搜索结果：Radiological and Ultrasound Technology

Quantitative assessment of myocardial perfusion using 13N-ammonia PET with compartmental modeling enables evaluation of myocardial flow reserve (MFR) and prediction of patient prognosis. However, the reliability of these assessments can depend on the analytical methods used for quantitation. The present study aimed to evaluate the variability and agreement of values obtained using three quantitative software tools and to assess the impact of kinetic model selection on myocardial blood flow (MBF) and MFR estimates in a clinical setting. We analyzed 100 patients who underwent 13N-ammonia PET/CT, including 60 with normal perfusion and 20, 10, and 10 with single-, two-, and three-vessel disease, respectively. We derived MBF and MFR at global (entire left ventricle) and regional (coronary territories) levels and evaluated five analytical pipelines: SyngoMBF, QPET, and three implementations of PMOD tools (1-tissue compartment, Hutchins, and UCLA models). MBF and MFR showed high correlations among the software tools, although stress MBF statistically differed between PMOD and QPET. Correlation coefficients between software tools ranged from 0.81 to 0.91 at the global level, and Bland-Altman analysis demonstrated overall agreement with residual variability. In contrast, MBF and MFR values varied depending on the compartment model. The UCLA model yielded the highest stress MBF and MFR, and correlation coefficients between models ranged from 0.43 to 0.99 at the global level. Although Bland-Altman analysis showed overall agreement, noticeable scatter persisted and the UCLA model exhibited a positive bias. Quantitative MBF and MFR estimates from 13N-ammonia PET show good overall agreement across commonly used software tools but remain strongly dependent on kinetic model selection. These findings indicate that quantitative results are not directly interchangeable across different software and modeling approaches, underscoring the importance of methodological consistency when interpreting myocardial perfusion PET in clinical practice.

Ovarian ectopic pregnancy (OEP) is a rare and life-threatening condition that is typically diagnosed post-rupture. Its diagnosis and management become more complex when it is concurrent with ovarian endometrioma, as the latter may mask the clinical and radiological features of OEP. We report the case of a 28-year-old woman (gravida 2, para 1) who presented with 44 days of amenorrhoea and lower abdominal pain. Transvaginal ultrasound (TVS) revealed an empty uterus, a complex right adnexal mass containing a yolk sac (with synchronous movement with the ovary and a negative "sliding organ sign", raising a strong suspicion of ovarian ectopic pregnancy), and a separate "ground-glass" cystic lesion (consistent with an endometrioma). Corpus luteum blood flow signals were detected in the left ovary. The patient's preoperative haemoglobin concentration was 128 g/L. Diagnostic laparoscopy confirmed a right ovarian pregnancy co-existing with an ipsilateral endometrioma. Both lesions were excised laparoscopically while preserving the ovary. Haemostasis was achieved by primary suturing supplemented with minimal bipolar coagulation to preserve ovarian function. The patient recovered well; her postoperative haemoglobin concentration was 122 g/L, and her menses resumed at 6 weeks post-operatively, which confirmed preserved ovarian function. This case reaffirms a fundamental clinical principle: any reproductive-age woman with a positive pregnancy test, an empty uterus, and an adnexal mass should be presumed to have an ectopic pregnancy, prompting immediate surgical evaluation. In our patient, this principle alone mandated surgery. The transvaginal ultrasound findings (a yolk sac and a negative "sliding organ sign") did not change the need for surgery, but they provided critical preoperative localization of the gestational sac to the ovary. This allowed us to anticipate an ovarian pregnancy, obtain specific consent for ovary-conserving surgery, and plan a suture-dominant haemostatic strategy. To our knowledge, this is the first reported case of pre-rupture diagnosis of an ovarian ectopic pregnancy masked by an endometrioma using these sonographic signs. Clinicians must prioritize the clinical triad; when available, meticulous ultrasound adds precision for fertility preservation.

