BMC Medical Imaging[JOURNAL]
Sun
200 papers
Multi-organ metabolic connectivity mapping in advanced pulmonary arterial hypertension: a group-level 18 F-FDG PET-controlled study.
Rey MA, Varet S, Roche A … +10 more , Ghidaglia J, Mas R, Jentreau Y, Montani D, Onephandara S, Humbert M, Amoin A, Savale L, Faure S, Besson FL
BMC Med Imaging · 2026 Jun · PMID 42310556 · Full text
BACKGROUND: To explore potential systemic alterations in carbohydrate metabolism associated with advanced pulmonary arterial hypertension (PAH) using group-level PET-based metabolic connectivity mapping. METHODS: This re...
BACKGROUND: To explore potential systemic alterations in carbohydrate metabolism associated with advanced pulmonary arterial hypertension (PAH) using group-level PET-based metabolic connectivity mapping. METHODS: This retrospective, controlled study analysed F-FDG PET-CT scans from 75 individuals (29 with PAH, 46 controls). Eleven major organs were segmented using AI-based tools and voxel-level PET data were extracted. Inter-organ metabolic profiles in PAH and control groups were evaluated using a Maximum mean discrepancy (MMD) framework with extensive permutation testing (1,000,000 permutations) to assess intra-group homogeneity and detect between-group distributional differences. Upon confirmation of within-group homogeneity, organ-level metabolic connectomes were derived from Spearman correlation matrices, with Holm-adjusted multiple testing correction and significance filtering (t-test, p < 0.05). Between-group comparisons were similarly performed using MMD, flowed by post-hoc organ-pair analyses and construction of differential connectomes to localize statistical metabolic networks divergence. RESULTS: Standard SUV analysis revealed no significant intergroup differences across most organs except for increased uptake in the right heart in PAH patients. MMD testing confirmed intra-group homogeneity in both controls (0,0048) and PAH (0,0465), with no rejection of H at α = 0.05, while demonstrating significant between-group differences (H rejected). Spearman-based PET connectomes, retaining only significant correlations (𝜌 ≠ 0; p < 0.05) revealed perturbated metabolic network in PAH. This altered network involved the heart, adipose tissue, liver, spleen, muscle, and bone marrow. CONCLUSION: Group-level whole-body F-FDG PET connectivity analysis may provide additional insights into systemic metabolic alterations in advanced PAH that are not readily captured by conventional regional SUV assessments. These findings suggest that PET-based connectomics could complement existing methods for assessing metabolic involvement in PAH.
Advancing fungal sinusitis diagnosis: a radiomics and machine learning approach.
Li S, Li R, Ma Q … +3 more , Zheng C, Yan X, Chen S
BMC Med Imaging · 2026 Jun · PMID 42304280 · Full text
BACKGROUND: Differentiating fungal from chronic sinusitis remains a diagnostic challenge due to overlapping symptoms. OBJECTIVES: This study aims to enhance diagnostic accuracy for fungal sinusitis by integrating radiomi...
BACKGROUND: Differentiating fungal from chronic sinusitis remains a diagnostic challenge due to overlapping symptoms. OBJECTIVES: This study aims to enhance diagnostic accuracy for fungal sinusitis by integrating radiomic analysis with machine learning. METHODS: Data from 106 fungal and 146 chronic sinusitis patients, confirmed by surgical pathology at the Integrated Traditional Chinese and Western Medicine Hospital of Wenzhou Medical University from January 2022 to December 2025, were analyzed. Radiomic features from CT scans were extracted and reduced to 19 key indicators using 3DSlicer, Minimum Redundancy Maximum Relevance (mRMR), and lasso regression. A comparative analysis of logistic regression, support vector machine, and random forest models selected the optimal model based on AUC performance. RESULTS: Using mRMR followed by LASSO regression, 19 key radiomic features were selected to distinguish fungal sinusitis from chronic sinusitis. Among logistic regression, SVM, and random forest, the random forest model performed best, achieving a training AUC of 96.04% and a test AUC of 94.91%, with corresponding accuracies of 92.51% and 91.30%. Calibration curves confirmed excellent agreement between predicted and actual outcomes, demonstrating strong diagnostic reliability. Calibration curves showed strong agreement between predicted probabilities and actual outcomes. CONCLUSIONS: Employing radiomic features and machine learning, specifically a refined random forest model, significantly improves fungal sinusitis diagnosis accuracy. This approach promises to minimize diagnostic errors and enhance therapeutic decisions, marking a significant advance in precise diagnostics for fungal sinusitis. However, external validation from independent institutions was not performed, representing a key limitation of the study. CLINICAL TRIAL NUMBER: Not applicable.
Radiomics-based unsupervised clustering and deep learning nomogram for response prediction after neoadjuvant chemotherapy in locally advanced laryngeal cancer.
Li Z, Dong H, Ma Z … +2 more , Shen Z, Xia S
BMC Med Imaging · 2026 Jun · PMID 42304259 · Full text
OBJECTIVES: To develop a nomogram model for individualized prediction of neoadjuvant chemotherapy (NAC) response in locally advanced laryngeal cancer (LALC). METHODS: A total of 175 patients who underwent CT examinations...
OBJECTIVES: To develop a nomogram model for individualized prediction of neoadjuvant chemotherapy (NAC) response in locally advanced laryngeal cancer (LALC). METHODS: A total of 175 patients who underwent CT examinations before NAC from two hospitals were retrospectively enrolled and divided into training (n = 112) and test (n = 63) sets. Significant radiomics and clinical features were selected sequentially, and consensus clustering was used to identify tumor subtypes. A nomogram model for predicting remission status was constructed by fusing clinical signature, radiomics signature, cluster result and deep learning signature. The predictive performance of junior and senior doctors for NAC response was evaluated with and without model assistance. RESULTS: Two radiomics subtypes (Subtype A and B) were identified. Compared with Subtype A, Subtype B had higher frequencies of heterogeneous tumor density (all P < 0.001) and marked venous-phase enhancement (all P < 0.05) in both sets. The nomogram-derived remission score showed good predictive performance for NAC response, with AUCs of 0.935 (training set) and 0.912 (test set). Survival analysis revealed that patients with high remission scores had better overall survival (OS) and locoregional control than those with low scores (all P < 0.05) in both sets. Model assistance improved the predictive performance of junior (AUC: 0.799 vs. 0.915) and senior doctors (AUC: 0.849 vs. 0.919, all P < 0.05). CONCLUSION: The nomogram model based on two-center databases achieved good performance in predicting NAC response and survival in LALC patients, which may contribute to the personalized treatment of LALC.
