Pub Date : 2025-12-07DOI: 10.1016/j.media.2025.103894
Jiawen Li , Qiehe Sun , Renao Yan , Yizhi Wang , Yuqiu Fu , Yani Wei , Tian Guan , Huijuan Shi , Yonghong He , Anjia Han
With the development of digital imaging in medical microscopy, artificial intelligent-based analysis of pathological whole slide images (WSIs) provides a powerful tool for cancer diagnosis. Limited by the expensive cost of pixel-level annotation, current research primarily focuses on representation learning with slide-level labels, showing success in various downstream tasks. However, given the diversity of lesion types and the complex relationships between each other, these techniques still deserve further exploration in addressing advanced pathology tasks. To this end, we introduce the concept of hierarchical pathological image classification and propose a representation learning called PathTree. PathTree considers the multi-classification of diseases as a binary tree structure. Each category is represented as a professional pathological text description, which messages information with a tree-like encoder. The interactive text features are then used to guide the aggregation of hierarchical multiple representations. PathTree uses slide-text similarity to obtain probability scores and introduces two extra tree-specific losses to further constrain the association between texts and slides. Through extensive experiments on three challenging hierarchical classification datasets: in-house cryosectioned lung tissue lesion identification, public prostate cancer grade assessment, and public breast cancer subtyping, our proposed PathTree is consistently competitive compared to the state-of-the-art methods and provides a new perspective on the deep learning-assisted solution for more complex WSI classification.
{"title":"Diagnostic text-guided representation learning in hierarchical classification for pathological whole slide image","authors":"Jiawen Li , Qiehe Sun , Renao Yan , Yizhi Wang , Yuqiu Fu , Yani Wei , Tian Guan , Huijuan Shi , Yonghong He , Anjia Han","doi":"10.1016/j.media.2025.103894","DOIUrl":"10.1016/j.media.2025.103894","url":null,"abstract":"<div><div>With the development of digital imaging in medical microscopy, artificial intelligent-based analysis of pathological whole slide images (WSIs) provides a powerful tool for cancer diagnosis. Limited by the expensive cost of pixel-level annotation, current research primarily focuses on representation learning with slide-level labels, showing success in various downstream tasks. However, given the diversity of lesion types and the complex relationships between each other, these techniques still deserve further exploration in addressing advanced pathology tasks. To this end, we introduce the concept of hierarchical pathological image classification and propose a representation learning called PathTree. PathTree considers the multi-classification of diseases as a binary tree structure. Each category is represented as a professional pathological text description, which messages information with a tree-like encoder. The interactive text features are then used to guide the aggregation of hierarchical multiple representations. PathTree uses slide-text similarity to obtain probability scores and introduces two extra tree-specific losses to further constrain the association between texts and slides. Through extensive experiments on three challenging hierarchical classification datasets: in-house cryosectioned lung tissue lesion identification, public prostate cancer grade assessment, and public breast cancer subtyping, our proposed PathTree is consistently competitive compared to the state-of-the-art methods and provides a new perspective on the deep learning-assisted solution for more complex WSI classification.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"109 ","pages":"Article 103894"},"PeriodicalIF":11.8,"publicationDate":"2025-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145689528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-07DOI: 10.1016/j.media.2025.103905
Mikaël Simard , Zhuoyan Shen , Konstantin Bräutigam , Rasha Abu-Eid , Maria A. Hawkins , Charles-Antoine Collins Fekete
With the advent of novel cancer treatment options such as immunotherapy, studying the tumour immune micro-environment (TIME) is crucial to inform on prognosis and understand potential response to therapeutic agents. A key approach to characterising the TIME involves combining digitised images of haematoxylin and eosin (H&E) stained tissue sections obtained in routine histopathology examination with automated immune cell detection and classification methods. In this work, we introduce a workflow to automatically generate robust single cell contours and labels from dually stained tissue sections with H&E and multiplexed immunofluorescence (IF) markers. The approach harnesses the Segment Anything Model and requires minimal human intervention compared to existing single cell databases. With this methodology, we create Immunocto, a massive, multi-million automatically generated database of 6,848,454 human cells and objects, including 2,282,818 immune cells distributed across 4 subtypes: CD4 T cell lymphocytes, CD8 T cell lymphocytes, CD20 B cell lymphocytes, and CD68/CD163 macrophages. For each cell, we provide a 64 × 64 pixels2 H&E image at 40 × magnification, along with a binary mask of the nucleus and a label. The database, which is made publicly available, can be used to train models to study the TIME on routine H&E slides. We show that deep learning models trained on Immunocto result in state-of-the-art performance for lymphocyte detection. The approach demonstrates the benefits of using matched H&E and IF data to generate robust databases for computational pathology applications.
