Diana Maria Sima

Background: Cognitive deterioration is common in multiple sclerosis (MS) and requires regular follow-up. Currently, cognitive status is measured in clinical practice using paper-and-pencil tests, which are both time-consuming and costly. Remote monitoring of cognitive status could offer a solution because previous studies on telemedicine tools have proved its feasibility and acceptance among people with MS. However, existing smartphone-based apps include designs that are prone to motor interference and focus primarily on information processing speed, although memory is also commonly affected. Objective: This study aims to validate a smartphone-based cognitive screening battery, icognition, to detect deterioration in both memory and information processing speed. Methods: The icognition screening battery consists of 3 tests: the Symbol Test for information processing speed, the Dot Test for visuospatial short-term memory and learning, and the visual Backward Digit Span (vBDS) for working memory. These tests are based on validated paper-and-pencil tests: the Symbol Digit Modalities Test, the 10/36 Spatial Recall Test, and the auditory Backward Digit Span, respectively. To establish the validity of icognition, 101 people with MS and 82 healthy participants completed all tests. Of the 82 healthy participants, 20 (24%) repeated testing 2 to 3 weeks later. For each icognition test, validity was established by the correlation with its paper-and-pencil equivalent (concurrent validity), the correlation and intraclass correlation coefficient (ICC) between baseline and follow-up testing (test-retest reliability), the difference between people with MS and healthy participants, and the correlation with other clinical parameters such as the Expanded Disability Status Scale. Results: All icognition tests correlated well with their paper-and-pencil equivalents (Symbol Test: r=0.67; P<.001; Dot Test: r=0.31; P=.002; vBDS: r=0.69; P<.001), negatively correlated with the Expanded Disability Status Scale (Symbol Test: ρ=–0.34; P<.001; Dot Test: ρ=−0.32; P=.003; vBDS: ρ=−0.21; P=.04), and showed moderate test-retest reliability (Symbol Test: ICC=0.74; r=0.85; P<.001; Dot Test: ICC=0.71; r=0.74; P<.001; vBDS: ICC=0.72; r=0.83; P<.001). Test performance was comparable between people with MS and healthy participants for all cognitive tests, both in icognition (Symbol Test: U=4431; P=.42; Dot Test: U=3516; P=.32; vBDS: U=3708; P=.27) and the gold standard paper-and-pencil tests (Symbol Digit Modalities Test: U=4060.5, P=.82; 10/36 Spatial Recall Test: U=3934; P=.74; auditory Backward Digit Span: U=3824.5, P=.37). Conclusions: icognition is a valid tool to remotely screen cognitive performance in people with MS. It is planned to be included in a digital health platform that includes volumetric brain analysis and patient-reported outcome measures. Future research should establish the usability and psychometric properties of icognition in a remote setting.

