Results
- Showing results for:
- Reset all filters
Search results
-
Journal articleWei X, Liao Y, Zhu J, et al., 2026,
A hybrid CNN–Mamba state space model with pyramid-pooled skip connections for prostate tumor segmentation
, Expert Systems with Applications, Vol: 327, ISSN: 0957-4174Automatic prostate tumor segmentation in multiparametric magnetic resonance imaging (mpMRI) can support targeted biopsy and therapy planning, but remains challenging due to the small, heterogeneous and low-contrast nature of the lesions. CNN-based encoder-decoder models capture local detail well, but struggle with long-range reasoning. Transformers excel at global interactions, but are computationally expensive. Mamba/state-space models enable efficient global modelling, but often lack boundary-sensitive local refinement. We propose ProMamba: a hybrid CNN-Mamba segmentation framework combining local detail extraction with efficient global dependency modelling within a hierarchical U-Net architecture. Its dual-branch Pro-SSM block incorporates a depthwise dilated convolution module (DDCM) to extract multi-scale boundary cues, as well as a Mamba-based visual state-space (VSS) branch to aggregate multi-directional long-range context. The two branches are fused via channel concatenation and shuffling while preserving the 2D convolutional pathway and avoiding spatial flattening. Pyramid-pooled skip connections further reduce the semantic gap between the encoder and decoder and improve small-lesion delineation. Using five-fold cross-validation at the patient level on Prostate158, PI-CAI 2022 and an ethics-approved private cohort of 110 cases, ProMamba achieves Dice scores of 51.39%, 59.46% and 48.49% respectively, outperforming strong CNN, nnU-Net, efficient Transformer/hybrid and Mamba baselines. ProMamba also shows competitive latency and memory usage and generalizes well to PROMISE12 and ACDC, achieving Dice scores of 91.40% and 92.14%, respectively.
-
Journal articleWen K, Ferreira PF, Di Biase Oemick A, et al., 2026,
Evaluation of Third-Order Motion-Compensated Cardiac Diffusion Tensor Imaging Across Cardiac Phases Using an Ultra-High-Performance Clinical Scanner.
, Magn Reson Med, Vol: 96, Pages: 1365-1378PURPOSE: To evaluate a third-order motion-compensated spin echo (M3-MCSE) sequence at multiple cardiac phases on a clinical 3 T MRI scanner with ultra-high performance (UHP) gradients (200 mT/m), compared with stimulated echo acquisition mode (STEAM) and second-order MCSE (M2-MCSE) for cardiac diffusion tensor imaging (cDTI). METHODS: Twenty healthy subjects underwent mid-ventricular short-axis cDTI at peak systole and diastasis using STEAM, M2-MCSE, and M3-MCSE. cDTI metrics and image quality were compared. In five additional healthy subjects, diffusion-weighted images were obtained at multiple trigger delays distributed over diastasis to assess motion-induced signal loss. RESULTS: Compared to M2-MCSE, M3-MCSE yielded higher systolic helix angle map scores ( p = 0.007 ) but lower diastolic scores ( p = 0.001 ), with no significant difference in mean diffusivity, fractional anisotropy, helix angle transmurality or sheetlet angle in systole/diastole. STEAM-derived apparent diffusion coefficients (ADC) were consistent across diastasis, while ADC for MCSE sequences increased at sub-optimal trigger delays. CONCLUSION: UHP gradients enabled in vivo evaluation of M3-MCSE, showing superior systolic cDTI but reduced diastolic performance versus M2-MCSE due to reduced signal-to-noise ratio and a longer motion-sensitive window. Future work may consider numerically optimized gradient designs to enhance MCSE robustness throughout the cardiac cycle.
-
Journal articleLuo Y, Ferreira PF, Wen K, et al., 2026,
Optimized Reduced Field of View and Fat Suppression Methods for Interleaved Multislice In Vivo Cardiac Diffusion Tensor Imaging.
, Magn Reson Med, Vol: 96, Pages: 1097-1110PURPOSE: Slice interleaving, a limited phase encode (PE) field of view (FOV), and effective fat suppression are vital for efficient cardiac diffusion tensor imaging (cDTI) with minimal artifacts. This study aimed to optimize reduced FOV and fat suppression methods for interleaved multislice cDTI to improve signal-to-noise ratio (SNR) and minimize artifacts. METHODS: Two-slice motion compensated spin echo datasets from 20 healthy volunteers were acquired. Four reduced PE FOV sequences were evaluated: 2DRF pulse; applying either 180 ° or 90 ° pulses in PE direction; and the proposed flip-back sequence with a nonselective 180 ° pulse after readout to restore inverted magnetization. Four fat suppression techniques were implemented: no fat suppression (standard); fat saturation; binomial water excitation and spectral attenuated inversion recovery (SPAIR). RESULTS: The proposed flip-back sequence with SPAIR achieved the highest median SNR, and its SNR values are significantly higher ( p < 0.01 ) than 2DRF with SPAIR as current state-of-the-art. SPAIR and water excitation demonstrated comparable performance when combined with the flip-back sequence, and both yielded superior image quality than with no suppression or fat saturation. SPAIR showed robust fat suppression across most subjects, whilst water excitation exhibited advantages in some subjects with a high body mass index. CONCLUSION: The proposed flip-back sequence with SPAIR enables efficient interleaved multislice imaging with reduced PE FOV and effective fat suppression, facilitating clinical translation of in vivo cDTI.