Insulin-derived amyloidosis (IDA) reduces insulin absorption, increasing the risk of poor glycemic control; however, early detection remains challenging. Although ultrasound can identify IDA, the effectiveness of portable pocket-sized devices has not been evaluated. This study was performed to evaluate the detectability of findings suggestive of IDA using a pocket-sized ultrasound device on the abdominal region, the most common site for stable insulin absorption in patients with diabetes mellitus. This cross-sectional observational study was conducted in the diabetes ward of a university hospital between July and December 2024. The participants were inpatients with diabetes who had been receiving insulin injections for more than 1 month. Findings suggestive of IDA were assessed through visual inspection and palpation, magnetic resonance imaging (MRI), high-performance ultrasonography, and a pocket-sized ultrasound device. The concordance rate between the pocket-sized ultrasound device and other assessment methods was calculated. Of the 20 participants enrolled, 5 met the exclusion criteria; thus, data from 15 participants were included in the final analysis. Findings suggestive of IDA were identified in nine patients (60.0%) by visual inspection and palpation and in eight patients (53.3%) by the pocket-sized ultrasound device, high-performance ultrasound, and MRI. The concordance rate among the pocket-sized ultrasound device, high-performance ultrasound, and MRI was 100%, while the concordance rate between visual inspection with palpation and the pocket-sized ultrasound device was 80%. These findings suggest that a pocket-sized ultrasound device has the potential to detect findings suggestive of IDA and may be suitable for use in clinical practice.

Osteoporosis is a chronic skeletal disorder characterized by reduced bone mineral density and disrupted bone microarchitecture, affecting over 200 million individuals worldwide. The lumbar spine, containing the largest volume of metabolically active trabecular bone, is particularly vulnerable to osteoporotic degeneration and compression fractures. This narrative review examines recent advances in imaging modalities for lumbar spine osteoporosis assessment, emphasizing the diagnostic utility and emerging clinical applications of ¹⁸F-sodium fluoride (NaF) positron emission tomography/computed tomography (PET/CT). A comprehensive narrative review was conducted, synthesizing findings from pivotal studies investigating conventional imaging methods and newer PET-based technologies for osteoporosis evaluation. Particular focus was given to studies utilizing quantitative and kinetic PET biomarkers for assessing bone metabolic activity with ¹⁸F-NaF. While dual-energy X-ray absorptiometry (DXA) remains the clinical standard for bone mineral density assessment, it has significant limitations including poor spatial resolution, lack of three-dimensional capability, and inability to differentiate cortical from trabecular bone. In contrast, ¹⁸F-NaF PET/CT demonstrates superior image quality, rapid tracer kinetics, and quantitative assessment of regional osteoblastic activity. Studies show strong correlations between ¹⁸F-NaF uptake and bone turnover markers, mineral density measurements, and therapeutic response. Kinetic modeling approaches provide detailed insights into bone remodeling dynamics, supporting personalized treatment planning and prognostic assessment. Diagnostic performance studies report area under the receiver operating characteristic curves as high as 0.96 for osteoporosis detection when evaluated against DXA-derived BMD, though no study has yet compared both modalities against an independent gold standard such as fracture outcomes. ¹⁸F-NaF PET/CT offers optimal clinical applications for early treatment response monitoring, evaluation of patients with discordant clinical risk and DXA findings, pre-surgical assessment in patients with borderline bone density, and investigation of complex metabolic bone disorders. Ideally, ¹⁸F-NaF PET/CT should be utilized in a complementary fashion to DXA. Primary barriers to clinical adoption include cost, limited accessibility, and absence of standardized kinetic modeling protocols. Future research should focus on establishing reference ranges across age and sex demographics, validating fracture prediction models, and determining cost-effectiveness thresholds for specific clinical scenarios such as high-risk patients with discordant DXA and fracture history.

