Deep Learning detection of Multiple Sclerosis Cortical Lesions
- Open Access
Fluid and White Matter Suppression Contrasts MRI Improves Deep Learning Detection of Multiple Sclerosis Cortical Lesions
Purpose: To investigate, for the first time, the efficacy of Fluid and White Matter Suppression (FLAWS) MRI sequence in improving Deep Learning (DL)-based detection and segmentation of cortical lesions in Multiple Sclerosis (MS) patients even, with applicability to clinical settings where only standard T1-weighted images are available. Materials and Methods: In this retrospective multi-site study, we analyzed 204 MS patients using DL models developed with FLAWS and Magnetization Prepared 2 Rapid Acquisition Gradient Echoes (MP2RAGE) sequences. Reference standard annotations were established through two approaches: (1) consensus of three expert raters across all contrasts, and (2) single-rater annotations for individual modalities. Models were validated on both internal and external datasets, with performance assessed using F 1 -score for detection and DSC for segmentation accuracy. Results: Models involving FLAWS demonstrated superior performance over MP2RAGE-only models. The combined MP2RAGE+FLAWS model achieved CL detection with median F 1 -score of 0.667[0.339 − 0.840] compared to multirater consensus. Models trained on comprehensive consensus annotations outperformed those trained on single-modality annotations. Notably, a model exclusively based on MP2RAGE images and trained with FLAWS-derived annotations showed, showed strong generalization to external Magnetization Prepared Rapid Gradient-Echo (MPRAGE) clinical datasets (median F 1 -score: 0.55[0.211 − 0.998]). Conclusion: Integration of FLAWS-derived contrasts and annotations significantly improves DL-based CL detection and segmentation. The models demonstrate capability in identifying lesions missed by individual raters and maintain robust performance even without FLAWS sequences in standard clinical settings. This advancement facilitates clinical translation, supported by publicly available inference models on DockerHub. Graphical Highlights: MP2RAGE+FLAWS MRI sequences achieves superior cortical lesion detection performance Quantitative validation shows median F 1 of 0.667[0.339 − 0.840] with combined sequences FLAWS-trained model generalizes well to standard MPRAGE images ( F 1 : 0.55[0.211 − 0.998) Model demonstrates immediate clinical applicability on standard MPRAGE without FLAWS input Implementation publicly accessible via DockerHub for widespread clinical adoption
Open AccessFluid and White Matter Suppression Contrasts MRI Improves Deep Learning Detection of Multiple Sclerosis Cortical Lesions
Purpose: To investigate, for the first time, the efficacy of Fluid and White Matter Suppression (FLAWS) MRI sequence in improving Deep Learning (DL)-based detection and segmentation of cortical lesions in Multiple Sclerosis (MS) patients even, with applicability to clinical settings where only standard T1-weighted images are available. Materials and Methods: In this retrospective multi-site study, we analyzed 204 MS patients using DL models developed with FLAWS and Magnetization Prepared 2 Rapid Acquisition Gradient Echoes (MP2RAGE) sequences. Reference standard annotations were established through two approaches: (1) consensus of three expert raters across all contrasts, and (2) single-rater annotations for individual modalities. Models were validated on both internal and external datasets, with performance assessed using F 1 -score for detection and DSC for segmentation accuracy. Results: Models involving FLAWS demonstrated superior performance over MP2RAGE-only models. The combined MP2RAGE+FLAWS model achieved CL detection with median F 1 -score of 0.667[0.339 − 0.840] compared to multirater consensus. Models trained on comprehensive consensus annotations outperformed those trained on single-modality annotations. Notably, a model exclusively based on MP2RAGE images and trained with FLAWS-derived annotations showed, showed strong generalization to external Magnetization Prepared Rapid Gradient-Echo (MPRAGE) clinical datasets (median F 1 -score: 0.55[0.211 − 0.998]). Conclusion: Integration of FLAWS-derived contrasts and annotations significantly improves DL-based CL detection and segmentation. The models demonstrate capability in identifying lesions missed by individual raters and maintain robust performance even without FLAWS sequences in standard clinical settings. This advancement facilitates clinical translation, supported by publicly available inference models on DockerHub. Graphical Highlights: MP2RAGE+FLAWS MRI sequences achieves superior cortical lesion detection performance Quantitative validation shows median F 1 of 0.667[0.339 − 0.840] with combined sequences FLAWS-trained model generalizes well to standard MPRAGE images ( F 1 : 0.55[0.211 − 0.998) Model demonstrates immediate clinical applicability on standard MPRAGE without FLAWS input Implementation publicly accessible via DockerHub for widespread clinical adoption
- Top Cited PapersOpen Access
- 21 January 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 101, 103473
Towards contrast-agnostic soft segmentation of the spinal cord
- 21 January 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 101, 103473
Top Cited PapersOpen AccessTowards contrast-agnostic soft segmentation of the spinal cord
- Open Access
- 24 September 2024
- Published by Wolters Kluwer Health in Neurology
- Vol. 103 (6) , e209604
Quantifying Remyelination Using χ-Separation in White Matter and Cortical Multiple Sclerosis Lesions
Myelin and iron play essential roles in remyelination processes of multiple sclerosis (MS) lesions. χ-separation, a novel biophysical model applied to multiecho T2*-data and T2-data, estimates the contribution of myelin and iron to the obtained susceptibility signal. We used this method to investigate myelin and iron levels in lesion and nonlesion brain areas in patients with MS and healthy individuals. This prospective MS cohort study included patients with MS fulfilling the McDonald Criteria 2017 and healthy individuals, aged 18 years or older, with no other neurologic comorbidities. Participants underwent MRI at baseline and after 2 years, including multiecho GRE-(T2*) and FAST-(T2) sequences. Using χ-separation, we generated myelin-sensitive and iron-sensitive susceptibility maps. White matter lesions (WMLs), cortical lesions (CLs), surrounding normal-appearing white matter (NAWM), and normal-appearing gray matter were segmented on fluid-attenuated inversion recovery and magnetization-prepared 2 rapid gradient echo images, respectively. Cross-sectional group comparisons used Wilcoxon rank-sum tests, longitudinal analyses applied Wilcoxon signed-rank tests. Associations with clinical outcomes (disease phenotype, age, sex, disease duration, disability measured by Expanded Disability Status Scale [EDSS], neurofilament light chain levels, and T2-lesion number and volume) were assessed using linear regression models. Of 168 patients with MS (median [interquartile range (IQR)] age 47.0 [21.7] years; 101 women; 6,898 WMLs, 775 CLs) and 103 healthy individuals (age 33.0 [10.5] years, 57 women), 108 and 62 were followed for a median of 2 years, respectively (IQR 0.1; 5,030 WMLs, 485 CLs). At baseline, WMLs had lower myelin (median 0.025 [IQR 0.015] parts per million [ppm]) and iron (0.017 [0.015] ppm) than the corresponding NAWM (myelin 0.030 [0.012]; iron 0.019 [0.011] ppm; both p < 0.001). After 2 years, both myelin (0.027 [0.014] ppm) and iron had increased (0.018 [0.015] ppm; both p < 0.001). Younger age (p < 0.001, b = −5.111 × 10−5), lower disability (p = 0.04, b = −2.352 × 10−5), and relapsing-remitting phenotype (RRMS, 0.003 [0.01] vs primary progressive 0.002 [IQR 0.01], p < 0.001; vs secondary progressive 0.0004 [IQR 0.01], p < 0.001) at baseline were associated with remyelination. Increment of myelin correlated with clinical improvement measured by EDSS (p = 0.015, b = −6.686 × 10−4). χ-separation, a novel mathematical model applied to multiecho T2*-images and T2-images shows that young RRMS patients with low disability exhibit higher remyelination capacity, which correlated with clinical disability over a 2-year follow-up.