{"title":"Immunocto: A massive immune cell database auto-generated for histopathology","authors":"Mikaël Simard , Zhuoyan Shen , Konstantin Bräutigam , Rasha Abu-Eid , Maria A. Hawkins , Charles-Antoine Collins Fekete","doi":"10.1016/j.media.2025.103905","DOIUrl":"10.1016/j.media.2025.103905","url":null,"abstract":"<div><div>With the advent of novel cancer treatment options such as immunotherapy, studying the tumour immune micro-environment (TIME) is crucial to inform on prognosis and understand potential response to therapeutic agents. A key approach to characterising the TIME involves combining digitised images of haematoxylin and eosin (H&E) stained tissue sections obtained in routine histopathology examination with automated immune cell detection and classification methods. In this work, we introduce a workflow to automatically generate robust single cell contours and labels from dually stained tissue sections with H&E and multiplexed immunofluorescence (IF) markers. The approach harnesses the Segment Anything Model and requires minimal human intervention compared to existing single cell databases. With this methodology, we create Immunocto, a massive, multi-million automatically generated database of 6,848,454 human cells and objects, including 2,282,818 immune cells distributed across 4 subtypes: CD4<span><math><msup><mrow></mrow><mo>+</mo></msup></math></span> T cell lymphocytes, CD8<span><math><msup><mrow></mrow><mo>+</mo></msup></math></span> T cell lymphocytes, CD20<span><math><msup><mrow></mrow><mo>+</mo></msup></math></span> B cell lymphocytes, and CD68<span><math><msup><mrow></mrow><mo>+</mo></msup></math></span>/CD163<span><math><msup><mrow></mrow><mo>+</mo></msup></math></span> macrophages. For each cell, we provide a 64 × 64 pixels<sup>2</sup> H&E image at <strong>40</strong> × magnification, along with a binary mask of the nucleus and a label. The database, which is made publicly available, can be used to train models to study the TIME on routine H&E slides. We show that deep learning models trained on Immunocto result in state-of-the-art performance for lymphocyte detection. The approach demonstrates the benefits of using matched H&E and IF data to generate robust databases for computational pathology applications.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"109 ","pages":"Article 103905"},"PeriodicalIF":11.8,"publicationDate":"2025-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145705011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-04DOI: 10.1016/j.media.2025.103903
Benjamin Sinclair , William Pham , Lucy Vivash , Jasmine Moses , Miranda Lynch , Karina Dorfman , Cassandra Marotta , Shaun Koh , Jacob Bunyamin , Ella Rowsthorn , Alex Jarema , Himashi Peiris , Zhaolin Chen , Sandy R Shultz , David K Wright , Dexiao Kong , Sharon L. Naismith , Terence J. O’Brien , Meng Law
Perivascular spaces (PVSs) form a central component of the brain’s waste clearance system, the glymphatic system. These structures are visible on MRIs when enlarged, and their morphology is associated with aging and neurological disease. Manual quantification of PVS is time consuming and subjective. Numerous deep learning methods for PVS segmentation have been developed for automated segmentation. However, the majority of these algorithms have been developed and evaluated on homogenous datasets and high resolution scans, perhaps limiting their applicability for the wide range of image qualities acquired in clinical and research settings. In this work we train a nnUNet, a top-performing task driven biomedical image segmentation deep learning algorithm, on a heterogenous training sample of manually segmented MRIs of a range of different qualities and resolutions from 7 different datasets acquired on 6 different scanners. These are compared to the two currently publicly available deep learning methods for 3D segmentation of PVS, evaluated on scans with a range of resolutions and qualities. The resulting model, PINGU (Perivascular space Identification Nnunet for Generalised Usage), achieved voxel and cluster level dice scores of 0.50(SD=0.15) and 0.63(0.17) in the white matter (WM), and 0.54 (0.11) and 0.66(0.17) in the basal ganglia (BG). Performance on unseen “external” sites’ data was substantially lower for both PINGU (0.20-0.38 [WM, voxel], 0.29-0.58 [WM, cluster], 0.22-0.36 [BG, voxel], 0.46-0.60 [BG, cluster]) and the publicly available algorithms (0.18-0.30 [WM, voxel], 0.29-0.38 [WM cluster], 0.10-0.20 [BG, voxel], 0.15-0.37 [BG, cluster]). Nonetheless, PINGU strongly outperformed the publicly available algorithms, particularly in the BG. PINGU stands out as broad-use PVS segmentation tool, with particular strength in the BG, an area of PVS highly related to vascular disease and pathology.
{"title":"Perivascular space identification nnUNet for generalised usage (PINGU)","authors":"Benjamin Sinclair , William Pham , Lucy Vivash , Jasmine Moses , Miranda Lynch , Karina Dorfman , Cassandra Marotta , Shaun Koh , Jacob Bunyamin , Ella Rowsthorn , Alex Jarema , Himashi Peiris , Zhaolin Chen , Sandy R Shultz , David K Wright , Dexiao Kong , Sharon L. Naismith , Terence J. O’Brien , Meng Law","doi":"10.1016/j.media.2025.103903","DOIUrl":"10.1016/j.media.2025.103903","url":null,"abstract":"<div><div>Perivascular spaces (PVSs) form a central component of the brain’s waste clearance system, the glymphatic system. These structures are visible on MRIs when enlarged, and their morphology is associated with aging and neurological disease. Manual quantification of PVS is time consuming and subjective. Numerous deep learning methods for PVS segmentation have been developed for automated segmentation. However, the majority of these algorithms have been developed and evaluated on homogenous datasets and high resolution scans, perhaps limiting their applicability for the wide range of image qualities acquired in clinical and research settings. In this work we train a nnUNet, a top-performing task driven biomedical image segmentation deep learning algorithm, on a heterogenous training sample of manually segmented MRIs of a range of different qualities and resolutions from 7 different datasets acquired on 6 different scanners. These are compared to the two currently publicly available deep learning methods for 3D segmentation of PVS, evaluated on scans with a range of resolutions and qualities. The resulting model, PINGU (Perivascular space Identification Nnunet for Generalised Usage), achieved voxel and cluster level dice scores of 0.50(SD=0.15) and 0.63(0.17) in the white matter (WM), and 0.54 (0.11) and 0.66(0.17) in the basal ganglia (BG). Performance on unseen “external” sites’ data was substantially lower for both PINGU (0.20-0.38 [WM, voxel], 0.29-0.58 [WM, cluster], 0.22-0.36 [BG, voxel], 0.46-0.60 [BG, cluster]) and the publicly available algorithms (0.18-0.30 [WM, voxel], 0.29-0.38 [WM cluster], 0.10-0.20 [BG, voxel], 0.15-0.37 [BG, cluster]). Nonetheless, PINGU strongly outperformed the publicly available algorithms, particularly in the BG. PINGU stands out as broad-use PVS segmentation tool, with particular strength in the BG, an area of PVS highly related to vascular disease and pathology.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"109 ","pages":"Article 103903"},"PeriodicalIF":11.8,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145689529","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-03DOI: 10.1016/j.media.2025.103892
Ruixi Zheng , Wei Zhang , Yijie Li , Xi Zhu , Zhou Lan , Jarrett Rushmore , Yogesh Rathi , Nikos Makris , Lauren J. O’Donnell , Fan Zhang
Diffusion MRI (dMRI) tractography is currently the only method for in vivo mapping of the brain’s white matter (WM) connections. Tractometry is an advanced tractography analysis technique for along-tract profiling to investigate the morphology and microstructural properties along the fiber tracts. Tractometry has become an essential tool for studying local along-tract differences between different populations (e.g., health vs disease). In this study, we propose a novel atlas-guided fine-scale tractometry method, namely AGFS-Tractometry, that leverages tract spatial information and permutation testing to enhance the along-tract statistical analysis between populations. There are two major contributions in AGFS-Tractometry. First, we create a novel atlas-guided tract profiling template that enables consistent, fine-scale, along-tract parcellation of subject-specific fiber tracts. Second, we propose a novel nonparametric permutation testing group comparison method to enable simultaneous analysis across all along-tract parcels while correcting for multiple comparisons. We perform experimental evaluations on synthetic datasets with known group differences and in vivo real data. We compare AGFS-Tractometry with two state-of-the-art tractometry methods, including Automated Fiber-tract Quantification (AFQ) and BUndle ANalytics (BUAN). Our results show that the proposed AGFS-Tractometry obtains enhanced sensitivity and specificity in detecting local WM differences. In the real data analysis experiments, AGFS-Tractometry can identify more regions with significant differences, which are anatomically consistent with the existing literature. Overall, these demonstrate the ability of AGFS-Tractometry to detect subtle or spatially localized WM group-level differences. The created tract profiling template and related code are available at: https://github.com/ZhengRuixi/AGFS-Tractometry.git.