Cardiac magnetic resonance imaging (MRI) is a gold standard to assess functional and anatomical properties of the living heart. Inflammation changes the myocardial tissue and, furthermore, MR relaxation properties. Continuous-wave (CW) longitudinal rotating frame relaxation time (T1ρ) mapping has been used to assess myocardial fibrosis and inflammation. Conventional T2 relaxation time is a known marker of edema in the myocardium. In this study, we assessed myocardial inflammation after viral infection in a mouse heart using CW-T1ρ and T2 relaxation times. Adenoviral human vascular endothelial growth factor-A165 (AdVEGF-A165) and empty control adenoviral vector with cytomegalovirus promoter (AdCMV) gene transfers were used to induce inflammation in the mouse myocardium. In vivo CW-T1ρ and T2 relaxation time measurements were performed in both groups (AdVEGF-A165 and AdCMV) after -1-, 1-, 3-, 7-, 14-, 21-, and 28-day post-injection. The inflammation associated with gene transfer was verified by hematoxylin and eosin staining after 14-day post-injection. One day after AdVEGF-A165 and AdCMV injections and inflammatory reactions, CW-T1ρ showed a significant increase, which stayed increased as a function of time. T2 also increased significantly after both injections and inflammatory reactions as compared to before injections. Contrast difference between inflammation and remote areas was visually observed in both groups in CW-T1ρ and T2 maps. Hematoxylin and eosin staining revealed the area of inflammation after Ad injection in both groups after 14-day post-injection. This study showed that both acute and chronic phases of the inflammatory reaction in mouse myocardium caused by myocardial adenoviral injections were associated with increased CW-T1ρ and T2 relaxation time constants. Furthermore, the inflammatory reaction can be followed up with rotating frame and conventional relaxation time mappings.

Intestinal-type and pancreatobiliary-type periampullary carcinomas exhibit distinct biological behaviours and prognostic outcomes, yet accurate preoperative subtyping remains a major clinical challenge. This study aimed to evaluate the value of clinical variables and computed tomography (CT) imaging features in the differential diagnosis of intestinal-type and pancreatobiliary-type periampullary carcinoma, and to develop a subtype prediction model. This retrospective study included 83 patients with pathologically confirmed periampullary carcinoma, comprising 21 intestinal-type and 62 pancreatobiliary-type periampullary carcinomas. Clinical variables and conventional CT imaging features were evaluated using univariable and multivariable logistic regression analyses to identify predictors associated with PAC subtype. Based on these predictors, clinical, radiologic, and combined prediction models were developed, and a nomogram was constructed from the combined model. Model performance was evaluated using receiver operating characteristic (ROC) curves and the area under the curve (AUC). The DeLong test was employed to compare the diagnostic performance among different models. The AUCs of the clinical model and radiologic model were 0.82 (95%CI, 0.70-0.92) and 0.85 (95%CI, 0.79-0.93), respectively. The combined model showed significantly better diagnostic performance than either model alone, with an AUC of 0.92 (95%CI, 0.84-0.97), a sensitivity of 87.1%, and a specificity of 90.5%. Serum carbohydrate antigen 19 - 9, total bilirubin, tumor location, and enhancement degree of lesion were identified as the most important predictors in the combined model. In addition, the nomogram derived from the combined model demonstrated good discriminative ability for predicting histologic subtype. The combined model integrating clinical and CT imaging features enables more accurate preoperative differentiation between intestinal-type and pancreatobiliary-type periampullary carcinoma and yields higher sensitivity and specificity than models based on clinical or imaging features alone.

Early diagnosis and accurate staging of endometrial cancer (EC) are crucial for effective treatment planning. Distinguishing stage IA from stage IB EC is challenging due to large variations in tumor and uterine morphology, as well as the limited availability of annotated magnetic resonance imaging (MRI) data for training robust models. A genetic programming (GP)-based framework was developed for the classification of FIGO stage IA and IB EC using a small number of MRI images. The framework consists of three main components: (1) automatic detection of regions of interest (ROI) on MRI images, (2) GP-based feature extraction and construction, and (3) EC stage classification. A fast single-shot detector (SSD) was employed to automatically localize the uterus as the ROI. The detected ROI images were cropped and resized to construct training and test datasets. Four GP-based methods with different structures and primitives were employed for feature extraction and construction: GP with convolutional operators (COGP), GP with image descriptors (IDGP), GP with flexible program structures and image-related operators (FlexGP), and GP with automatic simultaneous learning of features and evolutionary ensembles (FELGP). The best-performing GP individuals were used to generate discriminative features, which were subsequently used to train classifiers for stage IA and IB EC. Experimental results from three MRI datasets revealed that GP-based methods achieved competitive performance relative to both neural and traditional non-neural machine learning approaches. The proposed methods achieved classification accuracies of up to 0.92 on cropped axial diffusion-weighted imaging (DWI), 0.87 on cropped axial T2-weighted imaging (T2WI), and 0.83 on cropped sagittal T2WI images of EC patients. GP-based methods effectively classify FIGO stage IA and IB endometrial cancer using limited MRI data. By automatically extracting discriminative and interpretable features from ROI within lesions, the proposed framework provides a reliable and transparent solution for EC staging, highlighting the potential of GP in medical image analysis and clinical decision support.