-
Journal articleHasan MK, Wang Q, Shahard H, et al., 2026,
4-D Reconstruction of Fetal Left Ventricle From Echocardiography via 2.5-D Radial Segmentation and Graph-Fourier Reconstruction.
, IEEE Trans Med Imaging, Vol: 45, Pages: 4617-46344D (3D over time) fetal heart reconstruction improves detection and functional assessment of congenital malformations compared with 2D methods, but remains challenging due to the lack of publicly available 4D echocardiography datasets, the burden of full 3D/4D annotations, and the computational cost of volumetric networks. To address these challenges, we introduce a 2.5D radial-slicing paradigm that converts 3D volumes into a set of angularly structured long-axis slices, inducing a consistent U-shaped anatomical appearance that facilitates manual annotation and introduces an acquisition-induced angular symmetry as an effective inductive bias. Based on this slicing, we construct FeEcho4D, the first public benchmark for 4D fetal echocardiography (52 subjects, 1845 annotated volumes). Building on this inductive bias, we propose SCOPE-Net, a symmetry-consistent, prompt-enhanced segmentation network that encodes radial symmetry via novel learnable Flip-Consistent Radial Attention and Symmetry-induced Self-distillation through inter-slice augmentation invariance, enabling label-free representation-level self-supervision. Sparse radial segmentations are subsequently reconstructed into temporally coherent 3D meshes using graph-harmonic deformation, providing a geometry-aware alternative to volumetric segmentation and voxel-based surface extraction without requiring dense 3D annotations. Extensive experiments on FeEcho4D and the public MITEA dataset show that radial segmentation outperforms standard short-, long-axis, and volumetric views, while SCOPE-Net yields anatomically plausible 3D meshes over time. Our framework achieves an 88.8% correlation in ejection fraction, surpassing 3D volumetric segmentation (81.9%) and conventional 2D approaches (64.4%). These results indicate that geometry-aware 2.5D learning can outperform fully volumetric models for 4D fetal cardiac analysis, enabling accurate, efficient, and high-fidelity functional assessment without the need for dense
-
Journal articleLiao Y, Yue P, Wang Y, et al., 2026,
A global-to-local state space model with context-mixing dynamic kernels for medical image classification
, Expert Systems with Applications, Vol: 324, Pages: 132515-132515, ISSN: 0957-4174 -
Journal articleLee K, Jing P, Zhang Z, et al., 2026,
Seeing through experts' eyes: a foundational vision-language model trained on radiologists' gaze and reasoning
, npj Artificial Intelligence, ISSN: 3005-1460Large-scale vision-language models have shown promise in automating chest X-ray interpretation. However, their clinical utility remains limited, since most systems optimize for semantic information rather than emulating how experts visually examine and interpret medical images. As a result, current models often overlook critical findings, misrepresent anatomical context, or diverge from established diagnostic workflows. Radiologists, by contrast, follow structured protocols that sequentially assess anatomical regions, reducing missed findings and supporting reliable diagnostic reasoning. We therefore introduce Gaze-X, a vision-language model that leverages radiologists’ eye-tracking data as a behavioral prior for expert diagnostic reasoning. By incorporating gaze trajectories and fixation patterns into pretraining, Gaze-X learns to follow the spatial and temporal structure of radiologist attention. Using a curated dataset of over 30,000 key frames from five radiologists interpreting chest X-rays across diverse disease categories, we show that Gaze-X produces more accurate, interpretable, and expert-consistent outputs across a range of clinically relevant tasks. Unlike autonomous reporting systems, Gaze-X produces verifiable evidence artifacts, including inspection trajectories and finding-linked localized regions, enabling transparent and safe human-AI collaboration. This capability provides a practical route toward more trustworthy, explainable, and diagnostically robust AI for radiology and beyond.