{"title":"AGFS-tractometry: A novel atlas-guided fine-scale tractometry approach for enhanced along-tract group statistical comparison using diffusion MRI tractography","authors":"Ruixi Zheng , Wei Zhang , Yijie Li , Xi Zhu , Zhou Lan , Jarrett Rushmore , Yogesh Rathi , Nikos Makris , Lauren J. O’Donnell , Fan Zhang","doi":"10.1016/j.media.2025.103892","DOIUrl":"10.1016/j.media.2025.103892","url":null,"abstract":"<div><div>Diffusion MRI (dMRI) tractography is currently the only method for in vivo mapping of the brain’s white matter (WM) connections. Tractometry is an advanced tractography analysis technique for along-tract profiling to investigate the morphology and microstructural properties along the fiber tracts. Tractometry has become an essential tool for studying local along-tract differences between different populations (e.g., health vs disease). In this study, we propose a novel atlas-guided fine-scale tractometry method, namely AGFS-Tractometry, that leverages tract spatial information and permutation testing to enhance the along-tract statistical analysis between populations. There are two major contributions in AGFS-Tractometry. First, we create a novel atlas-guided tract profiling template that enables consistent, fine-scale, along-tract parcellation of subject-specific fiber tracts. Second, we propose a novel nonparametric permutation testing group comparison method to enable simultaneous analysis across all along-tract parcels while correcting for multiple comparisons. We perform experimental evaluations on synthetic datasets with known group differences and in vivo real data. We compare AGFS-Tractometry with two state-of-the-art tractometry methods, including Automated Fiber-tract Quantification (AFQ) and BUndle ANalytics (BUAN). Our results show that the proposed AGFS-Tractometry obtains enhanced sensitivity and specificity in detecting local WM differences. In the real data analysis experiments, AGFS-Tractometry can identify more regions with significant differences, which are anatomically consistent with the existing literature. Overall, these demonstrate the ability of AGFS-Tractometry to detect subtle or spatially localized WM group-level differences. The created tract profiling template and related code are available at: <span><span>https://github.com/ZhengRuixi/AGFS-Tractometry.git</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"109 ","pages":"Article 103892"},"PeriodicalIF":11.8,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145689531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-03DOI: 10.1016/j.media.2025.103881
Jonas M. Van Elburg , Natalia V. Korobova , Mohammad M. Islam , Marian A. Troelstra , Oliver J. Gurney-Champion
Dynamic contrast-enhanced (DCE) MRI is a powerful technique for detecting and characterising various diseases by quantifying tissue perfusion. However, accurate perfusion quantification remains challenging due to noisy data and the complexity of pharmacokinetic modelling. Conventional non-linear least squares (NLLS) fitting often yields noisy parameter maps. Although deep-learning algorithms generate smoother, more visually appealing maps, these may lure clinicians into a false sense of security when the maps are incorrect. Hence, reliable uncertainty estimation is crucial for assessing model performance and ensuring clinical confidence.
Therefore, we implemented an ensemble of mean-variance estimation (MVE) neural networks to quantify perfusion parameters alongside aleatoric (data-driven) and epistemic (model-driven) uncertainties in DCE-MRI. We compared MVE with NLLS and a physics-informed neural network (PINN), both of which used conventional covariance matrix-based uncertainty estimation.
Simulations demonstrated that MVE achieved the highest accuracy in perfusion and uncertainty estimates. MVE’s aleatoric uncertainty strongly correlated with true errors, whereas NLLS and PINN tended to overestimate uncertainty. Epistemic uncertainty was significantly higher for the data deviating from what was encountered in training (out-of-distribution) in both MVE and PINN ensembles. In vivo, MVE produced smoother and more reliable uncertainty maps than NLLS and PINN, which exhibited outliers and overestimation. Within a liver region of interest, MVE’s uncertainty estimates matched the standard deviation of the data more closely than NLLS and PINN, making it the most accurate method.
In conclusion, an MVE enhances quantitative DCE-MRI by providing robust uncertainty estimates alongside perfusion parameters. This approach improves the reliability of AI-driven MRI analysis, supporting clinical translation.