Agenesis of the corpus callosum (ACC) presents with highly heterogeneous clinical features. Common methods rarely achieve accurate prenatal or early postnatal diagnosis and prognosis. We aimed to develop and test an interpretable deep neural network (DNN) that combines multimodal clinical data to improve diagnostic accuracy and neurodevelopmental outcome prediction. We collected data from 205 pediatric patients with ACC at Wuhan Children's Hospital between 2016 and 2024. A total of 27 clinical features were extracted, including neuroimaging findings, perinatal risk factors, and follow-up developmental quotients (Gesell Developmental Schedules and Gross Motor Function scores). Five-fold cross-validation was adopted. We built an eight-layer fully connected DNN with ReLU activation in the hidden layers. For categorical endpoints, a sigmoid output layer with binary cross-entropy loss was used. For continuous endpoints, a linear output layer with mean squared error loss was used. SHAP (Shapley Additive Explanations) values were used to quantify the contribution of individual features to model predictions. Performance was compared with a support vector machine (SVM) baseline and across hyperparameter settings. Area under the receiver-operating-characteristic curve (AUC), F1 score, precision, recall, mean absolute error (MAE), mean squared error (MSE), and coefficient of determination (R2) served as primary metrics. Across 12 neurodevelopmental disorders, the model reached an average AUC of 0.97. AUCs for intellectual disability, autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), specific learning disorder and developmental coordination disorder ranged from 0.98 to 1.00. Prediction remained moderate for cerebral palsy (AUC&#x202f;=&#x202f;0.74) and epilepsy (AUC&#x202f;=&#x202f;0.67). MAE for both Gesell and Gross Motor Function scores was 0.10, with corresponding R2 values of 0.62 and 0.63. SHAP analysis identified extracranial malformation (clinical type III), facial dysmorphism and birth weight as the most influential features for developmental outcome. The DNN model outperformed the SVM baseline, with an AUC improvement of 0.16 for communication disorder and an R2 increase of 0.19 for Gesell score (p&#x202f;<&#x202f;0.001). Ablation experiments confirmed eight layers, sixteen neurons per layer, a learning rate of 0.01 and ten training epochs as the optimal configuration. Additional layers or higher learning rates caused overfitting. The proposed interpretable DNN framework outperforms traditional classifiers in early ACC diagnosis and developmental outcome prediction. It provides a potential tool for clinical decision support. Larger samples and integration of raw imaging data are needed to enhance prediction of complex phenotypes such as cerebral palsy and epilepsy.

Thermal ablation offers a safer, less invasive, and more cost-effective curative-intent treatment for selected patients with primary and metastatic liver tumours than surgery; when done with appropriate technique, ablation can deliver similar oncological outcomes. However, effectiveness in routine practice varies because structured training, planning, and procedural governance remain scarce. These international multidisciplinary, multi-society guidelines-formally endorsed by the European Society of Surgical Oncology, the Cardiovascular and Interventional Radiological Society of Europe, and the Society of Interventional Oncology-define key domains contributing to procedural difficulty and practice variation in liver tumour thermal ablation. A Delphi consensus initiative held in Innsbruck, Austria, engaged 72 experts across three iterative rounds of scoring across 135 statements grouped into five domains: credentialing, indications, approach, procedural factors, and safety measures. Consensus was achieved for 94 (70%) of 135 statements. The least invasive route-typically percutaneous-should be prioritised, and margin adequacy was reaffirmed as the principal technical goal. Procedural difficulty was considered context-dependent, shaped by tumour factors, institutional infrastructure, and operator experience. Organ displacement techniques were endorsed to maintain safety and expand treatable indications. Complex ablations should be done by experienced operators (more than 100 previous cases), with programmes underpinned by structured training, multidisciplinary team participation, and routine audit. Future efforts should develop and validate practical tools such as difficulty scoring systems, standardised procedural reporting templates, and comprehensive training curricula to improve consistency, standardisation, and clinical outcomes globally.

