Optimising brain age estimation through transfer learning: A suite of pre-trained foundation models for improved performance and generalisability in a clinical setting.

Estimated age from brain MRI data has emerged as a promising biomarker of neurological health. However, the absence of large, diverse, and clinically representative training datasets, along with the complexity of managing heterogeneous MRI data, presents significant barriers to the development of accurate and generalisable models appropriate for clinical use. Here, we present a deep learning framework trained on routine clinical data (N up to 18,890, age range 18-96 years). We trained five separate models for accurate brain age prediction (all with mean absolute error ≤4.0 years, R2 ≥ .86) across five different MRI sequences (T2 -weighted, T2 -FLAIR, T1 -weighted, diffusion-weighted, and gradient-recalled echo T2 *-weighted). Our trained models offer dual functionality. First, they have the potential to be directly employed on clinical data. Second, they can be used as foundation models for further refinement to accommodate a range of other MRI sequences (and therefore a range of clinical scenarios which employ such sequences). This adaptation process, enabled by transfer learning, proved effective in our study across a range of MRI sequences and scan orientations, including those which differed considerably from the original training datasets. Crucially, our findings suggest that this approach remains viable even with limited data availability (as low as N = 25 for fine-tuning), thus broadening the application of brain age estimation to more diverse clinical contexts and patient populations. By making these models publicly available, we aim to provide the scientific community with a versatile toolkit, promoting further research in brain age prediction and related areas.

Functional MRI but not white matter fibre dissection identifies language dominance.

OBJECTIVES: Lateralisation of some language pathways has been reported in the literature using diffusion tractography, which is more feasible than functional magnetic resonance imaging (fMRI) in challenging patients. Our retrospective study investigates whether a correlation exists between threshold-independent fMRI language lateralisation and structural lateralisation using tractography in healthy controls and brain tumour patients. METHODS: Fifteen healthy subjects and 61 patients underwent language fMRI and diffusion-weighted MRI. A regional fMRI laterality index (LI) was calculated. Tracts dissected were the arcuate fasciculus (long direct and short indirect tracts), uncinate fasciculus, inferior longitudinal fasciculus, inferior fronto-occipital fasciculus and frontal aslant tract. An asymmetry index (AI) for each tract was calculated using tract volume analysed with single tensor (ST) and spherical deconvolution (SD) models, as well as hindrance modulated orientational anisotropy (HMOA) for SD tracts. Linear regression assessed the correlation between LI and AI. RESULTS: In all subjects, there was no significant correlation between LI and AI for any of the dissected tracts. Significant correlations were only found when handedness for controls and tumour volume for patients were included as covariates. In handedness subgroups, the average AI of some tracts showed the same laterality as LI, and some the opposite. Discordant results were observed for ST- and SD-based AIs. CONCLUSIONS: Our results do not support using tractography in the assessment of language lateralisation. The discordant results between ST and SD indicate that either the structural lateralisation of dissected tracts is less robust than functional lateralisation, or tractography is not sensitive methodology. Other diffusion analysis approaches should be developed. CLINICAL RELEVANCE STATEMENT: Although diffusion tractography may be more feasible than fMRI in challenging tumour patients and where sedation or anaesthesia is required, our results do not currently recommend replacing fMRI with tractography using volume or HMOA in the assessment of language lateralisation. KEY POINTS: • No correlation found between fMRI and tractography in language lateralisation. • Discordance between asymmetry indices of different tractography models and metrics. • Tractography not currently recommended in language lateralisation assessment.

Early MRI Predictors of Relapse in Primary Central Nervous System Lymphoma Treated with MATRix Immunochemotherapy.

Primary Central Nervous System Lymphoma (PCNSL) is a highly malignant brain tumour. We investigated dynamic changes in tumour volume and apparent diffusion coefficient (ADC) measurements for predicting outcome following treatment with MATRix chemotherapy in PCNSL. Patients treated with MATRix (n = 38) underwent T1 contrast-enhanced (T1CE) and diffusion-weighted imaging (DWI) before treatment, after two cycles and after four cycles of chemotherapy. Response was assessed using the International PCNSL Collaborative Group (IPCG) imaging criteria. ADC histogram parameters and T1CE tumour volumes were compared among response groups, using one-way ANOVA testing. Logistic regression was performed to examine those imaging parameters predictive of response. Response after two cycles of chemotherapy differed from response after four cycles; of the six patients with progressive disease (PD) after four cycles of treatment, two (33%) had demonstrated a partial response (PR) or complete response (CR) after two cycles. ADCmean at baseline, T1CE at baseline and T1CE percentage volume change differed between response groups (0.005 < p < 0.038) and were predictive of MATRix treatment response (area under the curve: 0.672-0.854). Baseline ADC and T1CE metrics are potential biomarkers for risk stratification of PCNSL patients early during remission induction therapy with MATRix. Standard interim response assessment (after two cycles) according to IPCG imaging criteria does not reliably predict early disease progression in the context of a conventional treatment approach.