Importance: Amyloid-related imaging abnormalities (ARIA) are brain magnetic resonance imaging (MRI) findings associated with the use of amyloid-β–directed monoclonal antibody therapies in Alzheimer disease (AD). ARIA monitoring is important to inform treatment dosing decisions and might be improved through assistive software. Objective: To assess the clinical performance of an artificial intelligence (AI)–based software tool for assisting radiological interpretation of brain MRI scans in patients monitored for ARIA. Design, Setting, and Participants: This diagnostic study used a multiple-reader multiple-case design to evaluate the diagnostic performance of radiologists assisted by the software vs unassisted. The study enrolled 16 US Board of Radiology–certified radiologists to perform radiological reading with (assisted) and without the software (unassisted). The study encompassed 199 retrospective cases, where each case consisted of a predosing baseline and a postdosing follow-up MRI of patients from aducanumab clinical trials PRIME, EMERGE, and ENGAGE. Statistical analysis was performed from April to July 2023. Exposures: Use of icobrain aria, an AI-based assistive software for ARIA detection and quantification. Main Outcomes and Measures: Coprimary end points were the difference in diagnostic accuracy between assisted and unassisted detection of ARIA-E (edema and/or sulcal effusion) and ARIA-H (microhemorrhage and/or superficial siderosis) independently, assessed with the area under the receiver operating characteristic curve (AUC). Results: Among the 199 participants included in this study of radiological reading performance, mean (SD) age was 70.4 (7.2) years; 105 (52.8%) were female; 23 (11.6%) were Asian, 1 (0.5%) was Black, 157 (78.9%) were White, and 18 (9.0%) were other or unreported race and ethnicity. Among the 16 radiological readers included, 2 were specialized neuroradiologists (12.5%), 11 were male individuals (68.8%), 7 were individuals working in academic hospitals (43.8%), and they had a mean (SD) of 9.5 (5.1) years of experience. Radiologists assisted by the software were significantly superior in detecting ARIA than unassisted radiologists, with a mean assisted AUC of 0.87 (95% CI, 0.84-0.91) for ARIA-E detection (AUC improvement of 0.05 [95% CI, 0.02-0.08]; P = .001]) and 0.83 (95% CI, 0.78-0.87) for ARIA-H detection (AUC improvement of 0.04 [95% CI, 0.02-0.07]; P = .001). Sensitivity was significantly higher in assisted reading compared with unassisted reading (87% vs 71% for ARIA-E detection; 79% vs 69% for ARIA-H detection), while specificity remained above 80% for the detection of both ARIA types. Conclusions and Relevance: This diagnostic study found that radiological reading performance for ARIA detection and diagnosis was significantly better when using the AI-based assistive software. Hence, the software has the potential to be a clinically important tool to improve safety monitoring and management of patients with AD treated with amyloid-β–directed monoclonal antibody therapies.

Purpose: To assess the performance of the inferior lateral ventricle (ILV) to hippocampal (Hip) volume ratio on brain MRI, for Alzheimer’s disease (AD) diagnostics, comparing it to individual automated ILV and hippocampal volumes, and visual medial temporal lobe atrophy (MTA) consensus ratings. Methods: One-hundred-twelve subjects (mean age ± SD, 66.85 ± 13.64 years) with varying degrees of cognitive decline underwent MRI using a Philips Ingenia 3T. The MTA scale by Scheltens, rated on coronal 3D T1-weighted images, was determined by three experienced radiologists, blinded to diagnosis and sex. Automated volumetry was computed by icobrain dm (v. 5.10) for total, left, right hippocampal, and ILV volumes. The ILV/Hip ratio, defined as the percentage ratio between ILV and hippocampal volumes, was calculated and compared against a normative reference population (n = 1903). Inter-rater agreement, association, classification accuracy, and clinical interpretability on patient level were reported. Results: Visual MTA scores showed excellent inter-rater agreement. Ordinal logistic regression and correlation analyses demonstrated robust associations between automated brain segmentations and visual MTA ratings, with the ILV/Hip ratio consistently outperforming individual hippocampal and ILV volumes. Pairwise classification accuracy showed good performance without statistically significant differences between the ILV/Hip ratio and visual MTA across disease stages, indicating potential interchangeability. Comparison to the normative population and clinical interpretability assessments showed commensurability in classifying MTA “severity” between visual MTA and ILV/Hip ratio measurements. Conclusion: The ILV/Hip ratio shows the highest correlation to visual MTA, in comparison to automated individual ILV and hippocampal volumes, offering standardized measures for diagnostic support in different stages of cognitive decline.