Since its initial introduction to interventional cardiology over two decades ago, optical coherence tomography (OCT) has emerged as a powerful tool in neurovascular intervention. This intravascular imaging modality uses near-infrared light to provide cross-sectional visualization of the vessel wall with a resolution approaching 10 &#x3bc;m. The resolution of OCT far surpasses that of other imaging techniques. This higher resolution enables radiologists to directly assess arterial wall disease, including atherosclerotic plaque, aneurysm, and thrombus, as well as the interaction between therapeutic devices and the arterial wall in real time, providing actionable information during neurovascular interventions. The growing reliance on endovascular approaches to treat intracranial aneurysms and ischemia underscores the importance of precise vessel evaluation during treatment to provide accurate imaging guidance. However, digital subtraction angiography and cone beam computed tomography angiography often fail to reveal underlying arterial disease and other key features, such as the presence of thrombi, dissections, and malapposed stents, that could lead to incomplete treatment and acute and chronic complications. By enabling direct visualization of these microstructural details, OCT may overcome some of the most persistent challenges in neurovascular practice, ultimately improving diagnostic accuracy, procedural safety, and long-term patient outcomes. Nevertheless, integrating OCT into neurovascular settings remains challenging. There is still a lack of large-scale clinical validation, and existing coronary devices are not suitable for reliable use in tortuous intracranial vascular circulations. To overcome the technical limitations of current technologies, neuroOCT technology was designed specifically for neurovascular use and was evaluated in a first-in-human study. This technology will enable future clinical studies to investigate using neuroOCT to guide and optimize neurovascular treatments. This review article aims to provide a comprehensive perspective on the potential of neuroOCT in neurovascular practice. It highlights the technology's technical principles, current applications, limitations, and prospects for reshaping vascular imaging and therapy in the brain.

Melanoma recurrence risk is highest within the first 2 years after diagnosis and progressively declines thereafter. Current surveillance strategies remain largely guided by clinicopathologic risk stratification, with the comprehensive medical history, physical examination, and complete skin assessment forming the cornerstone of follow-up. Although cross-sectional imaging and lymph node ultrasound are used in selected higher-risk patients, routine intensive imaging has not consistently demonstrated survival benefit and may increase costs and false-positive findings. Emerging technologies are reshaping melanoma surveillance and clinical management. Circulating tumor DNA (ctDNA) has shown promise as a minimally invasive biomarker capable of detecting molecular residual disease and anticipating clinical recurrence. Persistent or newly positive ctDNA after surgery is consistently associated with inferior recurrence-free survival. However, ctDNA does not reliably detect all recurrence patterns and its sensitivity varies according to disease burden and metastatic site. Prospective validation and clarification of how ctDNA should guide adjuvant therapy or imaging strategies remain necessary. In parallel, CD8-targeted positron emission tomography (CD8 PET) has emerged as a novel functional imaging modality capable of noninvasively visualizing whole-body T-cell dynamics. By differentiating tumor burden from immune infiltration and capturing early T-cell recruitment, CD8 PET offers predictive insights into immunotherapy response. Nevertheless, limitations of this technique include dependence on optimal imaging timing, limited tracer availability, cost, and an inability to directly assess T-cell functionality. Together, ctDNA and immune-focused imaging approaches represent promising steps toward precision surveillance and management of melanoma. Further robust prospective studies are required to define their integration into clinical decision making and optimize patient outcomes.

Functional motor disorders (FMDs) represent a frequent and disabling neurological condition. The lack of reliable diagnostic biomarkers and their heterogeneity might affect diagnosis. We identified multimodal biomarkers distinguishing FMDs from healthy controls (HCs) using machine-learning approaches. In this multicenter cross-sectional study, consecutive adults with a clinically established FMDs diagnosis (n&#x2009;=&#x2009;75, 74.7% female; mean age 44.20&#x2009;&#xb1;&#x2009;12.92) and age- and sex-matched HCs (n&#x2009;=&#x2009;75; 58.6% female; 48.42&#x2009;&#xb1;&#x2009;11.67) were recruited. All participants underwent standardized behavioral, neurophysiological, and brain MRI assessment exploring motor, exteroceptive, and interoceptive domains. A Random Forest (RF) classifier combined with repeated stratified k-fold cross-validation was trained on the collected features. Predictive performance was evaluated using accuracy, sensitivity, specificity, precision, F1-score, and AUC-ROC. SHapley Additive exPlanations interpreted feature importance. The strongest diagnostic biomarkers were lower dual-task effect scores for postural sway area under eyes-closed motor and cognitive conditions, and gait speed during the motor dual-task, followed by increased vDMN and basal ganglia networks functional connectivity, reduced baseline ipsilateral-contralateral R2 blink reflex area, and higher DNIC-to-baseline N2P2 amplitude ratios for the lower limb. The RF classifier achieved robust performance (accuracy 85.0%, sensitivity 83.9%, specificity 86.1%, F1-score 85.7%, AUC-ROC 0.921). Motor, functional neuroimaging, and neurophysiological markers demonstrated diagnostic value in distinguishing FMDs from healthy controls, addressing the current lack of objective tools and supporting more confident and accurate diagnosis of these heterogeneous conditions. Trial registration number NCT06328790. Registered on 26 March 2024.