-
Journal articleDeng F, Chu X, Shi W, et al., 2026,
Radiogenomic modeling of EGFR mutation status in brain metastases from lung adenocarcinoma: a multicenter study with biological interpretability
, npj Digital Medicine, ISSN: 2398-6352Accurate prediction of epidermal growth factor receptor (EGFR) mutation status in lung adenocarcinoma (LUAD) with brain metastases (BMs) is crucial for guiding targeted therapy. However, noninvasive and biologically interpretable tools remain limited. In this multicenter radiogenomic study, we analyzed a total of 1303 BMs from 421 LUAD patients across three institutions. 3435 radiomic features were extracted from T1, T2, and contrast-enhanced T1 sequences. A four-task classification framework was developed to predict EGFR mutation status (EGFR+, 19Del, L858R, or sensitizing mutation) using an adaptive LightGBM-based modeling pipeline. The models achieved excellent performance in the internal cohort (AUCs up to 0.95) and were further validated in 94 lesions with pathologically confirmed EGFR status, reaching an accuracy of 83.0%, sensitivity of 84.7%, and specificity of 80.0%. SHAP and LIME analyses revealed that shape-based radiomic features, particularly original_shape_sphericity, were the most important predictors of EGFR mutational subtypes. Then, we conducted transcriptomic analysis on 38 matched surgical specimens. Radiogenomic correlation revealed that sphericity negatively correlated with RNF125 and SLC37A2. Downstream enrichment analysis identified EGFR-associated features linked to DNA replication, sister chromatid segregation, and ERBB signaling. The study demonstrates that radiogenomic modeling, grounded in interpretable biology, holds promise as a non-invasive, clinical strategy for precision stratification of LUAD BMs.
-
Journal articleVano LJ, Sedlacik J, Kaar SJ, et al., 2026,
Cortical Iron and Mesostriatal Dopamine Function in Schizophrenia: A Positron Emission Tomography and Magnetic Resonance Imaging Study.
, Schizophr Bull, Vol: 52BACKGROUND AND HYPOTHESIS: Elevated postmortem iron in Brodmann areas 10-11 has recently been linked to schizophrenia. Although in vivo studies have linked subcortical iron abnormalities to disease-related striatal hyperdopaminergia, in vivo cortical iron alterations have not been previously examined. We therefore used neuroimaging to test whether cortical iron is elevated in individuals with schizophrenia and whether this correlated with mesostriatal dopamine function. STUDY DESIGN: We acquired quantitative susceptibility mapping magnetic resonance imaging (MRI) to measure magnetic susceptibility (χ), a marker of iron, in 149 participants aged 18-45 (73 with schizophrenia and 76 matched healthy controls). Subsets of patients with schizophrenia underwent [18F]-DOPA PET to estimate striatal dopamine synthesis capacity (n = 39) and neuromelanin-sensitive MRI of the dopaminergic midbrain (n = 68), as neuromelanin is a byproduct of dopamine synthesis. STUDY RESULTS: Primary analyses showed no significant case-control differences in χ in the whole cortex (P = .675) or Brodmann areas 10-11 (P = .537). Exploratory analyses examined χ for 360 cortical regions, correcting for multiple comparisons. Two left temporo-parieto-occipital junction regions showed significantly elevated χ in schizophrenia: the posterior temporo-parieto-occipital junction the posterior temporo-parieto-occipital junction (d = 0.752, P < .001) and the superior temporal visual area (d = 0.638, P = .034). Mean χ across these regions inversely correlated with dopamine synthesis capacity in the associative (r = -0.37, P = .048) and limbic (r = -0.34, P = .048) striatum, and with neuromelanin-sensitive MRI values in the dopaminergic midbrain (r = -0.35, P = .005), corrected for multiple comparisons. CONCLUSIONS: This stu
-
Journal articleNonchev K, Andani S, Ficek-Pascual J, et al., 2026,
Representation learning for multi-modal spatially resolved transcriptomics data.
, Bioinformatics, Vol: 42MOTIVATION: Spatial transcriptomics enables in-depth molecular characterization of samples on a morphology and RNA level while preserving spatial location. Integrating the resulting multi-modal data is an unsolved problem, and developing new solutions in precision medicine depends on improved methodologies. RESULTS: We introduce AESTETIK, a convolutional deep learning model that jointly integrates spatial, transcriptomics, and morphology information to learn accurate spot representations. AESTETIK yielded substantially improved cluster assignments on widely adopted technology platforms (e.g. 10x Genomics™, NanoString™) across multiple datasets. We achieved performance enhancement on structured tissues (e.g. brain) with a 21% increase in median ARI over previous state-of-the-art methods. Notably, AESTETIK also demonstrated superior performance on cancer tissues with heterogeneous cell populations, showing a 2-fold increase in breast cancer, 79% in melanoma, and 21% in liver cancer. We expect that these advances will enable a multi-modal understanding of key biological processes. AVAILABILITY AND IMPLEMENTATION: AESTETIK is implemented in Python 3 and is available as open source software at http://www.github.com/ratschlab/aestetik. The Snakemake pipeline for reproducing the results is available at http://www.github.com/ratschlab/st-rep.
-
Journal articleHuang G, Wu C, Li M, et al., 2026,
Knowledge-guided multi-modality transformer for multi-label genetic mutation prediction
, Pattern Recognition, Vol: 175, Pages: 113047-113047, ISSN: 0031-3203
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.
Contact
For enquiries about the MRI Physics Collective, please contact:
Mary Finnegan
Senior MR Physicist at the Imperial College Healthcare NHS Trust
Pete Lally
Assistant Professor in Magnetic Resonance (MR) Physics at Imperial College
Jan Sedlacik
MR Physicist at the Robert Steiner MR Unit, Hammersmith Hospital Campus