Radiomics using contrast-enhanced T1-weighted imaging and clinical features for predicting response to EGFR-TKIs in EGFR-mutated non-small cell lung cancer patients with brain metastases.
Sui LY, Zhang TY, Cheng C … +8 more , Xing LH, Meng H, Liu C, Wang Q, Wang JN, Zhang TS, Liu K, Yin XP
BMC Med Imaging · 2026 Jun · PMID 42304254 · Full text
OBJECTIVES: To construct and validate a model based on clinical characteristics and magnetic resonance imaging (MRI) radiomics to predict 1-year efficacy of epidermal growth factor receptor tyrosine kinase inhibitors (EG...
OBJECTIVES: To construct and validate a model based on clinical characteristics and magnetic resonance imaging (MRI) radiomics to predict 1-year efficacy of epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) in patients with EGFR-mutant non-small cell lung cancer (NSCLC) brain metastases (BMs). METHODS: This study retrospectively analyzed data from 338 patients with EGFR-mutant NSCLC BMs from three centers, including MRI, clinical and pathological data, and radiological features. Based on the selected significant radiomic features from intratumoral regions extracted from CE-T1WI, while exploring the value of features in 3/5/8 mm peritumoral regions, seven commonly used machine learning algorithms were compared to select the optimal one for model construction, and the best algorithm was selected for model construction. In the model predicting 1-year therapeutic efficacy, clinical, radiomic, and combined models were constructed separately. The model performance was evaluated using receiver operating characteristic curves. RESULTS: The final development cohort comprised 285 patients from Center 1, while the external validation set included 57 patients from Centers 2 and 3. In the model predicting 1-year EGFR-TKIs efficacy, the random forest algorithm, which showed the best application, was used to construct the model. Compared with the radiomic and clinical models, the combined model exhibited superior area under the curve performance in the test set (0.756 vs. 0.644 vs. 0.668). In the external validation set, the combined model achieved an area under the curve of 0.743 (95% CI: 0.604-0.881). CONCLUSION: Compared to single clinical or radiomic models, the combined model was more effective in predicting the 1-year efficacy of EGFR-TKIs in patients with NSCLC BMs with EGFR mutations.
Assessment of cerebrovascular reactivity in middle cerebral artery stenosis using non-hypercapnic resting-state fMRI: a potential biomarker for hemodynamic alterations.
Zhang L, Wang L, Tang X … +4 more , Zhu Y, Wang R, Zhang H, Ding Z
BMC Med Imaging · 2026 Jun · PMID 42288826 · Full text
BACKGROUND: Cerebrovascular reactivity(CVR), a key indicator of cerebrovascular reserve, is crucial for evaluating cerebrovascular pathophysiology. This study employed resting-state MRI (rs-MRI) to assess CVR alteration...
BACKGROUND: Cerebrovascular reactivity(CVR), a key indicator of cerebrovascular reserve, is crucial for evaluating cerebrovascular pathophysiology. This study employed resting-state MRI (rs-MRI) to assess CVR alteration in patients with unilateral middle cerebral artery stenosis or occlusion (MCA-S) under non-hypercapnic conditions, comparing them with healthy controls. METHODS: A total of 41 patients with unilateral MCA-S and 50 age-, sex-, and education-matched normal controls (NC). All underwent rs-MRI and neuropsychological assessments. CVR was derived from rs-fMRI frequency band signals, and t-test was conducted to obtain the CVR-differentiated brain regions. Exploratory seed-based FC analysis was further performed to characterize the network context of regions showing altered CVR. Partial correlation analyses explored relationships between these differential brain regions and both neuropsychological assessments and clinical indicators. Discriminative performance of the CVR-related metric between MCA-S and controls was evaluated using receiver operating characteristic (ROC) curves. RESULTS: Compared with the NC group, patients with MCA-S exhibited increased CVR in the contralesional Cerebellum Crus1 (CC1) and decreased iCVR in the ipsilesional postcentral gyrus (PoCG). When contralesional CC1 served as regions of interest (ROIs), increased FC was observed in the ipsilesional middle frontal gyrus (MFG) and the contralesional precuneus of MCA-S patients. The partial correlation analysis indicated a positive correlation between the FC of the ipsilesional MFG and anxiety scores (r = 0.404, (95%CI: 0.106, 0.631), P = 0.012, P-FDR = 0.030). Using ipsilesional PoCG as the ROI, MCA-S patients showed significantly decreased FC in ipsilesional PoCG, contralesional precentral gyrus (PreCG), and ipsilesional supplementary motor area. The FC of the contralesional PreCG showed a positive correlation with anxiety scores (r = 0.436, (95%CI: 0.142, 0.658), P = 0.006, P-FDR = 0.030). ROC analysis demonstrated strong diagnostic accuracy for CVR in CC1 (AUC = 0.809) and PoCG (AUC = 0.787), with a combined AUC of 0.866. CONCLUSION: Non-hypercapnic rs-MRI effectively evaluates CVR alterations in MCA-S patients and may serve as a complementary physiological biomarker for characterizing hemodynamic alteration.
The application value of time-dependent DWI in the differential diagnosis of benign and malignant breast lesions and microstructural feature analysis.
Qi X, He Y, Wang W … +6 more , Zhai C, Yu S, Yang H, Duan S, Liu M, Chen M
BMC Med Imaging · 2026 Jun · PMID 42288784 · Full text
OBJECTIVE: To evaluate the incremental diagnostic value of various ADC sequences and microstructural parameters derived from time-dependent diffusion-weighted imaging (td-DWI) in differentiating benign and malignant brea...