Early detection of pancreatic cancer is crucial for survival, but detecting smalllesions remains challenging. Intraductal Ultrasound (IDUS) using intracardiac echocardiography (ICE) catheters for B-mode and Shear-Wave Elastography (SWE) potentially offers improved visualization and characterization of small tumors. This study assesses the feasibility of IDUS using ICE catheters to detect and visualize periampullary tumors in surgically resected specimens. In this two-phase ex-vivo feasibility study, 25 pancreatic specimens were included, of which the first 10 were used to establish and standardize the imaging protocol, followed by technical feasibility evaluation in the remaining 15 specimens. Catheters were introduced into the pancreatic duct, common bile duct, or positioned extraductally to enable tumor visualization with B-mode imaging and shear-wave elastography (SWE). Tumor visualization rates, catheter insertion success, SWE measurements in normal and tumor tissue, and image quality were assessed. ICE catheter insertion was successful in 12 of 15 specimens; unsuccessful access was primarily related to large tumor size (>4 cm) or unidentifiable ductal anatomy following surgical resection. However, extraluminal imaging successfully visualized tumors in one of these cases. Median shear-wave speed and elastic modulus for normal pancreatic parenchyma were 1.58 m/s and 7.6 kPa, respectively. SWE measurements in tumor tissue were suboptimal, likely due to ex-vivo tissue variability and catheter strain during repeated use. IDUS with ICE is feasible for qualitative B-mode visualization of periampullary tumors and enables SWE assessment of pancreatic parenchyma in an ex-vivo setting. Reliable elastography of tumor tissue remained challenging, indicating the need for further technical refinement and in-vivo validation.

The prediction of Epidermal Growth Factor Receptor (EGFR) mutation status in advanced lung adenocarcinoma is crucial for targeted therapy. Since EGFR mutations manifest as both macroscopic imaging features on CT and microscopic morphological changes in tissue, integrating these multiscale signals is essential for a comprehensive diagnostic assessment. However, current related research faces two key limitations: on one hand, unimodal deep learning models suffer from limited representational power; on the other hand, existing multimodal methods fail to address the inherent data structural discrepancies between continuous CT and discrete WSI, often losing critical fine-grained details due to forced data compression or shared semantic bottlenecks. To address the above limitations and improve the reliability of EGFR mutation status prediction, this study aims to propose a novel multimodal fusion framework (MFCA) that can effectively capture cross-modal semantic interactions and align imaging features across different scales. A novel MFCA based on Cross-Attention (MFCA) is proposed, and its implementation steps are as follows: 1. First, a region-of-interest-guided approach is utilized to coarsely segment whole-slide histopathology images (WSI) into three constituent regions, namely cancerous, stromal, and other regions; 2. Then, a dual-branch encoder is employed to separately extract features from two types of imaging data-global features from Computed Tomography (CT) scans and region-specific features from the segmented WSI; 3. Critically, a bidirectional cross-attention module is introduced into the framework, which is designed to facilitate deep semantic interaction and alignment between the macroscopic context of CT imaging and the microscopic context of histopathology, thereby achieving highly efficient and discriminative feature fusion. On the external validation set, our MFCA framework achieved robust performance, with Area Under the Curve (AUC) values of 0.758(95% CI: 0.683-0.832) for cancerous regions, 0.805(95% CI: 0.716-0.900) for stromal regions, and 0.760(95% CI: 0.686-0.833) for other regions. The model's performance, particularly in the stromal component, was statistically superior to all baseline and competing models. The proposed MFCA framework predicts EGFR mutation status by innovatively integrating macroscopic CT imaging with region-specific microscopic WSI features. It serves as a valuable computational tool to support precision oncology for patients with advanced lung adenocarcinoma.