BACKGROUND Cognitive deterioration is common in multiple sclerosis (MS) and requires regular follow-up. Currently, cognitive status is measured in clinical practice using paper-and-pencil tests, which are both time-consuming and costly. Remote monitoring of cognitive status could offer a solution because previous studies on telemedicine tools have proved its feasibility and acceptance among people with MS. However, existing smartphone-based apps include designs that are prone to motor interference and focus primarily on information processing speed, although memory is also commonly affected. OBJECTIVE This study aims to validate a smartphone-based cognitive screening battery, icognition, to detect deterioration in both memory and information processing speed. METHODS The icognition screening battery consists of 3 tests: the Symbol Test for information processing speed, the Dot Test for visuospatial short-term memory and learning, and the visual Backward Digit Span (vBDS) for working memory. These tests are based on validated paper-and-pencil tests: the Symbol Digit Modalities Test, the 10/36 Spatial Recall Test, and the auditory Backward Digit Span, respectively. To establish the validity of icognition, 101 people with MS and 82 healthy participants completed all tests. Of the 82 healthy participants, 20 (24%) repeated testing 2 to 3 weeks later. For each icognition test, validity was established by the correlation with its paper-and-pencil equivalent (concurrent validity), the correlation and intraclass correlation coefficient (ICC) between baseline and follow-up testing (test-retest reliability), the difference between people with MS and healthy participants, and the correlation with other clinical parameters such as the Expanded Disability Status Scale. RESULTS All icognition tests correlated well with their paper-and-pencil equivalents (Symbol Test: r=0.67; P<.001; Dot Test: r=0.31; P=.002; vBDS: r=0.69; P<.001), negatively correlated with the Expanded Disability Status Scale (Symbol Test: ρ=–0.34; P<.001; Dot Test: ρ=−0.32; P=.003; vBDS: ρ=−0.21; P=.04), and showed moderate test-retest reliability (Symbol Test: ICC=0.74; r=0.85; P<.001; Dot Test: ICC=0.71; r=0.74; P<.001; vBDS: ICC=0.72; r=0.83; P<.001). Test performance was comparable between people with MS and healthy participants for all cognitive tests, both in icognition (Symbol Test: U=4431; P=.42; Dot Test: U=3516; P=.32; vBDS: U=3708; P=.27) and the gold standard paper-and-pencil tests (Symbol Digit Modalities Test: U=4060.5, P=.82; 10/36 Spatial Recall Test: U=3934; P=.74; auditory Backward Digit Span: U=3824.5, P=.37). CONCLUSIONS icognition is a valid tool to remotely screen cognitive performance in people with MS. It is planned to be included in a digital health platform that includes volumetric brain analysis and patient-reported outcome measures. Future research should establish the usability and psychometric properties of icognition in a remote setting.

Background: Telemedicine is feasible and well-accepted by people with multiple sclerosis (MS). Objective: The aim of this study is to validate a smartphone-based cognitive screening battery, ico gnition , to faster signal cognitive deterioration. Methods: ico gnition consists of three tests (Symbol Test, Dot Test and visual Backwards Digit Span (vBDS)) that are equivalents of validated paper-pencil tests. These are the Symbol Digit Modalities Test (SDMT), the 10/36 Spatial Recall Test (SPART) and the auditory Backwards Digit Span (aBDS), respectively. To establish the validity of ico gnition , 101 people with MS and 82 healthy subjects completed all tests. 21 healthy subjects repeated testing 2 to 3 weeks later. Results: All tests in ico gnition correlate well with their paper-pencil equivalent (Symbol Test: r=.63, p<.001; Dot Test: r=.31, p=0.002; vBDS: r=.71, p<.001), negatively correlate with the Expanded Disability Status Scale (EDSS: Symbol Test: rho=-.27, p=.01; Dot Test: rho=-.29, p=.006; vBDS: rho=- .23, p=.027) and show high test-retest reliability (Symbol Test: r=.81, p<.001; Dot Test: r=.75, p<.001; vBDS: r=.84, p<.001). Test performance was not significantly different between people with MS and healthy subjects for all cognitive tests, both in ico gnition and their paper-pencil equivalents. Conclusion: ico gnition is a valid and reliable tool to remotely screen for cognitive functioning in persons with MS.