{"title":"Uncertainty estimates in pharmacokinetic modelling of DCE-MRI","authors":"Jonas M. Van Elburg , Natalia V. Korobova , Mohammad M. Islam , Marian A. Troelstra , Oliver J. Gurney-Champion","doi":"10.1016/j.media.2025.103881","DOIUrl":"10.1016/j.media.2025.103881","url":null,"abstract":"<div><div>Dynamic contrast-enhanced (DCE) MRI is a powerful technique for detecting and characterising various diseases by quantifying tissue perfusion. However, accurate perfusion quantification remains challenging due to noisy data and the complexity of pharmacokinetic modelling. Conventional non-linear least squares (NLLS) fitting often yields noisy parameter maps. Although deep-learning algorithms generate smoother, more visually appealing maps, these may lure clinicians into a false sense of security when the maps are incorrect. Hence, reliable uncertainty estimation is crucial for assessing model performance and ensuring clinical confidence.</div><div>Therefore, we implemented an ensemble of mean-variance estimation (MVE) neural networks to quantify perfusion parameters alongside aleatoric (data-driven) and epistemic (model-driven) uncertainties in DCE-MRI. We compared MVE with NLLS and a physics-informed neural network (PINN), both of which used conventional covariance matrix-based uncertainty estimation.</div><div>Simulations demonstrated that MVE achieved the highest accuracy in perfusion and uncertainty estimates. MVE’s aleatoric uncertainty strongly correlated with true errors, whereas NLLS and PINN tended to overestimate uncertainty. Epistemic uncertainty was significantly higher for the data deviating from what was encountered in training (out-of-distribution) in both MVE and PINN ensembles. In vivo, MVE produced smoother and more reliable uncertainty maps than NLLS and PINN, which exhibited outliers and overestimation. Within a liver region of interest, MVE’s uncertainty estimates matched the standard deviation of the data more closely than NLLS and PINN, making it the most accurate method.</div><div>In conclusion, an MVE enhances quantitative DCE-MRI by providing robust uncertainty estimates alongside perfusion parameters. This approach improves the reliability of AI-driven MRI analysis, supporting clinical translation.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"109 ","pages":"Article 103881"},"PeriodicalIF":11.8,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145689530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-30DOI: 10.1016/j.media.2025.103888
Wei Fu , Yufei Chen , Yuqi Liu , Xiaodong Yue
In computer-aided diagnosis, the extreme imbalance in disease incidence rates often results in the omission of rare conditions, leading to out-of-distribution (OOD) samples during testing. To prevent unreliable diagnostic outputs, detecting these OOD samples becomes essential for clinical jnsafety. While Evidential Deep Learning (EDL) and its variants have shown great promise in detecting outliers, their clinical application remains challenging due to the variability in medical images. We find that when encountering samples with high data uncertainty, the Kullback-Leibler divergence (KL) in EDL tends to suppress inherent ambiguity, resulting in an over-penalty effect in evidence estimation that impairs discrimination between ambiguous in-distribution cases and true outliers. Motivated by the confirmatory and differential diagnostic process in clinical practice, we propose Differential Evidential Deep Learning (D-EDL), a simple but effective method for robust OOD detection. Specifically, we treat KL as a confirmatory restriction and innovatively replace it with a Ruling Out Module (ROM) for differential restriction, which reduces over-penalty on ambiguous ID samples while maintaining OOD sensitivity. Considering extreme testing scenarios, we introduce test-time Raw evidence Inference (RI) to bypass instability in uncertainty estimation with refined evidence and further improve robustness and precision. Finally, we propose the Balanced Detection Score (BDS) to quantify the potential on clinical performance when optimally balancing misdiagnoses and missed diagnoses across varying sensitivities. Experimental results on ISIC2019, Bone Marrow Cytomorphology datasets and EDDFS dataset demonstrate that our D-EDL outperforms state-of-the-art OOD detection methods, achieving significant improvements in robustness and clinical applicability. Code for D-EDL is available at https://github.com/KellaDoe/Differential_EDL.