To develop and validate a CT-based radiomics model for differentiating primary gastric lymphoma (PGL) from Borrmann type IV gastric cancer (GC). A total of 136 patients with pathologically confirmed PGL (n&#x202f;=&#x202f;56) and Borrmann type IV GC (n&#x202f;=&#x202f;80) were retrospectively enrolled between January 2016 and May 2022. The cohort was randomly partitioned into a training set (n&#x202f;=&#x202f;95) and a testing set (n&#x202f;=&#x202f;41) at a 7:3 ratio. Radiomics features were extracted from unenhanced, arterial, venous, double-phase (arterial + venous), and three-phase (unenhanced + arterial + venous) CT images. After feature selection using the Least Absolute Shrinkage and Selection Operator, radiomics models were constructed via logistic regression. A clinical-radiomics model was developed through multivariate analysis. The models were evaluated using Receiver Operating Characteristic (ROC) curves, calibration curves with the Hosmer-Lemeshow test, and decision Curve Analysis (DCA) for clinical net benefit. Clinical model comprised of high-enhanced serosa sign, normalized CT value on venous phase, and perigastric fat infiltration showed good performance with AUCs of 0.902 (training set) and 0.878 (testing set). Among the radiomics models, the three-phase model outperformed others (AUC: 0.871 training, 0.865 testing). The clinical-radiomics combined model further improved discriminatory performance, achieving AUCs of 0.960 and 0.932 in the training and testing sets, respectively. DCA confirmed that the combined model provided the highest clinical net benefit. Clinical-radiomics model incorporating three-phase radiomics signatures and CT findings achieved satisfactory performance for differentiating PGL from Borrmann type IV GC, serving as a reliable non-invasive tool for clinical decision-making.

Digital clubbing is an important clinical sign associated with a range of cardiopulmonary diseases; however, its detection and severity assessment in routine practice largely rely on subjective visual inspection. This study proposes an automated, smartphone-based system for real-time detection and severity assessment of digital clubbing using deep learning techniques. The system integrates the YOLOv8 object detection model for initial clubbing classification, the KeypointNet model for anatomical landmark localization, and a novel Clubbing Fingers Severity Analysis (CFSA) algorithm to quantify the Lovibond angle and grade disease severity. Finger images were acquired using a smartphone camera with an OpenCV-based preprocessing strategy to standardize finger-to-camera distance and improve image consistency. Model performance was evaluated using publicly available anonymized datasets. The proposed system achieved an overall accuracy of 94.7% for digital clubbing detection and severity classification. The YOLOv8 model attained a classification accuracy of 92.5%, while the KeypointNet model achieved a landmark localization accuracy of 96.5%. Notably, the recall for severe digital clubbing reached 94.0%, indicating strong sensitivity for identifying high-risk cases. By providing real-time, non-invasive, and reproducible assessments, the proposed system addresses the limitations of conventional visual examination and supports objective severity grading. Although further clinical validation is required, this smartphone-based approach demonstrates strong potential as a preliminary screening support tool for early identification of digital clubbing in clinical and community-based settings, particularly in resource-limited environments.

Objectives: Our objectives were to develop and validate the habitat model based on low-dose computed tomography (LDCT) for noninvasive prediction of the visceral pleural invasion (VPI) in subpleural nodules with solid component. Methods: A total of 313 patients with subpleural lung adenocarcinoma nodules from three centers were retrospectively enrolled and divided into training (n = 192), validation (n = 82), and external test (n = 39) sets. All patients underwent preoperative LDCT scan. The habitat model was constructed using unsupervised clustering to partition each tumor into three distinct habitats, from which radiomic features were extracted and selected. Its diagnostic performance was compared with a whole-lesion radiomic model and radiological model. Statistical analysis included receiver operating characteristic (ROC) analysis and DeLong test. Results: The habitat model significantly outperformed both the radiomic and radiological models across the validation and external test sets, with areas under the ROC curve of 0.893 and 0.908, respectively (all p < 0.05). In contrast, the radiomic model achieved 0.833 and 0.772, while the radiological model yielded 0.746 and 0.624. The corresponding software tool has been made publicly available to facilitate broader clinical application. Conclusions: The habitat imaging model based on LDCT effectively predicts the VPI in subpleural lung adenocarcinoma by quantifying intratumoral spatial heterogeneity and demonstrates promising diagnostic performance compared to conventional radiomic and radiological methods. This approach offers a noninvasive preoperative tool to assist in risk stratification and guide personalized therapeutic decision-making for subpleural nodules detected during lung cancer screening.