OBJECTIVE: To evaluate the incremental diagnostic value of various ADC sequences and microstructural parameters derived from time-dependent diffusion-weighted imaging (td-DWI) in differentiating benign and malignant breast lesions. METHODS: In this study, a total of 52 patients with breast lesions were included, comprising 24 cases of benign lesions and 28 cases of malignant lesions. MRI examinations were conducted to measure four ADC values: ADC_25Hz, ADC_50Hz, ADC_PGSE, and ADC_Zoomit. Additionally, microstructural parameters fin, Dex, d, and Cellularity were calculated. Clinical and imaging characteristics between two groups were compared using independent samples t-tests and chi-square tests, and the diagnostic performance of each parameter was evaluated. RESULTS: Age, maximum diameter, FGT, morphology, TIC curve, edema, T2WI signal, DWI signal, and BI-RADS score showed significant differences between the benign and malignant groups (p < 0.05). The differences in ADC_25Hz, ADC_PGSE, and ADC_Zoomit between the two groups were statistically significant (p = 0.042, 0.020, < 0.001), with ADC_Zoomit exhibiting the highest AUC of 0.808. Among the microstructural parameters, fin, Dex, and d showed significant differences between the groups (p = 0.005, 0.001, < 0.001), with d having the highest AUC of 0.816 and an accuracy, specificity, and sensitivity of 0.769, 0.792, and 0.75, respectively. A logistic regression model based on diffusion-derived parameters demonstrated good diagnostic performance, and the combined model incorporating clinical and imaging features achieved the highest diagnostic accuracy. CONCLUSION: Microstructural parameters derived from td-DWI, particularly d, together with ADC_Zoomit, demonstrated promising diagnostic performance for differentiating benign and malignant breast lesions. These findings suggest that td-DWI may provide incremental diagnostic value beyond conventional MRI features for breast cancer.
Hybrid metaheuristic feature selection for breast cancer detection in digital mammography: a radiomics and deep learning pilot feasibility study.
Alshreef BS
BMC Med Imaging · 2026 Jun · PMID 42288782 · Full text
BACKGROUND: Artificial intelligence (AI) can improve breast cancer detection in mammography, but high-dimensional feature spaces and feature-selection instability remain challenging. This study developed a hybrid metaheu...
BACKGROUND: Artificial intelligence (AI) can improve breast cancer detection in mammography, but high-dimensional feature spaces and feature-selection instability remain challenging. This study developed a hybrid metaheuristic feature-selection framework that combines radiomics and deep learning features and evaluated its methodological feasibility on a small real mammography pilot and a controlled synthetic comparison designed to test behavior under collapse-prone conditions. METHODS: Using the public CBIS-DDSM dataset, 2,051 Image Biomarker Standardization Initiative (IBSI)-compliant radiomic features and 2,048-dimensional deep features from a pretrained, non-fine-tuned EfficientNet-B5 model were extracted for each lesion region of interest (ROI). A hybrid Grasshopper Optimization Algorithm and Crow Search Algorithm (GOA-CSA) with a proposed multi-constraint fitness function was used to select an optimal feature subset for a multilayer perceptron (MLP) classifier. Performance was assessed on a small CBIS-DDSM pilot subset (n = 22, 5-fold stratified cross-validation) and on a synthetic dataset (N = 16, D = 1114) designed to compare the proposed fitness against a legacy fitness under collapse-prone conditions. RESULTS: On the CBIS-DDSM pilot, the hybrid GOA-CSA model selected an average of 486 features, achieving a cross-validated area under the receiver operating characteristic curve (AUC) of 0.750 ± 0.433 and sensitivity of 0.433 ± 0.435, compared with an all-features baseline AUC of 0.900 ± 0.224 and sensitivity of 0.667 ± 0.471. In the synthetic comparison, the proposed fitness achieved an AUC of 0.810 ± 0.115 and sensitivity of 0.571 ± 0.198 versus 0.476 ± 0.210 and 0.286 ± 0.241, respectively, for the legacy fitness. The collapse-prevention penalty was implemented but was not empirically triggered in this pilot because both models maintained non-zero sensitivity. CONCLUSIONS: This pilot feasibility study demonstrates that the hybrid GOA-CSA framework can successfully identify compact feature subsets combining radiomic and deep features. The results are exploratory and hypothesis-generating, and the small real-data sample size limits definitive performance evaluation. The synthetic experiment supports the conceptual value of the multi-constraint fitness design, but the collapse-prevention penalty remains empirically unvalidated on real mammography data. External validation on independent cohorts such as VinDr-Mammo remains a crucial subject for future work.
Photon-counting CT for paraspinal muscle fat quantification compared with MRI proton density fat fraction.
Shu D, Wang J, Liang D … +2 more , Zhao L, Zhao Z
BMC Med Imaging · 2026 Jun · PMID 42286523 · Full text
BACKGROUND: To evaluate photon-counting CT (PCCT) derived 70 keV attenuation values and virtual noncontrast fat fraction (VNC FF) in quantifying paraspinal muscle fat infiltration, using MRI proton density fat fraction (...
BACKGROUND: To evaluate photon-counting CT (PCCT) derived 70 keV attenuation values and virtual noncontrast fat fraction (VNC FF) in quantifying paraspinal muscle fat infiltration, using MRI proton density fat fraction (PDFF) as the reference standard. METHODS: In this prospective study, 76 adults with low back pain underwent same-day lumbar PCCT and 6-echo q-Dixon MRI within a 2-hour interval. The cohort consisted of 76 participants (38 men, 38 women), with a mean age of 47.7 ± 14.0 years and a mean body mass index (BMI) of 24.6 ± 3.3 kg/m². VNC FF represents a material decomposition-based fat fraction obtained from PCCT. Regions of interest (ROI) were bilaterally drawn in the multifidus, erector spinae, and psoas major at four intervertebral disc levels (L2/3-L5/S1). Correlation analysis, linear mixed-effects regression, Bland-Altman analysis, and receiver operating characteristic analysis were performed. RESULTS: The 70 keV CT values showed a strong correlation with MRI PDFF at the ROI level (r = - 0.931), outperforming VNC FF (r = 0.876). At the subject level, correlations were consistently strong across intervertebral disc levels (r range, - 0.964 to - 0.975 for CT values; 0.766 to 0.896 for VNC FF) and muscle groups (r range, - 0.881 to - 0.984 for CT values; 0.827 to 0.966 for VNC FF). Regression modeling enabled derivation of an internally calibrated CT fat fraction (CTFF), which closely approximated MRI PDFF within the study cohort. Agreement in categorical fat infiltration grading (< 10%, 10-30%, 30-50%, > 50%) was moderate (κ = 0.623). For binary classification at the 30% threshold, CTFF demonstrated excellent diagnostic performance (AUC = 0.993, 95% CI: 0.990-0.997). CONCLUSION: PCCT-derived 70 keV CT values showed strong agreement with MRI PDFF and enabled internal regression-based estimation of paraspinal muscle fat, with excellent diagnostic performance for fat infiltration classification.