{"title":"D-EDL: Differential evidential deep learning for robust medical out-of-distribution detection","authors":"Wei Fu , Yufei Chen , Yuqi Liu , Xiaodong Yue","doi":"10.1016/j.media.2025.103888","DOIUrl":"10.1016/j.media.2025.103888","url":null,"abstract":"<div><div>In computer-aided diagnosis, the extreme imbalance in disease incidence rates often results in the omission of rare conditions, leading to out-of-distribution (OOD) samples during testing. To prevent unreliable diagnostic outputs, detecting these OOD samples becomes essential for clinical jnsafety. While Evidential Deep Learning (EDL) and its variants have shown great promise in detecting outliers, their clinical application remains challenging due to the variability in medical images. We find that when encountering samples with high data uncertainty, the Kullback-Leibler divergence (KL) in EDL tends to suppress inherent ambiguity, resulting in an over-penalty effect in evidence estimation that impairs discrimination between ambiguous in-distribution cases and true outliers. Motivated by the confirmatory and differential diagnostic process in clinical practice, we propose Differential Evidential Deep Learning (D-EDL), a simple but effective method for robust OOD detection. Specifically, we treat KL as a confirmatory restriction and innovatively replace it with a Ruling Out Module (ROM) for differential restriction, which reduces over-penalty on ambiguous ID samples while maintaining OOD sensitivity. Considering extreme testing scenarios, we introduce test-time Raw evidence Inference (RI) to bypass instability in uncertainty estimation with refined evidence and further improve robustness and precision. Finally, we propose the Balanced Detection Score (BDS) to quantify the potential on clinical performance when optimally balancing misdiagnoses and missed diagnoses across varying sensitivities. Experimental results on ISIC2019, Bone Marrow Cytomorphology datasets and EDDFS dataset demonstrate that our D-EDL outperforms state-of-the-art OOD detection methods, achieving significant improvements in robustness and clinical applicability. Code for D-EDL is available at <span><span>https://github.com/KellaDoe/Differential_EDL</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"109 ","pages":"Article 103888"},"PeriodicalIF":11.8,"publicationDate":"2025-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145619565","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-29DOI: 10.1016/j.media.2025.103890
Veronika Spieker , Hannah Eichhorn , Wenqi Huang , Jonathan K. Stelter , Tabita Catalan , Rickmer F. Braren , Daniel Rueckert , Francisco Sahli Costabal , Kerstin Hammernik , Dimitrios C. Karampinos , Claudia Prieto , Julia A. Schnabel
Neural implicit k-space representations (NIK) have shown promising results for dynamic magnetic resonance imaging (MRI) at high temporal resolutions. Yet, reducing acquisition time, and thereby available training data, results in severe performance drops due to overfitting. To address this, we introduce a novel self-supervised k-space loss function , applicable for regularization of NIK-based reconstructions. The proposed loss function is based on the concept of parallel imaging-inspired self-consistency (PISCO), enforcing a consistent global k-space neighborhood relationship without requiring additional data. Quantitative and qualitative evaluations on static and dynamic MR reconstructions show that integrating PISCO significantly improves NIK representations, making it a competitive dynamic reconstruction method without constraining the temporal resolution. Particularly at high acceleration factors (R ≥ 50), NIK with PISCO can avoid temporal oversmoothing of state-of-the-art methods and achieves superior spatio-temporal reconstruction quality. Furthermore, an extensive analysis of the loss assumptions and stability shows PISCO’s potential as versatile self-supervised k-space loss function for further applications and architectures. Code is available at: https://github.com/compai-lab/2025-pisco-spieker
{"title":"PISCO: Self-supervised k-space regularization for improved neural implicit k-space representations of dynamic MRI","authors":"Veronika Spieker , Hannah Eichhorn , Wenqi Huang , Jonathan K. Stelter , Tabita Catalan , Rickmer F. Braren , Daniel Rueckert , Francisco Sahli Costabal , Kerstin Hammernik , Dimitrios C. Karampinos , Claudia Prieto , Julia A. Schnabel","doi":"10.1016/j.media.2025.103890","DOIUrl":"10.1016/j.media.2025.103890","url":null,"abstract":"<div><div>Neural implicit k-space representations (NIK) have shown promising results for dynamic magnetic resonance imaging (MRI) at high temporal resolutions. Yet, reducing acquisition time, and thereby available training data, results in severe performance drops due to overfitting. To address this, we introduce a novel self-supervised k-space loss function <span><math><msub><mi>L</mi><mtext>PISCO</mtext></msub></math></span>, applicable for regularization of NIK-based reconstructions. The proposed loss function is based on the concept of parallel imaging-inspired self-consistency (PISCO), enforcing a consistent global k-space neighborhood relationship without requiring additional data. Quantitative and qualitative evaluations on static and dynamic MR reconstructions show that integrating PISCO significantly improves NIK representations, making it a competitive dynamic reconstruction method without constraining the temporal resolution. Particularly at high acceleration factors (R ≥ 50), NIK with PISCO can avoid temporal oversmoothing of state-of-the-art methods and achieves superior spatio-temporal reconstruction quality. Furthermore, an extensive analysis of the loss assumptions and stability shows PISCO’s potential as versatile self-supervised k-space loss function for further applications and architectures. Code is available at: <span><span>https://github.com/compai-lab/2025-pisco-spieker</span><svg><path></path></svg></span></div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"109 ","pages":"Article 103890"},"PeriodicalIF":11.8,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145619566","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-28DOI: 10.1016/j.media.2025.103891
Siddharth Mittal, Michael Woletz, David Linhardt, Christian Windischberger
Population receptive field (pRF) mapping is a fundamental technique for understanding retinotopic organisation of the human visual system. Since its introduction in 2008, however, its scalability has been severely hindered by the computational bottleneck of iterative parameter refinement. Current state-of-the-art implementations either sacrifice precision for speed or rely on slow iterative parameter updates, limiting their applicability to large-scale datasets. Here, we present a novel mathematical reformulation of the General Linear Model (GLM), wrapped in a GPU-Empowered Mapping of population Receptive Fields (GEM-pRF) software implementation. By orthogonalizing the design matrix, our approach enables the direct and fast computation of the objective function’s derivatives, which are used to eliminate the iterative refinement process. This approach dramatically accelerates pRF estimation with high accuracy. Validation using empirical and simulated data confirms GEM-pRF’s accuracy, and benchmarking against established tools demonstrates a reduction in computation time of almost two orders of magnitude. With its modular and extensible design, GEM-pRF provides a critical advancement for large-scale fMRI retinotopic mapping. Furthermore, our reformulated GLM approach in combination with GPU-based implementation offers a broadly applicable solution that may extend beyond visual neuroscience, accelerating computational modelling across various domains in neuroimaging and beyond.