To evaluate the performance of T1&#x3c1; mapping for myocardial fibrosis detection across distinct cardiomyopathy entities of ischemic and non-ischemic origin against native T1 mapping. A total of 14 healthy controls and 39 patients [15 with ischemic cardiomyopathy [ICM], 12 with hypertrophic cardiomyopathy [HCM], and 12 with dilated cardiomyopathy [DCM]] underwent cardiac magnetic resonance (CMR), including T1&#x3c1; mapping, native T1 mapping, late gadolinium enhancement (LGE), and extracellular volume (ECV) mapping. Segments were visually classified as segments with or without LGE. T1&#x3c1; values were significantly higher in patients with ICM, HCM, and DCM than controls (all P&#x2009;<&#x2009;0.05). T1&#x3c1; showed favorable diagnostic performance compared with native T1 in distinguishing ICM and HCM patients from controls [area under the curve (AUC): 0.959 vs. 0.739 for ICM; AUC: 0.878 vs. 0.763 for HCM], while both parameters exhibited comparable performance in DCM (AUC: 0.814 vs. 0.814). Notably, elevated T1&#x3c1; values were observed in segments with or without LGE. T1&#x3c1; mapping demonstrated good performance in distinguishing segments with LGE from those without LGE in ICM and HCM (AUC: 0.820 and 0.726) and exhibited a significant correlation with ECV across all disease types (ICM: r&#x2009;=&#x2009;0.598; HCM: r&#x2009;=&#x2009;0.577; DCM: r&#x2009;=&#x2009;0.648; all P&#x2009;<&#x2009;0.05). This exploratory study demonstrates the potential of T1&#x3c1; mapping as a non-contrast CMR technique for detecting myocardial fibrosis in ischemic and non-ischemic cardiomyopathies. T1&#x3c1; showed favorable diagnostic performance compared with native T1 in ICM and HCM with comparable performance in DCM. These findings warrant validation in larger cohorts with diverse cardiac conditions to establish the clinical utility of T1&#x3c1; imaging.

Response to neoadjuvant chemoradiotherapy (NACRT) in locally advanced rectal cancer (LARC) is highly heterogeneous. Reliable pretreatment prediction of tumor regression and prognosis remains an unmet clinical need to optimize personalized management. A multicenter retrospective study of 434 Stage II-III LARC patients was conducted across three tertiary hospitals. Pretreatment T2-weighted MRI was used to build intratumoral and peritumoral macrohabitats. Local radiomic features within the tumor were clustered using K-means to generate intratumoral habitats, with the optimal cluster number determined by the Calinski-Harabasz score. Radiomic and 3D deep-learning features from each habitat and the peritumoral region were fused after LASSO-based selection. Machine-learning classifiers (support vector machine, logistic regression, multilayer perceptron) were trained to predict tumor regression grade (TRG). Performance was assessed by ROC and decision-curve analyses, and prognostic value for progression-free survival (PFS) was evaluated using Kaplan-Meier analysis. The macro habitat-based fusion model demonstrated superior performance compared with intratumoral, peritumoral, or single-habitat models, achieving AUCs of 0.807-0.830 in the external validation cohort. The derived risk score showed a significant association with progression-free survival (PFS) (p&#x2009;=&#x2009;0.011 in the training and p&#x2009;=&#x2009;0.030 in the validation cohorts). The macro habitat-based MRI radiomics and deep learning fusion model provides a noninvasive, interpretable, and robust biomarker for predicting treatment response and prognosis in LARC. It holds potential to guide personalized therapeutic strategies, including organ-preserving approaches and tailored surveillance after NACRT.