Preoperative assessment of intratumor heterogeneity using intravoxel incoherent motion MRI for survival prediction in high-grade gliomas: a feasibility study.
Wang X, Wang Y, Wang X … +5 more , Liu F, Suo S, Song Y, Zhou Y, Cao M
BMC Med Imaging · 2026 Jun · PMID 42286511 · Full text
BACKGROUND: Intratumor heterogeneity (ITH) is closely associated with poor prognosis in high-grade gliomas (HGGs). This study aimed to characterize ITH and explore potential imaging markers that predict overall survival...
BACKGROUND: Intratumor heterogeneity (ITH) is closely associated with poor prognosis in high-grade gliomas (HGGs). This study aimed to characterize ITH and explore potential imaging markers that predict overall survival (OS) in HGGs using intravoxel incoherent motion magnetic resonance imaging (IVIM MRI)-based spatially explicit analysis. METHODS: Sixty-five HGG patients who underwent surgical resection were analyzed. Preoperative IVIM MRI images were collected and processed to obtain true diffusion coefficient (D) and perfusion fraction (f) maps. Tumor regions of interest were segmented, and the k-means algorithm was applied to cluster the D and f image voxels for generating spatial habitats and extracting quantitative image features. Kaplan-Meier analysis and Cox proportional hazards were used to compare variables and patient subgroups. RESULTS: Three spatial habitats were identified: Habitat 1 (hypo-vascular, hyper-cellular), Habitat 2 (hypo-cellular), and Habitat 3 (hyper-vascular). In the multivariate Cox regression analysis, isocitrate dehydrogenase (IDH) genotype (hazard ratio [HR] = 0.298, P = 0.003) and volume percentage (pVol) of Habitat 1 (HR = 6.155, P = 0.01) showed prognostic significance, with the model yielding a concordance index of 0.756. A pVol value of Habitat 1 below 47.6% predicted survival benefits in patients with HGG and IDH wild-type gliomas, as well as in those with HGG who underwent subtotal resection (median OS improvement: 11, 11, and 8 months, respectively). CONCLUSIONS: Spatial habitats identified via IVIM MRI may aid in characterizing cellular and vascular heterogeneity in HGGs, with the pVol of hypo-vascular, hyper-cellular habitat potentially serving as an independent predictor of patient survival.
Deep learning for the diagnosis of lumbar disc herniation: a systematic review and meta-analysis.
Li Y, Zhang C, Qi Z … +5 more , Wang Q, Li D, Gao F, Ren X, Chen C
BMC Med Imaging · 2026 Jun · PMID 42277705 · Full text
BACKGROUND: Lumbar disc herniation (LDH) is a major cause of low back pain and disability worldwide. Although magnetic resonance imaging (MRI) is the standard modality for diagnosis, interpretation remains subject to int...
BACKGROUND: Lumbar disc herniation (LDH) is a major cause of low back pain and disability worldwide. Although magnetic resonance imaging (MRI) is the standard modality for diagnosis, interpretation remains subject to interobserver variability. Deep learning (DL)-based approaches have been increasingly applied to improve diagnostic accuracy; however, their overall performance and sources of heterogeneity remain unclear. METHODS: PubMed, Web of Science, and the Cochrane Library were searched from inception to March 2026. Studies evaluating imaging-based DL models for LDH diagnosis were included if sufficient diagnostic performance data were available. Two reviewers independently performed study selection, data extraction, and quality assessment using QUADAS-2. Pooled sensitivity and specificity were estimated using random-effects models, and summary receiver operating characteristic (SROC), subgroup, and sensitivity analyses were performed. The primary subgroup analysis used one primary standalone DL result per study to reduce non-independence. RESULTS: Ten retrospective studies were included. The pooled sensitivity and specificity were 0.94 (95% CI: 0.90-0.96) and 0.94 (95% CI: 0.90-0.97), respectively, with an area under the SROC curve of 0.99. Substantial heterogeneity was observed (I² > 97%), with no obvious threshold effect (ρ = -0.188, P = 0.603), indicating that the pooled estimates should be interpreted as exploratory. External validation studies showed lower specificity than internal or same-center temporally independent validation studies (0.87 vs. 0.96; P = 0.034), while sensitivity was similar. Sensitivity analyses suggested that differences in model task and output structure contributed to heterogeneity. At a pretest probability of 20%, a positive DL result increased the posttest probability to approximately 80%, whereas a negative result reduced it to approximately 2%. CONCLUSION: DL-based imaging models show promising diagnostic potential for LDH and may support assisted screening, triage, and lesion localization. However, the evidence is limited by substantial heterogeneity, retrospective study designs, non-patient-level analytical units, variable reference standards, and limited external validation. Future studies should use standardized task definitions, annotation procedures, AI reporting frameworks, and multicenter prospective patient-level validation before routine clinical implementation. CLINICAL TRIAL REGISTRATION: Not applicable. This study is a systematic review and meta-analysis based on previously published literature and did not involve any prospective intervention involving human participants. REGISTRATION: This systematic review and meta-analysis was registered in PROSPERO (CRD420261353452).