{"title":"GEM-pRF: GPU-empowered mapping of population receptive fields for large-scale fMRI analysis","authors":"Siddharth Mittal, Michael Woletz, David Linhardt, Christian Windischberger","doi":"10.1016/j.media.2025.103891","DOIUrl":"10.1016/j.media.2025.103891","url":null,"abstract":"<div><div>Population receptive field (pRF) mapping is a fundamental technique for understanding retinotopic organisation of the human visual system. Since its introduction in 2008, however, its scalability has been severely hindered by the computational bottleneck of iterative parameter refinement. Current state-of-the-art implementations either sacrifice precision for speed or rely on slow iterative parameter updates, limiting their applicability to large-scale datasets. Here, we present a novel mathematical reformulation of the General Linear Model (GLM), wrapped in a GPU-Empowered Mapping of population Receptive Fields (GEM-pRF) software implementation. By orthogonalizing the design matrix, our approach enables the direct and fast computation of the objective function’s derivatives, which are used to eliminate the iterative refinement process. This approach dramatically accelerates pRF estimation with high accuracy. Validation using empirical and simulated data confirms GEM-pRF’s accuracy, and benchmarking against established tools demonstrates a reduction in computation time of almost two orders of magnitude. With its modular and extensible design, GEM-pRF provides a critical advancement for large-scale fMRI retinotopic mapping. Furthermore, our reformulated GLM approach in combination with GPU-based implementation offers a broadly applicable solution that may extend beyond visual neuroscience, accelerating computational modelling across various domains in neuroimaging and beyond.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"109 ","pages":"Article 103891"},"PeriodicalIF":11.8,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145611717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-25DOI: 10.1016/j.media.2025.103889
Junzhuo Liu , Markus Eckstein , Zhixiang Wang , Friedrich Feuerhake , Dorit Merhof
Spatial transcriptomics is a technology that captures gene expression at different spatial locations, widely used in tumor microenvironment analysis and molecular profiling of histopathology, providing valuable insights into resolving gene expression and clinical diagnosis of cancer. Due to the high cost of data acquisition, large-scale spatial transcriptomics data remain challenging to obtain. In this study, we develop a contrastive learning-based deep learning method to predict spatially resolved gene expression from the whole-slide images (WSIs). Unlike existing end-to-end prediction frameworks, our method leverages multi-modal contrastive learning to establish a correspondence between histopathological morphology and spatial gene expression in the feature space. By computing cross-modal feature similarity, our method generates spatially resolved gene expression directly from WSIs. Furthermore, to enhance the standard contrastive learning paradigm, a cross-modal masked reconstruction is designed as a pretext task, enabling feature-level fusion between modalities. Notably, our method does not rely on large-scale pretraining datasets or abstract semantic representations from either modality, making it particularly effective for scenarios with limited spatial transcriptomics data. Evaluation across six different disease datasets demonstrates that, compared to existing studies, our method improves Pearson Correlation Coefficient (PCC) in the prediction of highly expressed genes, highly variable genes, and marker genes by 6.27 %, 6.11 %, and 11.26 % respectively. Further analysis indicates that our method preserves gene-gene correlations and applies to datasets with limited samples. Additionally, our method exhibits potential in cancer tissue localization based on biomarker expression. The code repository for this work is available at https://github.com/ngfufdrdh/CMRCNet.