- 24 September 2024
- Published by Wolters Kluwer Health in Neurology
- Vol. 103 (6) , e209604
Open AccessQuantifying Remyelination Using χ-Separation in White Matter and Cortical Multiple Sclerosis Lesions
Myelin and iron play essential roles in remyelination processes of multiple sclerosis (MS) lesions. χ-separation, a novel biophysical model applied to multiecho T2*-data and T2-data, estimates the contribution of myelin and iron to the obtained susceptibility signal. We used this method to investigate myelin and iron levels in lesion and nonlesion brain areas in patients with MS and healthy individuals. This prospective MS cohort study included patients with MS fulfilling the McDonald Criteria 2017 and healthy individuals, aged 18 years or older, with no other neurologic comorbidities. Participants underwent MRI at baseline and after 2 years, including multiecho GRE-(T2*) and FAST-(T2) sequences. Using χ-separation, we generated myelin-sensitive and iron-sensitive susceptibility maps. White matter lesions (WMLs), cortical lesions (CLs), surrounding normal-appearing white matter (NAWM), and normal-appearing gray matter were segmented on fluid-attenuated inversion recovery and magnetization-prepared 2 rapid gradient echo images, respectively. Cross-sectional group comparisons used Wilcoxon rank-sum tests, longitudinal analyses applied Wilcoxon signed-rank tests. Associations with clinical outcomes (disease phenotype, age, sex, disease duration, disability measured by Expanded Disability Status Scale [EDSS], neurofilament light chain levels, and T2-lesion number and volume) were assessed using linear regression models. Of 168 patients with MS (median [interquartile range (IQR)] age 47.0 [21.7] years; 101 women; 6,898 WMLs, 775 CLs) and 103 healthy individuals (age 33.0 [10.5] years, 57 women), 108 and 62 were followed for a median of 2 years, respectively (IQR 0.1; 5,030 WMLs, 485 CLs). At baseline, WMLs had lower myelin (median 0.025 [IQR 0.015] parts per million [ppm]) and iron (0.017 [0.015] ppm) than the corresponding NAWM (myelin 0.030 [0.012]; iron 0.019 [0.011] ppm; both p < 0.001). After 2 years, both myelin (0.027 [0.014] ppm) and iron had increased (0.018 [0.015] ppm; both p < 0.001). Younger age (p < 0.001, b = −5.111 × 10−5), lower disability (p = 0.04, b = −2.352 × 10−5), and relapsing-remitting phenotype (RRMS, 0.003 [0.01] vs primary progressive 0.002 [IQR 0.01], p < 0.001; vs secondary progressive 0.0004 [IQR 0.01], p < 0.001) at baseline were associated with remyelination. Increment of myelin correlated with clinical improvement measured by EDSS (p = 0.015, b = −6.686 × 10−4). χ-separation, a novel mathematical model applied to multiecho T2*-images and T2-images shows that young RRMS patients with low disability exhibit higher remyelination capacity, which correlated with clinical disability over a 2-year follow-up.
- Open Access
- 2 February 2023
- Published by Elsevier in Multiple Sclerosis and Related Disorders
- Vol. 71, 104545
Longitudinal assessment of cervical spinal cord compartments in multiple sclerosis
- 2 February 2023
- Published by Elsevier in Multiple Sclerosis and Related Disorders
- Vol. 71, 104545
Open AccessLongitudinal assessment of cervical spinal cord compartments in multiple sclerosis
- Open Access
- 26 January 2023
- Published by American Society of Neuroradiology (ASNR) in American Journal of Neuroradiology
- Vol. 44 (2) , 218-227
Fully Automatic Method for Reliable Spinal Cord Compartment Segmentation in Multiple Sclerosis
BACKGROUND AND PURPOSE: Fully automatic quantification methods of spinal cord compartments are needed to study pathologic changes of the spinal cord GM and WM in MS in vivo. We propose a novel method for automatic spinal cord compartment segmentation (SCORE) in patients with MS. MATERIALS AND METHODS: The cervical spinal cords of 24 patients with MS and 24 sex- and age-matched healthy controls were scanned on a 3T MR imaging system, including an averaged magnetization inversion recovery acquisition sequence. Three experienced raters manually segmented the spinal cord GM and WM, anterior and posterior horns, gray commissure, and MS lesions. Subsequently, manual segmentations were used to train neural segmentation networks of spinal cord compartments with multidimensional gated recurrent units in a 3-fold cross-validation fashion. Total intracranial volumes were quantified using FreeSurfer. RESULTS: The intra- and intersession reproducibility of SCORE was high in all spinal cord compartments (eg, mean relative SD of GM and WM: ≤ 3.50% and ≤1.47%, respectively) and was better than manual segmentations (all P < .001). The accuracy of SCORE compared with manual segmentations was excellent, both in healthy controls and in patients with MS (Dice similarity coefficients of GM and WM: ≥ 0.84 and ≥0.92, respectively). Patients with MS had lower total WM areas (P < .05), and total anterior horn areas (P < .01 respectively), as measured with SCORE. CONCLUSIONS: We demonstrate a novel, reliable quantification method for spinal cord tissue segmentation in healthy controls and patients with MS and other neurologic disorders affecting the spinal cord. Patients with MS have reduced areas in specific spinal cord tissue compartments, which may be used as MS biomarkers.