The diagnostic performance of salivary gland shear wave elastography to differentiate Sjögren disease from type 2 diabetes mellitus with sicca: a cross-sectional controlled study.
Ozkan E, Kahveci A
BMC Med Imaging · 2026 Jun · PMID 42277700 · Full text
BACKGROUND: Xerostomia is a shared clinical manifestation of both Sjögren disease (SjD) and type 2 diabetes mellitus (DM), but their underlying pathophysiology differs markedly. The differentiation between autoimmune gla...
BACKGROUND: Xerostomia is a shared clinical manifestation of both Sjögren disease (SjD) and type 2 diabetes mellitus (DM), but their underlying pathophysiology differs markedly. The differentiation between autoimmune glandular damage and metabolic dysfunction is of clinical significance. This study aimed to evaluate the diagnostic performance of shear wave elastography (2D-SWE) for discriminating SjD from DM patients with sicca symptoms. METHODS: This cross-sectional study included three groups: 61 patients with SjD, 42 patients with type 2 DM presenting with sicca symptoms, and 43 age-matched healthy controls (HC). All participants underwent bilateral parotid and submandibular gland ultrasonography. Parenchymal inhomogeneity was graded using the Outcome Measures in Rheumatology (OMERACT) scoring system, and tissue stiffness (Young's modulus, kPa) was quantified via 2D-SWE. Pairwise comparison between groups were conducted, and the receiver operating characteristic (ROC) analysis was used to assess the performance of SWE in identifying SjD patients. RESULTS: The SjD group demonstrated higher stiffness values in the bilateral parotid and submandibular glands compared with both the DM and control groups. Additionally, the OMERACT total scores for all salivary glands were elevated in patients with SjD compared with the DM and HC groups (p < 0.001 for both groups). In contrast, the DM group exhibited tissue stiffness and gray-scale ultrasound profiles that were comparable to those of the HC group (p > 0.0167 for all comparisons). The ROC analysis indicated that SWE possessed moderate discriminatory power for distinguishing SjD from DM (Area Under the Curve (AUC) range for glands: 0.598-0.628). CONCLUSION: The combined use of SWE and gray-scale assessment may serve as a multiparametric approach that could contribute to differentiating autoimmune sialadenitis in SjD patients from metabolic sialosis in DM patients presenting with dry mouth.
Development and validation of a deep learning model for automatic detection of depressed skull fractures from CT scans.
Angkurawaranon S, Chatchavan S, Iangkoonchorn T … +3 more , Singlor T, Jarunnarumol N, Inkeaw P
BMC Med Imaging · 2026 Jun · PMID 42277670 · Full text
BACKGROUND: Depressed skull fractures with bone depression greater than in one cortex might cause major consequences and require surgery in traumatic head injury patients. Therefore, skull fractures with depression in mo...
BACKGROUND: Depressed skull fractures with bone depression greater than in one cortex might cause major consequences and require surgery in traumatic head injury patients. Therefore, skull fractures with depression in more than one cortex must be identified quickly and accurately. METHODS: This study proposes using a deep learning model to deal with the task. Cranial CT scans of traumatic head injury patients with and without depressed skull fractures were collected for this retrospective investigation. A real-time object detection model, You Only Look Once (YOLO), was adopted to detect depressed skull fractures in more than one cortex. We proposed a two-phase training strategy for training the model. The model was evaluated using internal and external test datasets. The detection performance was reported in terms of accuracy, sensitivity, specificity, precision, negative predictive value, F1-score, and area under the receiver operating characteristic curve. RESULTS: The deep learning model demonstrated strong performance on an internal test dataset (accuracy = 0.957); however, its performance declined on two external test datasets (accuracy = 0.884 and 0.857). CONCLUSION: This model enables automated detection of depressed skull fractures, streamlining the clinical workflow by flagging high-priority cases for expedited radiologist review.
Yolov7 neural network for pelvic X-rays after total hip arthroplasty: automatically measuring component position parameters- a retrospective feasibility study.
Li J, Ding L, Zhang K … +10 more , Zhang Y, Gao S, Jin F, Ji Q, Chen Q, Guo Z, Lan W, Wang H, Zhang L, Li X
BMC Med Imaging · 2026 Jun · PMID 42271275 · Full text
BACKGROUND: After total hip arthroplasty (THA), regular follow-up X-rays are required to evaluate the position of the prosthesis. However, measuring the parameters of prosthesis position consumes a significant amount of...
BACKGROUND: After total hip arthroplasty (THA), regular follow-up X-rays are required to evaluate the position of the prosthesis. However, measuring the parameters of prosthesis position consumes a significant amount of time, and there may be variations in the results obtained by different doctors. PURPOSE: To validate the feasibility of an artificial intelligence model for the automated measurement of component position parameters on anteroposterior X-rays after THA. MATERIALS AND METHODS: Post-THA anteroposterior X-rays were collected from five hospitals in Inner Mongolia, China. They were divided into training, validation and test sets. The model was divided into left, right, and bilateral THA modules for training. Statistical analysis was performed using Python 3.8 and SPSS 25.0. The accuracy of the model in keypoint detection and measurement were assessed. In addition, we compared the time difference between the model and manual measurements. RESULTS: A total of 1050 X-rays were included in the training and validation sets (350 X-rays for each module), and 105 in the test set (35 X-rays for each module). Overall, 80.00-100.00% of the model-predicted keypoints fell within 3 mm of the standard keypoints. The model-predicted values were highly consistent with the standard reference values (intra-class correlation coefficient = 0.89-0.99, r = 0.83-0.99, root mean square error = 0.96-3.43, mean absolute error = 0.78-2.87, mean difference = -1.29-1.17). The model-predicted values were highly consistent with the measurements of senior attending radiologists. The model took an average of 3.2 s to measure an X-ray image, which is significantly shorter than the 751.76 s in manual measurement. CONCLUSIONS: This study validates the feasibility of the model for rapid and accurate keypoint identification and component position parameter measurement on anteroposterior X-rays. The model holds promise as a diagnostic aid to alleviate clinicians' workload and facilitate the monitoring of prosthetic positioning changes following THA. CLINICAL TRIAL NUMBER: Not applicable.
