Recent advances in the longitudinal segmentation of multiple sclerosis lesions on magnetic resonance imaging: a review

Diaz-Hurtado, Marcos; Martínez‐Heras, Eloy; Solana, Elisabeth; Casas-Roma, Jordi; Llufriú, Sara; Kanber, Baris; Carrasco, Ferran Prados

doi:10.1007/s00234-022-03019-3

Cited by 19 publications

(22 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Indeed, lack of new lesions in the CNS is a key indicator of medication effectiveness 5 . Monitoring these lesions, however, is often monotonous and repetitive for neuroradiologists 6 , and this combined with radiology's supply-demand challenges 7 has led to increased interest in methods for automating lesion detection 8 . Over the past two decades, research has heavily focused on computer-assisted segmentation methods 8 , with a recent surge in AI methodologies 9 .…”

Section: Introductionmentioning

confidence: 99%

“…Monitoring these lesions, however, is often monotonous and repetitive for neuroradiologists 6 , and this combined with radiology's supply-demand challenges 7 has led to increased interest in methods for automating lesion detection 8 . Over the past two decades, research has heavily focused on computer-assisted segmentation methods 8 , with a recent surge in AI methodologies 9 . Current research trends are evolving from simple identification of MS lesions on T2/FLAIR to analysing images over different times 8 .…”

Section: Introductionmentioning

confidence: 99%

“…Over the past two decades, research has heavily focused on computer-assisted segmentation methods 8 , with a recent surge in AI methodologies 9 . Current research trends are evolving from simple identification of MS lesions on T2/FLAIR to analysing images over different times 8 . The MSSEG2 challenge by The Medical Image Computing and Computer Assisted Intervention Society (MICCAI), targeting new lesion detection, has significantly boosted interest in this research domain 3 .…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Siamese U-Transformer for change detection on MRI brain for multiple sclerosis, a model development and external validation study

Kelly,

Mathur,

Killeen

et al. 2024

Preprint

View full text Add to dashboard Cite

Background Multiple Sclerosis (MS), is a chronic idiopathic demyelinating disorder of the CNS. Imaging plays a central role in diagnosis and monitoring. Monitoring for progression however, can be repetitive for neuroradiologists, and this has led to interest in automated lesion detection. Simultaneously, in the computer science field of Remote Sensing, Change Detection (CD), the identification of change between co-registered images at different times, has been disrupted by the emergence of Vision Transformers. CD offers an alternative to semantic segmentation leveraging the temporal information in the data. Methods In this retrospective study with external validation we reframe the clinical radiology task of new lesion identification as a CD problem. Consecutive patients who had MRI studies for MS at our institution between 2019 and 2022 were reviewed and those with new lesion(s) were included. External data was obtained from the MSSEG2 challenge and OpenMS. Multiple CD models, and a novel model (NeUFormer), were trained and tested. Results were analysed on both paired slices and at the patient level. Expected Cost (EC) and F2 were independently and prospectively chosen as our primary evaluation metrics. For external data we report DICE and F1 to allow for comparison with existing data. For each test set 1000 bootstrapping simulations were performed by sampling 10 patient samples with replacement giving a non parametric estimate of the confidence interval. Wilcoxon statistics were calculated to test for significance. Findings 43,440 MR images were included for analysis (21,720 pairs). The internal set comprised of 170 patients (110 for training, 30 for tuning, 30 testing) with 120 females and 50 males, average age of 42 (range 21,74). 60 (40 + 20) patients were included for external validation. In the CD experiments (2D) our proposed NeuFormer model achieved the best (lowest) Expected Cost (EC) (p=0.0095), the best F2 and second best DICE (p<0.0001). At the patient level our NeUFormer model had the joint highest number of True Positive lesions, and lowest number of False negatives (p<0.002). For CD on external data, NeUFormer achieved the highest DICE on both datasets (p<0.0001). NeUFormer had the lowest or joint lowest number of False Positives on external data (p<0.0001 in all cases). Interpretation Reformulating new lesion identification as a CD problem allows the use of new techniques and methods of evaluation. We introduce a novel Siamese U-Transformer, NeUFormer, which combines concepts from U-Net, Siamese Networks, and vision transformers to create a model with improved small lesion detection and the consistently best EC. Its ability to increase detection of small lesions, balanced with relatively few false positives, and superior generalisability has the potential to greatly impact the field of the identification of radiologic progression of MS with AI.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Siamese U-Transformer for change detection on MRI brain for multiple sclerosis, a model development and external validation study

Kelly,

Mathur,

Killeen

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…3 A number of automated and semi-automated algorithms have been developed to overcome these challenges. [4][5][6] In the classical approach, serially acquired images are intensitynormalized and co-registered, from which a dissimilarity map (e.g., obtained by subtraction) is calculated and then automatically segmented (e.g., by thresholding or statistical inference methods) or reviewed by humans to yield the final lesion change mask. 2,[7][8][9][10][11][12][13][14] More recently, supervised deep learning-based convolutional neural network models have become the predominant approach.…”

Section: Introductionmentioning

confidence: 99%

“…[15][16][17][18][19] Despite rapid advances in research, the detection sensitivity and specificity remain moderate on a voxel-or lesion-level (less than 0.8). 3,6 We previously introduced the statistical detection of change algorithm (SDC) algorithm as an automated lesion change detection tool to visually assist human readers. This algorithm applies an optimal binary change detector to the subtraction of two longitudinally registered FLAIR images to delineate brain areas with potential new lesions.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of statistical detection of change algorithm for triaging multiple sclerosis patients with new lesion activity on longitudinal brain MRI

Homssi¹,

Sweeney²,

Demmon³

et al. 2023

Preprint

View full text Add to dashboard Cite

Background and Purpose: Identification of new MS lesions on longitudinal MRI by human readers is time-consuming and prone to error. Our objective was to evaluate the improvement in a subject-level detection performance by readers when assisted by the automated statistical detection of change (SDC) algorithm. Materials and Methods: A total of 200 MS patients with mean inter-scan interval of 13.2 ± 2.4 months were included. SDC was applied to the baseline and follow-up FLAIR images to detect potential new lesions for confirmation by readers (Reader+SDC method). This method was compared with readers operating in the clinical workflow (Reader method) for a subject-level detection of new lesions. Results: Reader+SDC found 30 subjects (15.0%) with at least one new lesion, while Reader detected 16 subjects (8.0%). As a subject-level triage tool, SDC achieved a perfect sensitivity of 1.00 (95% CI: [0.88, 1.00]) and a moderate specificity of 0.67 (95% CI: [0.59, 0.74]). The agreement on a subject-level was 0.91 (95% CI: [0.87, 0.95]) between Reader+SDC and Reader, and 0.72 (95% CI: [0.66, 0.78]) between Reader+SDC and SDC. Conclusion: SDC improves the detection accuracy of human readers and can serve as a time-saving patient triage tool for detecting new MS lesion activity on longitudinal FLAIR images.

show abstract

Predicting multiple sclerosis disease progression and outcomes with machine learning and MRI-based biomarkers: a review

Yousef,

Malagurski Tortei,

Castiglione

2024

J Neurol

View full text Add to dashboard Cite

Recent advances in the longitudinal segmentation of multiple sclerosis lesions on magnetic resonance imaging: a review

Abstract: This study has been approved by the Ethics Committee of the University Oberta de Catalunya (UOC) stating that this research does non include human subjects participation or any processing of personal data and the research fulfils current legislation on data protection.

Cited by 19 publications

References 63 publications

A Siamese U-Transformer for change detection on MRI brain for multiple sclerosis, a model development and external validation study

A Siamese U-Transformer for change detection on MRI brain for multiple sclerosis, a model development and external validation study

Evaluation of statistical detection of change algorithm for triaging multiple sclerosis patients with new lesion activity on longitudinal brain MRI

Predicting multiple sclerosis disease progression and outcomes with machine learning and MRI-based biomarkers: a review

Contact Info

Product

Resources

About