Diverse imaging features of adolescent glioblastoma.

We highlight an unusual case of multifocal glioblastoma in an adolescent patient, manifesting as four discrete brain lesions, each distinct in appearance. Familiarity with the diverse imaging features of glioblastoma can reduce misdiagnosis and avoid treatment delays.

Imaging Biomarkers of Glioblastoma Treatment Response: A Systematic Review and Meta-Analysis of Recent Machine Learning Studies.

OBJECTIVE: Monitoring biomarkers using machine learning (ML) may determine glioblastoma treatment response. We systematically reviewed quality and performance accuracy of recently published studies. METHODS: Following Preferred Reporting Items for Systematic Reviews and Meta-Analysis: Diagnostic Test Accuracy, we extracted articles from MEDLINE, EMBASE and Cochrane Register between 09/2018-01/2021. Included study participants were adults with glioblastoma having undergone standard treatment (maximal resection, radiotherapy with concomitant and adjuvant temozolomide), and follow-up imaging to determine treatment response status (specifically, distinguishing progression/recurrence from progression/recurrence mimics, the target condition). Using Quality Assessment of Diagnostic Accuracy Studies Two/Checklist for Artificial Intelligence in Medical Imaging, we assessed bias risk and applicability concerns. We determined test set performance accuracy (sensitivity, specificity, precision, F1-score, balanced accuracy). We used a bivariate random-effect model to determine pooled sensitivity, specificity, area-under the receiver operator characteristic curve (ROC-AUC). Pooled measures of balanced accuracy, positive/negative likelihood ratios (PLR/NLR) and diagnostic odds ratio (DOR) were calculated. PROSPERO registered (CRD42021261965). RESULTS: Eighteen studies were included (1335/384 patients for training/testing respectively). Small patient numbers, high bias risk, applicability concerns (particularly confounding in reference standard and patient selection) and low level of evidence, allow limited conclusions from studies. Ten studies (10/18, 56%) included in meta-analysis gave 0.769 (0.649-0.858) sensitivity [pooled (95% CI)]; 0.648 (0.749-0.532) specificity; 0.706 (0.623-0.779) balanced accuracy; 2.220 (1.560-3.140) PLR; 0.366 (0.213-0.572) NLR; 6.670 (2.800-13.500) DOR; 0.765 ROC-AUC. CONCLUSION: ML models using MRI features to distinguish between progression and mimics appear to demonstrate good diagnostic performance. However, study quality and design require improvement.

MRI morphology of surgically treated lumbar canal stenosis: a retrospective study.

STUDY DESIGN: Retrospective cohort study. OBJECTIVE: The aim of this study was to identify morphologic features on magnetic resonance imaging that might correlate with lumbar canal stenosis severe enough to warrant surgery. SUMMARY OF BACKGROUND DATA: None of the quantitative parameters measured on x-rays, CT scans, or magnetic resonance imaging correlates well with the severity of clinical symptoms in lumbar canal stenosis (LCS). In a patient with neurogenic claudication, we need to define what would constitute radiologic LCS and whether he needs surgical intervention. This paper attempts to define MRI features of LCS addressing the morphology rather than canal dimensions in any direction. MATERIALS AND METHODS: A total of 64 consecutive patients who were operated at 113 levels of LCS were reviewed retrospectively. Their clinical notes and MRI were analyzed. Only the axial T2-weighted images were utilized for this study. The images were reviewed by 1 orthopedic surgeon and 1 radiologist and segregated into morphologic categories. No interobserver and intraobserver studies were undertaken. RESULTS: Two types of axial image features were identified in LCS symmetrical and asymmetrical with 5 subtypes. They were trefoil, triangular, "cat's eye," "pinhole," and complete obliteration. Several subtypes were also described. Of the operated cases, 70.8% had a triangular configuration of the canal with symmetrical large triangular canal shape occurring in 49/80 levels. It was impossible to correlate the severity of symptoms, their duration, and the presence of objective neurological deficits with the morphologic picture from the documentation available. CONCLUSIONS: LCS seems to produce predictable patterns on T2 axial MRI. The triangular configuration correlates most frequently with surgical LCS. Further studies are needed in normal individuals, in prospective patients, and to determine the outcome of treatment based on MRI morphology.