Microstructural MRI combined with metabolic MRI to evaluate the progression-free survival of patients with glioblastoma at the early follow-up after tumor treating fields: a pilot study.
Lu J, Zhang X, Guo X … +4 more , Qin Q, Chen J, Sun P, Wang J
BMC Med Imaging · 2026 Jun · PMID 42271271 · Full text
BACKGROUND: Non-invasive assessment of treatment response in Tumor Treating Fields (TTFields) and exploration of progression-free survival (PFS) by imaging techniques remains a major challenge. OBJECTIVES: This explorato...
BACKGROUND: Non-invasive assessment of treatment response in Tumor Treating Fields (TTFields) and exploration of progression-free survival (PFS) by imaging techniques remains a major challenge. OBJECTIVES: This exploratory pilot study investigated the feasibility of using time-dependent diffusion MRI (t-dMRI) combined with chemical exchange saturation transfer (CEST) MRI to assess the TTFields treatment efficacy and explore associations with PFS in postoperative glioblastoma (GBM) patients. METHODS: This study employed t-dMRI and CEST techniques to scan patients before and 1-6 months after receiving TTFields or temozolomide treatment between May 2023 and November 2024. Microstructural metrics, including Diameter, intracellular volume fraction (V), Cellularity index and extracellular diffusivity (D), were calculated using the IMPULSED method. Multiple metabolic parameters were derived through magnetization transfer ratio asymmetry analysis and pH-weighted analysis. Longitudinal changes in these parameters were analyzed using linear mixed-effects models (LMM), with sensitivity analysis conducted via the Mann-Whitney U test. Spearman correlation analysis assessed relationships between parameters and PFS. RESULTS: Ten patients were included (age range 29-74 years; 8 men). The TTFields group showed a trend toward greater early reduction in cellularity index at first follow‑up. LMM analysis showed different dynamic evolution trends in Diameter (F = 5.042, P = 0.035) and pH-weighted values (F = 5.291, P = 0.055) between the progression and non-progression groups. At the third month post-treatment, the Diameter in the progression group increased from baseline, while it decreased in the non-progression group (estimated value = 3.17, 95% CI: 0.29-6.06, P = 0.036). Sensitivity analyses supported these descriptive between‑group differences. Correlation analysis showed that at the third month post-treatment, the change in diameter was associated with the pH-weighted change (r = 0.697, P = 0.025). During the early treatment phase, both the change in diameter and the pH-weighted change relative to baseline showed negative associations with patient PFS (r = -0.636, P = 0.048; r = -0.697, P = 0.025). CONCLUSIONS: This exploratory pilot study suggested that combining t-dMRI with CEST MRI may help describe microstructural and metabolic changes that could be associated with TTFields efficacy during early treatment. Changes in Diameter and pH-weighted values, both in trend and magnitude, showed correlations with patient PFS. These observations support further exploratory investigations in larger cohorts to clarify the role of multiparametric MRI in evaluating early TTFields‑associated changes and prognosis. CRITICAL RELEVANCE STATEMENT: Early assessment of the efficacy of TTFields for GBM is key to individualized therapy, but challenges remain. This exploratory pilot study combined microstructural and metabolic MRI to describe early longitudinal changes linked to TTFields response. The preliminary findings provided initial insights that may inform future larger-scale studies aimed at better understanding treatment response in GBM patients. CLINICAL TRIAL NUMBER: Not applicable.
Swin UNETR-based prediction of [Ga]Ga-FAPI-46 PET/CT dose rate maps in cancer patients: quantitative comparison with [F]FDG PET/CT.
Izadi-Yazdi S, Babapour-Mofrad F, Yazdani E … +3 more , Karamzade-Ziarati N, Arabi H, Sadeghi M
BMC Med Imaging · 2026 Jun · PMID 42271267 · Full text
BACKGROUND: Fibroblast activation protein (FAP) is a promising theranostic target due to its high stroma expression in numerous malignancies. This study presents the first DL-based framework for predicting dose rate maps...
BACKGROUND: Fibroblast activation protein (FAP) is a promising theranostic target due to its high stroma expression in numerous malignancies. This study presents the first DL-based framework for predicting dose rate maps (DRMs) from [Ga]Ga-FAPI-46 PET/CT, following a quantitative and dosimetric comparison with [F]FDG PET/CT. The approach supports personalized pre-therapeutic planning in radiopharmaceutical therapies (RPTs). METHODS: PET/CT scans with [Ga]Ga-FAPI-46 and [F]FDG from 22 cancer patients were retrospectively analyzed. DRMs were generated using the dose voxel kernel (DVK) method using the GATE toolkit. Absorbed doses (ADs) obtained from DVK-based DRMs for tumors and organs at risk (OARs) were compared with those derived from full Monte Carlo (MC), and local energy deposition (LED), based DRMs, and organ-level estimates calculated using OLINDA/EXM v1.1. Tracer uptake of [Ga]Ga-FAPI-46 and [F]FDG was compared using SUV, SUV, and tumor-to-background ratio (TBR). A shifted windows UNEt TRansformers (Swin UNETR) model was trained to predict DRMs and benchmarked against ResNet-32 using R², RMSE, and gamma index. RESULTS: [Ga]Ga-FAPI-46 PET/CT demonstrated higher TBRs and lower OARs uptake and ADs compared to [F]FDG, indicating its promise in enhancing lesion detectability. The Swin UNETR model achieved an RMSE of 0.0598 Gy/s, R of 0.960, and a gamma pass rate of 98.71%, outperforming ResNet-32. CONCLUSION: Compared to [F]FDG, [Ga]Ga-FAPI-46 PET offers higher image contrast, better lesion detectability, and improved dosimetric profiles, supporting personalized RPT planning. While Swin UNETR enables fast and accurate DRM prediction from [Ga]Ga-FAPI-46, broader validation in larger, multicenter cohorts is needed to ensure reproducibility and clinical impact.
Nomogram prediction model for pathological complete response of metastatic axillary lymph nodes of breast cancer after neoadjuvant chemotherapy.