Introduction Classical deep learning research requires lots of centralised data. However, data sets are often stored at different clinical centers, and sharing sensitive patient data such as brain images is difficult. In this manuscript, we investigated the feasibility of federated learning, sending models to the data instead of the other way round, for research on brain magnetic resonant images of people with multiple sclerosis (MS). Methods Using transfer learning on a previously published brain age model, we trained a model to decode performance on the symbol digit modalities test (SDMT) of patients with MS from structural T1 weighted MRI. Three international centers in Brussels, Greifswald and Prague participated in the project. In Brussels, one computer served as the server coordinating the FL project, while the other served as client for model training on local data (n=97). The other two clients were Greifswald (n=104) and Prague (n=100). Each FL round, the server sent a global model to the clients, where its fully connected layer was updated on the local data. After collecting the local models, the server applied a weighted average of two randomly picked clients, yielding a new global model. Results After 22 federated learning rounds, the average validation loss across clients reached a minimum. The model appeared to have learned to assign SDMT values close to the mean with a mean absolute error of 9.04, 10.59 and 10.71 points between true and predicted SDMT on the test data sets of Brussels, Greifswald and Prague respectively. The overall test MAE across all clients was 10.13 points. Conclusion Federated learning is feasible for machine learning research on brain MRI of persons with MS, setting the stage for larger transfer learning studies to investigate the utility of brain age latent representations in cognitive decoding tasks.

Convolutional Neural Networks (CNNs) with U-shaped architectures have dominated medical image segmentation, which is crucial for various clinical purposes. However, the inherent locality of convolution makes CNNs fail to fully exploit global context, essential for better recognition of some structures, e.g., brain lesions. Transformers have recently proven promising performance on vision tasks, including semantic segmentation, mainly due to their capability of modeling long-range dependencies. Nevertheless, the quadratic complexity of attention makes existing Transformer-based models use self-attention layers only after somehow reducing the image resolution, which limits the ability to capture global contexts present at higher resolutions. Therefore, this work introduces a family of models, dubbed Factorizer, which leverages the power of low-rank matrix factorization for constructing an end-to-end segmentation model. Specifically, we propose a linearly scalable approach to context modeling, formulating Nonnegative Matrix Factorization (NMF) as a differentiable layer integrated into a U-shaped architecture. The shifted window technique is also utilized in combination with NMF to effectively aggregate local information. Factorizers compete favorably with CNNs and Transformers in terms of accuracy, scalability, and interpretability, achieving state-of-the-art results on the BraTS dataset for brain tumor segmentation and ISLES'22 dataset for stroke lesion segmentation. Highly meaningful NMF components give an additional interpretability advantage to Factorizers over CNNs and Transformers. Moreover, our ablation studies reveal a distinctive feature of Factorizers that enables a significant speed-up in inference for a trained Factorizer without any extra steps and without sacrificing much accuracy. The code and models are publicly available at https://github.com/pashtari/factorizer.

Background and purpose: Data from neuro‐imaging techniques allow us to estimate a brain's age. Brain age is easily interpretable as ‘how old the brain looks’ and could therefore be an attractive communication tool for brain health in clinical practice. This study aimed to investigate its clinical utility by investigating the relationship between brain age and cognitive performance in multiple sclerosis (MS).Methods: A linear regression model was trained to predict age from brain magnetic resonance imaging volumetric features and sex in a healthy control dataset (HC_train, n = 1673). This model was used to predict brain age in two test sets: HC_test (n = 50) and MS_test (n = 201). Brain‐predicted age difference (BPAD) was calculated as BPAD = brain age minus chronological age. Cognitive performance was assessed by the Symbol Digit Modalities Test (SDMT).Results: Brain age was significantly related to SDMT scores in the MS_test dataset (r = −0.46, p < 0.001) and contributed uniquely to variance in SDMT beyond chronological age, reflected by a significant correlation between BPAD and SDMT (r = −0.24, p < 0.001) and a significant weight (−0.25, p = 0.002) in a multivariate regression equation with age.Conclusions: Brain age is a candidate biomarker for cognitive dysfunction in MS and an easy to grasp metric for brain health.