- 26 January 2023
- Published by American Society of Neuroradiology (ASNR) in American Journal of Neuroradiology
- Vol. 44 (2) , 218-227
Open AccessFully Automatic Method for Reliable Spinal Cord Compartment Segmentation in Multiple Sclerosis
BACKGROUND AND PURPOSE: Fully automatic quantification methods of spinal cord compartments are needed to study pathologic changes of the spinal cord GM and WM in MS in vivo. We propose a novel method for automatic spinal cord compartment segmentation (SCORE) in patients with MS. MATERIALS AND METHODS: The cervical spinal cords of 24 patients with MS and 24 sex- and age-matched healthy controls were scanned on a 3T MR imaging system, including an averaged magnetization inversion recovery acquisition sequence. Three experienced raters manually segmented the spinal cord GM and WM, anterior and posterior horns, gray commissure, and MS lesions. Subsequently, manual segmentations were used to train neural segmentation networks of spinal cord compartments with multidimensional gated recurrent units in a 3-fold cross-validation fashion. Total intracranial volumes were quantified using FreeSurfer. RESULTS: The intra- and intersession reproducibility of SCORE was high in all spinal cord compartments (eg, mean relative SD of GM and WM: ≤ 3.50% and ≤1.47%, respectively) and was better than manual segmentations (all P < .001). The accuracy of SCORE compared with manual segmentations was excellent, both in healthy controls and in patients with MS (Dice similarity coefficients of GM and WM: ≥ 0.84 and ≥0.92, respectively). Patients with MS had lower total WM areas (P < .05), and total anterior horn areas (P < .01 respectively), as measured with SCORE. CONCLUSIONS: We demonstrate a novel, reliable quantification method for spinal cord tissue segmentation in healthy controls and patients with MS and other neurologic disorders affecting the spinal cord. Patients with MS have reduced areas in specific spinal cord tissue compartments, which may be used as MS biomarkers.
- Open Access
- 29 July 2022
- Published by Elsevier in NeuroImage: Clinical
- Vol. 35, 103137
Longitudinal changes of deep gray matter shape in multiple sclerosis
- 29 July 2022
- Published by Elsevier in NeuroImage: Clinical
- Vol. 35, 103137
Open AccessLongitudinal changes of deep gray matter shape in multiple sclerosis
- Open Access
- 3 May 2022
- Published by Wolters Kluwer Health in Investigative Radiology
- Vol. 57 (9) , 592-600
Fluid and White Matter Suppression
Objective: Cortical lesions are common in multiple sclerosis (MS), but their visualization is challenging on conventional magnetic resonance imaging. The uniform image derived from magnetization prepared 2 rapid acquisition gradient echoes (MP2RAGEuni) detects cortical lesions with a similar rate as the criterion standard sequence, double inversion recovery. Fluid and white matter suppression (FLAWS) provides multiple reconstructed contrasts acquired during a single acquisition. These contrasts include FLAWS minimum image (FLAWSmin), which provides an exquisite sensitivity to the gray matter signal and therefore may facilitate cortical lesion identification, as well as high contrast FLAWS (FLAWShco), which gives a contrast that is similar to one of MP2RAGEuni. In this study, we compared the manual detection rate of cortical lesions on MP2RAGEuni, FLAWSmin, and FLAWShco in MS patients. Furthermore, we assessed whether the combined detection rate on FLAWSmin and FLAWShco was superior to MP2RAGEuni for cortical lesions identification. Last, we compared quantitative T1 maps (qT1) provided by both MP2RAGE and FLAWS in MS lesions. Materials and Methods: We included 30 relapsing-remitting MS patients who underwent MP2RAGE and FLAWS magnetic resonance imaging with isotropic spatial resolution of 1 mm at 3 T. Cortical lesions were manually segmented by consensus of 3 trained raters and classified as intracortical or leukocortical lesions on (1) MP2RAGE uniform/flat images, (2) FLAWSmin, and (3) FLAWShco. In addition, segmented lesions on FLAWSmin and FLAWShco were merged to produce a union lesion map (FLAWSmin + hco). Number and volume of all cortical, intracortical, and leukocortical lesions were compared among MP2RAGEuni, FLAWSmin, and FLAWShco using Friedman test and between MP2RAGEuni and FLAWSmin + hco using Wilcoxon signed rank test. The FLAWS T1 maps were then compared with the reference MP2RAGE T1 maps using relative differences in percentage. In an exploratory analysis, individual cortical lesion counts of the 3 raters were compared, and interrater variability was quantified using Fleiss ϰ. Results: In total, 633 segmentations were made on the 3 contrasts, corresponding to 355 cortical lesions. The median number and volume of single cortical, intracortical, and leukocortical lesions were comparable among MP2RAGEuni, FLAWSmin, and FLAWShco. In patients with cortical lesions (22/30), median cumulative lesion volume was larger on FLAWSmin (587 μL; IQR, 1405 μL) than on MP2RAGEuni (490 μL; IQR, 990 μL; P = 0.04), whereas there was no difference between FLAWSmin and FLAWShco, or FLAWShco and MP2RAGEuni. FLAWSmin + hco showed significantly greater numbers of cortical (median, 4.5; IQR, 15) and leukocortical (median, 3.5; IQR, 12) lesions than MP2RAGEuni (median, 3; IQR, 10; median, 2.5; IQR, 7; both P < 0.001). Interrater agreement was moderate on MP2RAGEuni (ϰ = 0.582) and FLAWShco (ϰ = 0.584), but substantial on FLAWSmin (ϰ = 0.614). qT1 in lesions was similar between MP2RAGE and FLAWS. Conclusions: Cortical lesions identification in FLAWSmin and FLAWShco was comparable to MP2RAGEuni. The combination of FLAWSmin and FLAWShco allowed to identify a higher number of cortical lesions than MP2RAGEuni, whereas qT1 maps did not differ between the 2 acquisition schemes.