Chen Y, Song G, Xu X … +1 more , Yang Y
BMC Med Imaging · 2026 Jun · PMID 42265639 · Full text
OBJECTIVE: To evaluate the predictive value of a nomogram model combining ultrasound features and clinical indicators for pathological complete response (pCR) of metastatic axillary lymph nodes in breast cancer after neo...
OBJECTIVE: To evaluate the predictive value of a nomogram model combining ultrasound features and clinical indicators for pathological complete response (pCR) of metastatic axillary lymph nodes in breast cancer after neoadjuvant chemotherapy (NACT). METHODS: A total of 109 women with histologically confirmed breast cancer and axillary lymph node metastasis were retrospectively analyzed. Patients were divided into pCR and non-pCR groups after NACT. A nomogram model was established using univariate and multivariate logistic regression analyses. The predictive performance of the clinical model, ultrasound model, and combined model was compared using the area under the curve (AUC). RESULTS: Molecular subtype, clinical stage, pre-NACT metastatic axillary lymph node subtype, post-NACT reduction rate of lymph node short diameter, and post-NACT reduction rate of primary tumor maximum diameter were independent predictors of axillary lymph node pCR. The combined model (AUC 0.844, 95%CI: 0.773-0.915, P < 0.05) showed significantly improved predictive ability compared with the clinical model (AUC 0.725, 95%CI: 0.628-0.823, P < 0.05) and ultrasound model (AUC 0.754, 95%CI: 0.665-0.844, P < 0.05). CONCLUSION: The nomogram combining ultrasound and clinical indicators effectively predicts pCR of metastatic axillary lymph nodes after NACT and provides a reference for surgical selection. However, the single‑center design, small sample size, and lack of external and prospective validation limit its generalizability. Multi‑center prospective validation is warranted.
Repeatability and reproducibility of APT and GluCEST imaging at 5 T: phantom validation and clinical evaluation in patients with brain tumors.
Chen T, Wang L, Wang T … +5 more , Zheng C, Qin Z, Cai W, Li G, Hu C
BMC Med Imaging · 2026 Jun · PMID 42265609 · Full text
PURPOSE: To evaluate the repeatability and reproducibility of amide proton transfer (APT) and glutamate chemical exchange saturation transfer (GluCEST) imaging at 5 T and to investigate their clinical value in patients w...
PURPOSE: To evaluate the repeatability and reproducibility of amide proton transfer (APT) and glutamate chemical exchange saturation transfer (GluCEST) imaging at 5 T and to investigate their clinical value in patients with brain tumors. METHODS: Phantom experiments under varying pH conditions and bovine serum albumin (BSA) concentrations were performed to assess the within-session repeatability, between-session and between-day reproducibility of APT and GluCEST measurements using intraclass correlation coefficients (ICCs), mean absolute difference, coefficient of repeatability (CoR) and coefficient of variation (CV). In vivo APT and GluCEST imaging was performed in 96 patients with brain tumors. Intra- and inter-observer agreements of APTmean and GluCESTmean measurements were evaluated using ICCs and Bland-Altman analysis across tumor, peritumoral edema, and normal brain tissue regions. Quantitative tumor-region APT and GluCEST metrics (APT/GluCESTmean, APT/GluCEST difference, and APT/GluCEST change ratio) were compared among major tumor subtypes using the Kruskal-Wallis test, followed by post hoc pairwise comparisons using Dunn's test with Bonferroni correction. Receiver operating characteristic (ROC) analysis was performed to evaluate the diagnostic performance for differentiating high-grade gliomas from low-grade gliomas. RESULTS: In phantom studies, both APT and GluCEST demonstrated robust repeatability and reproducibility with high ICCs (> 0.96) and low mean absolute difference, CoR and CV. Both APT and GluCEST-weighted signals showed nonlinear pH-dependent changes and increased with higher BSA concentrations. In patients, intra- and inter-observer agreements were excellent for APTmean and GluCESTmean in tumor, peritumoral edema, and normal brain regions (ICCs > 0.93). Tumor and peritumoral edema regions showed higher APT and GluCEST values than normal brain tissue (p < 0.05). After multiple-comparison correction, high-grade gliomas showed significantly higher APTmean, APT difference, GluCESTmean, GluCEST difference and GluCEST change ratio than low-grade gliomas. GluCEST-derived metrics showed favorable performance for glioma grading, with GluCEST change ratio achieving the highest area under the curve (AUC) of 0.846, followed by GluCEST difference (AUC = 0.838) and GluCESTmean (AUC = 0.836). CONCLUSION: APT and GluCEST imaging at 5 T showed excellent repeatability and reproducibility in phantom experiments and robust measurement agreement in patients with brain tumors. APT and GluCEST measurements demonstrated subtype-related differences, and GluCEST-derived metrics showed favorable performance for differentiating high-grade from low-grade gliomas.
A novel description of dynamic organ-specific trajectories in pelvic organ prolapse based on magnetic resonance imaging traces.
Hou X, Zhou S, Feng F … +3 more , Wang X, Li A, Liu T
BMC Med Imaging · 2026 Jun · PMID 42260382 · Full text
BACKGROUND: The dynamic motion trajectories of pelvic organs during prolapse remain incompletely characterized. Detailing these trajectories could enhance the understanding and treatment of pelvic organ prolapse (POP). T...