{"title":"Spatial transcriptomics expression prediction from histopathology based on cross-modal mask reconstruction and contrastive learning","authors":"Junzhuo Liu , Markus Eckstein , Zhixiang Wang , Friedrich Feuerhake , Dorit Merhof","doi":"10.1016/j.media.2025.103889","DOIUrl":"10.1016/j.media.2025.103889","url":null,"abstract":"<div><div>Spatial transcriptomics is a technology that captures gene expression at different spatial locations, widely used in tumor microenvironment analysis and molecular profiling of histopathology, providing valuable insights into resolving gene expression and clinical diagnosis of cancer. Due to the high cost of data acquisition, large-scale spatial transcriptomics data remain challenging to obtain. In this study, we develop a contrastive learning-based deep learning method to predict spatially resolved gene expression from the whole-slide images (WSIs). Unlike existing end-to-end prediction frameworks, our method leverages multi-modal contrastive learning to establish a correspondence between histopathological morphology and spatial gene expression in the feature space. By computing cross-modal feature similarity, our method generates spatially resolved gene expression directly from WSIs. Furthermore, to enhance the standard contrastive learning paradigm, a cross-modal masked reconstruction is designed as a pretext task, enabling feature-level fusion between modalities. Notably, our method does not rely on large-scale pretraining datasets or abstract semantic representations from either modality, making it particularly effective for scenarios with limited spatial transcriptomics data. Evaluation across six different disease datasets demonstrates that, compared to existing studies, our method improves Pearson Correlation Coefficient (PCC) in the prediction of highly expressed genes, highly variable genes, and marker genes by 6.27 %, 6.11 %, and 11.26 % respectively. Further analysis indicates that our method preserves gene-gene correlations and applies to datasets with limited samples. Additionally, our method exhibits potential in cancer tissue localization based on biomarker expression. The code repository for this work is available at <span><span>https://github.com/ngfufdrdh/CMRCNet</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"108 ","pages":"Article 103889"},"PeriodicalIF":11.8,"publicationDate":"2025-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145592833","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-24DOI: 10.1016/j.media.2025.103887
Junyu Chen , Shuwen Wei , Yihao Liu , Zhangxing Bian , Yufan He , Aaron Carass , Harrison Bai , Yong Du
Spatially varying regularization accommodates the deformation variations that may be necessary for different anatomical regions during deformable image registration. Historically, optimization-based registration models have harnessed spatially varying regularization to address anatomical subtleties. However, most modern deep learning-based models tend to gravitate towards spatially invariant regularization, wherein a homogenous regularization strength is applied across the entire image, potentially disregarding localized variations. In this paper, we propose a hierarchical probabilistic model that integrates a prior distribution on the deformation regularization strength, enabling the end-to-end learning of a spatially varying deformation regularizer directly from the data. The proposed method is straightforward to implement and easily integrates with various registration network architectures. Additionally, automatic tuning of hyperparameters is achieved through Bayesian optimization, allowing efficient identification of optimal hyperparameters for any given registration task. Comprehensive evaluations on publicly available datasets demonstrate that the proposed method significantly improves registration performance and enhances the interpretability of deep learning-based registration, all while maintaining smooth deformations. Our code is freely available at http://bit.ly/3BrXGxz.
{"title":"Unsupervised learning of spatially varying regularization for diffeomorphic image registration","authors":"Junyu Chen , Shuwen Wei , Yihao Liu , Zhangxing Bian , Yufan He , Aaron Carass , Harrison Bai , Yong Du","doi":"10.1016/j.media.2025.103887","DOIUrl":"10.1016/j.media.2025.103887","url":null,"abstract":"<div><div>Spatially varying regularization accommodates the deformation variations that may be necessary for different anatomical regions during deformable image registration. Historically, optimization-based registration models have harnessed spatially varying regularization to address anatomical subtleties. However, most modern deep learning-based models tend to gravitate towards spatially invariant regularization, wherein a homogenous regularization strength is applied across the entire image, potentially disregarding localized variations. In this paper, we propose a hierarchical probabilistic model that integrates a prior distribution on the deformation regularization strength, enabling the end-to-end learning of a spatially varying deformation regularizer directly from the data. The proposed method is straightforward to implement and easily integrates with various registration network architectures. Additionally, automatic tuning of hyperparameters is achieved through Bayesian optimization, allowing efficient identification of optimal hyperparameters for any given registration task. Comprehensive evaluations on publicly available datasets demonstrate that the proposed method significantly improves registration performance and enhances the interpretability of deep learning-based registration, all while maintaining smooth deformations. Our code is freely available at <span><span>http://bit.ly/3BrXGxz</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"108 ","pages":"Article 103887"},"PeriodicalIF":11.8,"publicationDate":"2025-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145592834","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号