- 3 May 2022
- Published by Wolters Kluwer Health in Investigative Radiology
- Vol. 57 (9) , 592-600
Open AccessFluid and White Matter Suppression
Objective: Cortical lesions are common in multiple sclerosis (MS), but their visualization is challenging on conventional magnetic resonance imaging. The uniform image derived from magnetization prepared 2 rapid acquisition gradient echoes (MP2RAGEuni) detects cortical lesions with a similar rate as the criterion standard sequence, double inversion recovery. Fluid and white matter suppression (FLAWS) provides multiple reconstructed contrasts acquired during a single acquisition. These contrasts include FLAWS minimum image (FLAWSmin), which provides an exquisite sensitivity to the gray matter signal and therefore may facilitate cortical lesion identification, as well as high contrast FLAWS (FLAWShco), which gives a contrast that is similar to one of MP2RAGEuni. In this study, we compared the manual detection rate of cortical lesions on MP2RAGEuni, FLAWSmin, and FLAWShco in MS patients. Furthermore, we assessed whether the combined detection rate on FLAWSmin and FLAWShco was superior to MP2RAGEuni for cortical lesions identification. Last, we compared quantitative T1 maps (qT1) provided by both MP2RAGE and FLAWS in MS lesions. Materials and Methods: We included 30 relapsing-remitting MS patients who underwent MP2RAGE and FLAWS magnetic resonance imaging with isotropic spatial resolution of 1 mm at 3 T. Cortical lesions were manually segmented by consensus of 3 trained raters and classified as intracortical or leukocortical lesions on (1) MP2RAGE uniform/flat images, (2) FLAWSmin, and (3) FLAWShco. In addition, segmented lesions on FLAWSmin and FLAWShco were merged to produce a union lesion map (FLAWSmin + hco). Number and volume of all cortical, intracortical, and leukocortical lesions were compared among MP2RAGEuni, FLAWSmin, and FLAWShco using Friedman test and between MP2RAGEuni and FLAWSmin + hco using Wilcoxon signed rank test. The FLAWS T1 maps were then compared with the reference MP2RAGE T1 maps using relative differences in percentage. In an exploratory analysis, individual cortical lesion counts of the 3 raters were compared, and interrater variability was quantified using Fleiss ϰ. Results: In total, 633 segmentations were made on the 3 contrasts, corresponding to 355 cortical lesions. The median number and volume of single cortical, intracortical, and leukocortical lesions were comparable among MP2RAGEuni, FLAWSmin, and FLAWShco. In patients with cortical lesions (22/30), median cumulative lesion volume was larger on FLAWSmin (587 μL; IQR, 1405 μL) than on MP2RAGEuni (490 μL; IQR, 990 μL; P = 0.04), whereas there was no difference between FLAWSmin and FLAWShco, or FLAWShco and MP2RAGEuni. FLAWSmin + hco showed significantly greater numbers of cortical (median, 4.5; IQR, 15) and leukocortical (median, 3.5; IQR, 12) lesions than MP2RAGEuni (median, 3; IQR, 10; median, 2.5; IQR, 7; both P < 0.001). Interrater agreement was moderate on MP2RAGEuni (ϰ = 0.582) and FLAWShco (ϰ = 0.584), but substantial on FLAWSmin (ϰ = 0.614). qT1 in lesions was similar between MP2RAGE and FLAWS. Conclusions: Cortical lesions identification in FLAWSmin and FLAWShco was comparable to MP2RAGEuni. The combination of FLAWSmin and FLAWShco allowed to identify a higher number of cortical lesions than MP2RAGEuni, whereas qT1 maps did not differ between the 2 acquisition schemes.