BACKGROUND: The dynamic motion trajectories of pelvic organs during prolapse remain incompletely characterized. Detailing these trajectories could enhance the understanding and treatment of pelvic organ prolapse (POP). This study aimed to generate and visualize generalized dynamic trajectories for the bladder, uterus, and rectum. METHODS: This observational study enrolled 63 symptomatic patients with Stage III or worse POP (57 with cystocele, 35 with uterine prolapse, and 29 with rectocele). Patients underwent dynamic pelvic MRI scans during three consecutive maximum Valsalva maneuvers. Pelvic organs were segmented by contouring to calculate their centroids. Centroid displacement was tracked with a coordinate system based on the sacrococcygeal-inferior pubic point (SCIPP) line. Trajectories were analyzed using generalized additive mixed models (GAMMs) to generate composite fitting curves, followed by segmented linear regression. RESULTS: The fitted trajectories revealed distinct, organ-specific motion patterns. Based on the angles between the centroid motion trails and the SCIPP line, the bladder formed an inverse C-shape (three breakpoints, with slopes decreasing from 99.3° to 63.5°), the uterus exhibited an inverse S-shape (four breakpoints, with slopes fluctuating between 56.3° and 93.3°), and the rectum followed a near-linear path (slopes ranging from 62.3° to 67.0°). Furthermore, measurement approaches utilizing these organ-specific trajectories quantified prolapse severity more accurately than conventional, one-size-fits-all methods. CONCLUSIONS: Generalized POP trajectories are organ-specific, nonlinear, and segmented, challenging the clinical utility of universal reference lines. The heterogeneous motion patterns of these organs highlight the need for organ-specific MRI assessment protocols and tailored surgical strategies to optimize clinical outcomes.
Development and internal validation of a clinical nomogram incorporating quantitative CT features for predicting malignancy in pulmonary nodules ≤ 3 cm.
Ruan Y, Ma Y, Han J … +5 more , Long C, Cao W, Yang A, Sun P, Zhang T
BMC Med Imaging · 2026 Jun · PMID 42260372 · Full text
BACKGROUND: Accurate differentiation between benign and malignant pulmonary nodules ≤ 3 cm remains a clinical challenge. This study aimed to develop and internally validate a clinically interpretable nomogram integrating...
BACKGROUND: Accurate differentiation between benign and malignant pulmonary nodules ≤ 3 cm remains a clinical challenge. This study aimed to develop and internally validate a clinically interpretable nomogram integrating clinical variables and quantitative computed tomography (CT) features for predicting malignancy in pulmonary nodules. METHODS: This retrospective single-center study included 1,419 patients with pulmonary nodules ≤ 3 cm who underwent surgical resection between January 2012 and July 2025 with pathologic confirmation. The cohort was randomly divided into a training set (n = 994) for model development and a validation set (n = 425) for internal validation. Clinical data, conventional imaging findings, serum biomarkers, and quantitative CT measurements from preoperative thin-section CT were collected. Multivariable logistic regression was used to identify variables associated with malignancy and construct the nomogram. RESULTS: Among the 1,419 nodules, 1,150 (81.0%) were malignant and 269 (19.0%) were benign. The final nomogram incorporated seven variables: suspicious radiologic features, nodule size, sex, symptoms at detection, consolidation-to-tumor ratio, minimum CT attenuation, and age. Age was retained in the final model on clinical grounds despite lacking statistical significance in multivariable analysis. Suspicious radiologic features (adjusted odds ratio [aOR] = 6.61, 95% confidence interval [CI]: 4.51-9.84; P < 0.001), nodule diameter > 2 cm (aOR = 4.07, 95% CI: 2.16-7.62; P < 0.001), female sex (aOR = 1.69, 95% CI: 1.23-2.33; P = 0.001), asymptomatic presentation (aOR = 0.48, 95% CI: 0.34-0.69; P < 0.001), consolidation-to-tumor ratio > 0.50 (aOR = 0.20, 95% CI: 0.06-0.61; P = 0.005), and minimum CT attenuation per 100-HU increase (aOR = 0.82, 95% CI: 0.74-0.92; P < 0.001) were independently associated with malignancy. The nomogram showed good discrimination, with area under the receiver operating characteristic curve values of 0.809 in the training set and 0.782 in the validation set. Calibration analysis showed agreement between predicted and observed risks, and decision curve analysis supported usefulness. CONCLUSIONS: We developed and internally validated a clinical nomogram incorporating quantitative CT features for malignancy risk estimation in surgically resected pulmonary nodules ≤ 3 cm. The model showed good discrimination, calibration, and potential utility in a malignancy-enriched preoperative cohort. External validation in broader, less selected, screening-detected, incidental, and multicenter populations is warranted before routine clinical application.
The utilization of myocardial perfusion imaging prior to coronary angiography to improve outcomes and cost-effectiveness of patients with stable coronary artery disease: a comparative cohort study in China.
Yang L, Jia Y, Hu Y … +7 more , Wu P, Zan C, Li L, Xiao Y, Wei H, Wu Z, Li S
BMC Med Imaging · 2026 Jun · PMID 42251324 · Full text
BACKGROUND: The incremental value of myocardial perfusion imaging (MPI) before coronary angiography (CAG) for suspected or known stable coronary artery disease (SCAD) has not been validated in China. This study investiga...
BACKGROUND: The incremental value of myocardial perfusion imaging (MPI) before coronary angiography (CAG) for suspected or known stable coronary artery disease (SCAD) has not been validated in China. This study investigated the necessity of MPI for these patients regarding improvements in cardiac outcomes and cost-effectiveness in China. METHODS: The initial cohort comprised 7,437 patients with suspected or known SCAD between 2018 and 2019. After excluding patients with acute coronary syndrome, previous myocardial infarction (MI) and revascularization, enrolled patients were divided into the CAG group and the MPI group according to the initial strategy (CAG or MPI). Then, two groups were matched by propensity score. The cost, revascularization, MI, and all-cause mortality of patients were followed. RESULTS: The MPI and CAG groups each included 130 patients after matching, with similar basic characteristics (P > 0.05). Significantly decreased cardiac events were observed in the MPI group compared to the CAG group (6 vs. 16, P < 0.05). Furthermore, the number of revascularization (5 vs. 12), MI (1 vs. 3), and all-cause mortality (0 vs. 1) in the MPI group was also lower. The Cox model showed that fewer patients in the MPI group had cardiac events (HR 0.27, 95% CI 0.10-0.71). Moreover, patients in the MPI group had fewer first-visit costs ($718 vs. $1389) and fewer downstream costs ($120 vs. $344) than those in the CAG group (P < 0.001). CONCLUSIONS: Our findings suggest that an MPI-first strategy is associated with fewer cardiac events and lower costs compared with an initial CAG strategy. TRIAL REGISTRATION: Not applicable. This study is a retrospective analysis.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe